Polymeric Surfactants in Disperse System

19
Polymeric surfactants in disperse systems Tharwat Tadros 89 Nash Grove Lane, Wokingham, Berkshire RG40 4HE, UK a b s t r a c t a r t i c l e i n f o Available online 5 November 2008 Keywords: Homopolymers Block and graft copolymers Solution properties Adsorption and conformation Steric stabilization Emulsions Suspensions Latexes Nano-emulsions This ov ervi ew star ts with a section on gen era l classi cation of polym eric surfac tants. Both homop olymers, block and graft copolymers are described. The solution properties of polymeric surfactants is described by using the FloryHuggin s theo ry. Par tic ularatten tio n is giv en to theeffect of sol ven cy of themediu m forthe polymer chains. The adsorption and conformation of homopolymers, block and graft copolymers at the solid/liquid interface is described. The theories of polymer adsorption and their predictions are brie y described . This is followed by a descri ption of the experimen tal technique s that can be appli ed to study polymer ic surfac tant adsorption. Examples of adsorption isotherms of non-ionic polymeric surfactants are given. The effect of solvency on the adsorption amount is also described. Results for the adsorbed layer thickness of polymeric surfactants are given with particu lar attent ion to the effectof mole cular weigh t. The inter actionbetweenparticles contain ing adsorbed layer s is descri bed in terms of the unfav orable mixing of the stabili zing chains when these are in good solvent conditions. The entropic, volume restriction or elastic interaction that occurs on considerable overlap is also described. Combination of these two effects forms the basis of the theory of steric stabilization. The energy distan ce curveproducedwith thesestericall y stabili zed syste ms is descri bedwith partic ular atten tion of the effect of the ratio of ads orb ed lay er thi ckn ess to dro ple t rad ius.Exampl es of oil -in- wa ter (O/W) and wat er -in- oil (W/O ) emulsions stabilized with polymeric surfactants are given. Of particular int erest is the O/Wemulsions stabilized using hydr ophobi cally modi ed inu lin (INUTE C ® SP1). The emul sions produced are highly stable against coales cenceboth in wate r and high electr olyte concen tratio ns. This is accou ntedfor by the multip oint attachment of the pol yme ric sur fac tantto the oil dro ple ts withsever al alky l grou ps and the str ong ly hyd rat ed loo ps and tail s of linear polyfructose. Evidence of this high stability was obtained from disjoining pressure measurements. Stabilization of suspensions using INUTEC ® SP1 was described with particular reference to latexes that were pre par ed usin g emulsion pol yme riz atio n. Thehigh stab ilit y of thelatex es is attr ibu tedto thestrongadsorptionof the polymeric surfactant on the particle surfaces and the enhanced steric stabilization produced by the strongly hydr ated polyfru ctose loops and tails. Evidence for such high stability was obtained using Atom ic Force Microscopy (AFM) measurements. The last part of the overview described the preparation and stabilization of nano-emulsio ns using INUTE C ® SP1 . In par ticu lar thepolymericsurfa cta nt wasveryeffect ivein red uci ng Ost wal d ripening as a result of its strong adsorption and the Gibbs elasticity produced by the polymeric surfactant. © 2008 Elsevier B.V. All rights reserved. Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 1.1. Genera l classi cation of polymeric surfactants. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 1.2. Solution properties of polymeric surfactants. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 1. 3. Ads o rp t ion an d co nf o rmation of pol ymer ic sur fa ct an ts at i nt er fa ces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 2. Theories of polymer adsorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 3. Experimental techniq ues for studying polymeric surfactant adsorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286 3.1. Measurement of the adsorption isotherm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287 3.2. Measurement of the fraction of segments p. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287 3.3 . Dete rmina tion of th e seg ment dens ity d istri buti on ρ(z) and adsorbed layer thickness  δ h . . . . . . . . . . . . . . . . . . . . . . . 288 4. Examples of the adsorption isotherms of nonionic polymeric surf actants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289 5. Inte ract ion be twee n pa rt ic le s cont ai ni ng adsorbed polyme ri c surf actant la ye rs st er ic st ab il iz ati on. . . . . . . . . . . . . . . . . . . . . . 28 9 5.1. Mixing inter acti on G mix  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290 5.2. Ela sti c int era cti on G el  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290 Advances in Colloid and Interface Science 147148 (2009) 281299 E-mail address: tharwat@tadr os.fsnet.co.uk . 288 289 290 290 0001-8686/$  see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.cis.2008.10.005 Contents lists available at  ScienceDirect Advances in Colloid and Interface Science  j o urna l h omep a g e: ww w. e l se v i e r. c o m/ l o cat e / c i s

Transcript of Polymeric Surfactants in Disperse System

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 119

Polymeric surfactants in disperse systems

Tharwat Tadros

89 Nash Grove Lane Wokingham Berkshire RG40 4HE UK

a b s t r a c ta r t i c l e i n f o

Available online 5 November 2008

Keywords

Homopolymers

Block and graft copolymers

Solution properties

Adsorption and conformation

Steric stabilization

Emulsions

Suspensions

Latexes

Nano-emulsions

This overview starts with a section on general classi1047297cation of polymeric surfactants Both homopolymers block

and graft copolymers are described The solution properties of polymeric surfactants is described by using the

FloryndashHuggins theory Particularattention is given to theeffect of solvency of themedium forthe polymer chains

The adsorption and conformation of homopolymers block and graft copolymers at the solidliquid interface isdescribed The theories of polymer adsorption and their predictions are brie1047298y described This is followed by a

description of the experimental techniques that can be applied to study polymeric surfactant adsorption

Examples of adsorption isotherms of non-ionic polymeric surfactants are given The effect of solvency on the

adsorption amount is also described Results for the adsorbed layer thickness of polymeric surfactants are given

with particular attention to the effectof molecular weight The interactionbetweenparticles containing adsorbed

layers is described in terms of the unfavorable mixing of the stabilizing chains when these are in good solvent

conditions The entropic volume restriction or elastic interaction that occurs on considerable overlap is also

described Combination of these two effects forms the basis of the theory of steric stabilization The energyndash

distance curveproducedwith thesesterically stabilized systems is describedwith particular attention of the effect

of the ratio of adsorbed layer thickness to droplet radiusExamples of oil-in-water (OW) and water-in-oil (WO)

emulsions stabilized with polymeric surfactants are given Of particular interest is the OW emulsions stabilized

using hydrophobically modi1047297ed inulin (INUTECregSP1) The emulsions produced are highly stable against

coalescenceboth in water and high electrolyte concentrations This is accountedfor by the multipoint attachment

of the polymeric surfactantto the oil droplets withseveral alkyl groups and the strongly hydrated loops and tails

of linear polyfructose Evidence of this high stability was obtained from disjoining pressure measurements

Stabilization of suspensions using INUTECregSP1 was described with particular reference to latexes that wereprepared using emulsion polymerization Thehigh stability of thelatexes is attributedto thestrongadsorptionof

the polymeric surfactant on the particle surfaces and the enhanced steric stabilization produced by the strongly

hydrated polyfructose loops and tails Evidence for such high stability was obtained using Atomic Force

Microscopy (AFM) measurements The last part of the overview described the preparation and stabilization of

nano-emulsions using INUTECregSP1 In particular thepolymericsurfactant wasveryeffectivein reducing Ostwald

ripening as a result of its strong adsorption and the Gibbs elasticity produced by the polymeric surfactant

copy 2008 Elsevier BV All rights reserved

Contents

1 Introduction 282

11 General classi1047297cation of polymeric surfactants 282

12 Solution properties of polymeric surfactants 283

13 Adsorption and conformation of polymeric surfactants at interfaces 285

2 Theories of polymer adsorption 285

3 Experimental techni ques for studying polymeric surfactant adsorption 286

31 Measurement of the adsorption isotherm 287

32 Measurement of the fraction of segments p 287

33 Determination of the segment density distribution ρ(z) and adsorbed layer thickness δh 288

4 Examples of the adsorption isotherms of nonionic polymeric sur factants 289

5 Interaction between particles containing adsorbed polymeric surfactant layers steric stabilization 289

51 Mixing interaction Gmix 290

52 Elastic interaction Gel 290

Advances in Colloid and Interface Science 147ndash148 (2009) 281ndash299

E-mail address tharwattadrosfsnetcouk

288

289

290

290

0001-8686$ ndash see front matter copy 2008 Elsevier BV All rights reserved

doi101016jcis200810005

Contents lists available at ScienceDirect

Advances in Colloid and Interface Science

j o u r n a l h o m e p a g e w w w e l s e v i e r c o m l o c a t e c i s

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 219

6 Emulsions stabilized by polymeric surfactants 292

61 WO emulsions stabilized with PHSndashPEOndashPHS block copolymer 293

7 Suspensions stabilised using polymeric surfactants 295

71 Polymeric surfactants in emulsion polymerization 296

711 Dispersion polymerization 299

72 Polymeric surfactants for stabilisation of preformed latex dispersions

73 Use of polymeric surfactants for preparation and stabilisation of nano-emulsions

References

1 Introduction

Polymeric surfactants are essential materials for preparation of

many disperse systems of which we mention dyestuffs paper

coatings inks agrochemicals pharmaceuticals personal care pro-

ducts ceramics and detergents [1] One of the most important

applications of polymeric surfactants is in the preparation of oil-in-

water (OW) and water-in-oil (WO) emulsions as well as solidliquid

dispersions [23] In this case the hydrophobic portion of the

surfactant molecule should adsorb ldquostronglyrdquo at the OW or becomes

dissolved in the oil phase leaving the hydrophilic components in the

aqueous medium whereby they become strongly solvated by the

water molecules

The other major application of surfactants is for the preparation of

solidliquid dispersions (usually referred to as suspensions) There are

generally two methods for preparation of suspensions referred to as

condensation and dispersions methods In the 1047297rst case one starts

with molecular units and build-up the particles by a process of

nucleation and growth [4] A typical example is the preparation of

polymer lattices In this case the monomer (such as styrene or

methylmethacrylate) is emulsi1047297ed in water using an anionic or non-

ionic surfactant (such as sodium dodecyl sulphate or alcohol

ethoxylate) An initiator such as potassium persulphate is added and

when the temperature of the system is increased initiation occurs

resulting in the formation of the latex (polystyrene or polymethyl-

methacrylate) In the dispersion methods preformed particles

(usually powders) are dispersed in an aqueous solution containing asurfactant The latter is essential for adequate wetting of the powder

(both external and internal surfaces of the powder aggregates and

agglomerates must be wetted) [1] This is followed by dispersion of

the powder using high speed stirrers and 1047297nally the dispersion is

ldquomilledrdquo to reduce the particle size to the appropriate range

For stabilization of emulsionsand suspensions against 1047298occulation

coalescence and Ostwald ripening the following criteria must be

satis1047297ed (i) Complete coverage of the droplets or particles by the

surfactant Any bare patches may result in 1047298occulation as a result of

van der Waals attraction or bridging (ii) Strong adsorption (or

lsquorsquoanchoringrsquorsquo) of the surfactant molecule to the surface of droplet or

particle (iii) Strong solvation (hydration) of the stabilizing chain to

provide effective steric stabilization (iv) Reasonably thick adsorbed

layer to prevent weak 1047298occulation [1]Most of the above criteria for stability are best served by using a

polymeric surfactant In particular molecules of the AndashB AndashBndashA

blocks and BAn (or ABn) grafts (see below) are the most ef 1047297cient for

stabilisation of emulsions and suspensions In this case the B chain

(referred to as the ldquoanchoringrdquo chain) is chosen to be highly insoluble

in the medium and with a high af 1047297nity to the surface in the case of

suspensions or soluble in the oil in the case of emulsions The A chain

is chosen to be highly soluble in the medium and strongly solvated by

its molecules These block and graft copolymers are ideal for

preparation of concentrated emulsions and suspensions which are

needed in many industrial applications

In this overview I will start with a section on the general

classi1047297cation of polymeric surfactants This is followed by a discussion

of their solution properties The next section will be devoted to the

adsorption of polymeric surfactants at the solidliquid interface

whereby a summary will be given to some of the theoretical

treatments and the methods that may be applied for studying

polymeric surfactant adsorption and its conformation at the SL

interface The same principles may be applied to the adsorption of

polymeric surfactants at the LL interface although theoretical

treatments of this problem are not as developed as those for the SL

interface Thenext section will be devoted to theprinciples involved in

stabilisation of emulsions and suspensions using polymeric surfac-

tants ie the general theoryof steric stabilisation Two sections will be

devoted to describe the use of polymeric surfactants for stabilisation

of suspensions and emulsions The use of polymeric surfactants for

stabilization of nano-emulsions against Ostwald ripening will also be

described

11 General classi 1047297cation of polymeric surfactants

Perhaps the simplest type of a polymeric surfactant is a homo-

polymer that is formed from the same repeating units such as poly

(ethylene oxide) or poly(vinyl pyrrolidone) These homopolymers

have little surface activity at the OW interface since the homo-

polymer segments (ethylene oxide or vinylpyrrolidone) are highly

water soluble and have little af 1047297nity to the interface However such

homopolymers may adsorb signi1047297cantly at the SL interface Even if

the adsorption energy per monomer segment to the surface is small

(fraction of kT where k is the Boltzmann constant and T is the

absolute temperature) the total adsorption energy per molecule maybe suf 1047297cient to overcome the unfavourable entropy loss of the

molecule at the SL interface

Clearly homopolymers are not the most suitable emulsi1047297ers or

dispersants A small variant is to use polymers that contain speci1047297c

groups that have high af 1047297nity to the surface This is exempli1047297ed by

partially hydrolysed poly(vinyl acetate) (PVAc) technically referred to

as poly(vinyl alcohol) (PVA) The polymer is prepared by partial

hydrolysis of PVAc leaving some residual vinyl acetate groups Most

commercially available PVA molecules contain 4ndash12 acetate groups

These acetate groups which are hydrophobic give the molecule its

amphipathic character On a hydrophobic surface such as polystyrene

the polymer adsorbs with preferential attachment of the acetate

groups on the surface leaving the more hydrophilic vinyl alcohol

segments dangling in the aqueous medium These partially hydro-lysed PVA molecules also exhibit surface activity at the OW interface

[5]

The most convenient polymeric surfactants are those of the block

and graft copolymer type A block copolymer is a linear arrangement

of blocks of variable monomer composition The nomenclature for a

diblock is poly-A-block-poly-B and for a triblock is poly-A-block-poly-

B-poly-A One of the most widely used triblock polymeric surfactants

are the ldquoPluronicsrdquo (BASF Germany) or ldquoSynperonic PErdquo (ICI UK)

which consists of two poly-A blocks of poly(ethylene oxide) (PEO) and

one block of poly(propylene oxide) (PPO) Several chain lengths of PEO

and PPO are available More recently triblocks of PPOndashPEOndashPPO

(inverse Pluronics) became available for some speci1047297c applications

The above polymeric triblocks can be applied as emulsi1047297ers or

dispersants whereby the assumption is made that the hydrophobicPPO

291

292

293

294

296

296

297

300

282 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 319

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 419

Clearly when χb12 B2 is positive and mixing is non-ideal leading to

positive deviation (repulsion) this occurs when the polymer chains

are in ldquogoodrdquo solvent conditions In contrast when χN12 B2 is

negative and mixing is non-ideal leading to negative deviation

(attraction) this occurs when the polymer chains are in ldquopoorrdquo

solvent conditions (precipitation of the polymer may occur under

these conditions) Since the polymer solvency depends on tempera-

ture one can also de1047297ne a theta temperature θ at which χ=12

The function [(12)minus

χ] can also be expressed in terms of twomixing parameters an enthalpy parameter κ 1 and an entropy

parameter ψ1 ie

1

2minusχ

= κ 1minusψ1 eth8THORN

The θ-temperature can also be de1047297ned in terms of κ 1 and ψ1

θ = κ 1T

ψ1

eth9THORN

Alternatively one can write

1

2

minusχ = ψ1 1minusθ

T eth10THORN

Although the FloryndashHuggins theory is sound in principle several

experimental results cannot be accounted for For example it was

found that the χ parameter depends on the polymer concentration in

solution Most serious is the fact that many polymer solutions (such as

PEO) show phase separation on heating when the theory predictsthat

it should happen only on cooling

The solution properties of copolymers are much more complicated

This is due to the fact that the two copolymer components A and B

behave differently in different solvents Only when the two compo-

nents are both soluble in the same solvent then they exhibit similar

solution properties This is the case for example for a non-polar

copolymer in a non-polar solvent It should also be emphasised that

the FloryndashHuggins theory was developed for ideal linear polymers

Indeed with branched polymers consisting of high monomer density

(eg star branched polymers) the θ-temperature depends on the

lengthof the armsand isin general lower thanthatof a linearpolymer

with the same molecular weight

Another complication arises from speci1047297c interaction with the

solvent eg hydrogen bonding between polymer and solvent

molecules (eg with PEO and PVA in water) Also aggregation insolution (lack of complete dissolution) may present another problem

One of the most useful parameters for characterising the

conformation of a polymer in solution is the root mean-square (rms)

end to end length br 2N12 which represents a con1047297guration character r

as the distance from one end group to the other of a chain molecule

Another useful parameter is the radius of gyration bs2N

12 which is a

measure of the effective size of a polymer molecule (it is the root

mean-square distance of the elements of the chain from its centre of

gravity)

For linear polymers

bs2N

1=2 = br 2N1=2

61=2 eth11THORN

The radius of gyration of a polymer in solution can be determined

from light scattering measurements

As mentioned above dilute solutions of copolymers is solvents that

are good for both components exhibit similar behaviour to hompo-

lymer chains However in a selective solvent whereby the medium is

a good solvent for one component say A and a poor solvent for the

second component B the very different solvent af 1047297nities to the two

components will have a large effect on the conformation of the

isolated chain This results in formation of aggregates involving

several macromolecules in dilute solutions (low Critical Aggregation

Concentration CAC) It is believed that the polymeric aggregates are

Fig 3 Various conformations of macromolecules on a plane surface

284 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 519

spherical [5] The CAC of these block and graft copolymers is usually

very low

Several methods may be applied to obtain the aggregate size and

shape of block and graft copolymers of which light scattering small

angle X-ray and neutron scattering are probably the most direct

Dynamic light scattering (photon correlation spectroscopy) can also

be applied to obtain the hydrodynamic radius of the aggregate This

technique is relatively easy to perform when compared with static

light scattering since it does not require special preparation of thesample

13 Adsorption and conformation of polymeric surfactants at interfaces

Understanding the adsorption and conformation of polymeric

surfactants at interfaces is key to knowing how these molecules act as

stabilizers Most basic ideas on adsorption and conformation of

polymers have been developed for the solidliquid interface [10ndash17]

The process of polymer adsorption is fairly complicated mdash it

involves PolymerSurface Interaction PolymerSolvent Interaction

SurfaceSolvent Interaction The con1047297guration (conformation) of the

polymer at the solidliquid interface The polymersurface interaction

is described in terms of adsorption energy per segment χs The

polymersolvent interaction is described in terms of the Floryndash

Huggins interaction parameter χ The polymer con1047297guration is

described by sequences of Trains segments in direct contact with

the surface Loops segments in between the trains that extend into

solution Tails ends of the molecules that also extend into solution A

schematic representation of the various polymer con1047297gurations is

given in Fig 3

For homopolymers eg poly(ethylene oxide) (PEO) or polyvinyl

pyrrolidone (PVP) a train-loop-tail con1047297guration is the case For

adsorption to occur a minimum energy of adsorption per segment χs

is required When a polymer molecule adsorbs on a surface it looses

con1047297gurational entropy and this must be compensated by an

adsorption energy χs per segment This is schematically shown in

Fig 4 where the adsorbed amount Γ is plotted versus χs The

minimum value of χs can be very small (b01kT ) since a large number

of segments per molecule are adsorbed For a polymer with say 100segments and 10 of these are in trains the adsorption energy per

molecule now reaches 1kT (with χs =01kT ) For 1000 segments the

adsorption energy per molecule is now 10kT

Homopolymers are not the most suitable for stabilisation of

dispersions For strong adsorption one needs the molecule to be

ldquoinsolublerdquo in the medium and has strong af 1047297nity (ldquoanchoringrdquo) to the

surface For stabilisation one needs the molecule to be highly soluble

in themedium andstronglysolvated by itsmolecules mdash This requires a

FloryndashHuggins interaction parameter less than 05 The above

opposing effects can be resolved by introducing ldquoshortrdquo blocks in the

moleculewhichare insolublein themedium andhavea strong af 1047297nity

to the surface as illustrated in Fig 3b Example Partially hydrolysed

polyvinyl acetate (88 hydrolysed ie with 12 acetate groups)

usually referred to as polyvinyl alcohol (PVA)

The above requirements are better satis1047297ed using AndashB AndashBndashA and

BAn graft copolymers B is chosento be highly insolublein themedium

and it should have high af 1047297nity to the surface This is essential to

ensure strong ldquoanchoringrdquo to the surface (irreversible adsorption) A is

chosen to be highly soluble in the medium and strongly solvated by

its molecules The FloryndashHuggins χ parameter can be applied in this

case For a polymer in a good solvent χ has to be lower than 05 the

smaller the χ value the better the solvent for the polymer chains

Examples of B for a hydrophobic particles in aqueous media are

polystyrene polymethylmethacrylate Examples of A in aqueous

media are polyethylene oxide polyacrylic acid polyvinyl pyrollidone

and polysaccharides For non-aqueous media such as hydrocarbons

the A chain(s) could be poly(12-hydroxystearic acid)

For full description of polymer adsorption one needs to obtain

information on the following (i) The amount of polymer adsorbed Γ (in mg or moles) per unit area of the particles It is essential to know

thesurface area of theparticles in thesuspensionNitrogenadsorption

on the powder surface may give such information (by application of

the BET equation) provided there will be no change in area on

dispersing the particles in the medium For many practical systems a

change in surface area may occur on dispersing the powder in which

case one has to use dye adsorption to measure the surface area (some

assumptions have to be made in this case) (ii)The fraction of segments

in direct contact with thesurface ie the fraction of segmentsin trains

p ( p = (Number of segments in direct contact with the surface) Total

Number) (iii)The distribution of segments in loops and tails ρ(z)

which extend in several layersfrom the surface ρ(z) is usually dif 1047297cult

to obtain experimentally although recently application of small angle

neutron scattering could obtain such information An alternative anduseful parameter for assessing ldquosteric stabilisationrdquo is the hydro-

dynamic thickness δh (thickness of the adsorbed or grafted polymer

layer plus any contribution from the hydration layer) Several methods

can be applied to measure δh as will be discussed below

2 Theories of polymer adsorption

Two main approaches have been developed to treat the problem of

polymer adsorption (i) Random Walk approach This is based on

Florys treatment of the polymer chain in solution the surface was

considered as a re1047298ecting barrier (ii) Statistical mechanical approach

Thepolymer con1047297gurationwas treated as being made of three types of

structures trains loops and tails each having a different energy state

The random walk approach is an unrealistic model to the problemof polymer adsorption since the polymer interacts in a speci1047297c

manner with the surface and the solvent The statistical mechanical

approach is a more realistic model for the problem of polymer

adsorption since it takes into account the various interactions

involved A useful model for treating polymer adsorption and

con1047297guration was suggested by Scheutjens and Fleer (SF theory)

[14ndash16] that is referred to as the step weighted random walk approach

which is summarized below

The SF theory starts without any assumption for the segment

density distribution The partition functions were derived for the

mixture of free and adsorbed polymer molecules as well as for the

solvent molecules All chain conformations were described as step

weighted random walks on a quasi-crystalline lattice which extends in

parallel layers from the surface mdash

this is schematically shown in Fig 5Fig 4 Variation of adsorption amount Γ with adsorption energy per segment χs

285T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 619

The partition function is written in terms of a number of con1047297gura-

tions These were treated as connectedsequencesof segments In each

layer random mixing between segments and solvent molecules was

assumed (mean 1047297eld approximation) Each step in the random walk

was assigned a weighting factor pi

pi consists of three contributions (i) An adsorption energyχs (which

exists only for the segments that are near the surface) (ii) Con1047297gura-

tional entropy of mixing (that exists in each layer) (iii) Segment-solvent

interaction parameterχ (the FloryndashHuggins interaction parameter note

that χ=0 for an athermal solvent χ=05 for a θ-solvent)

From the weighting factors the statistical weight of any chain

conformation can be calculated for a given segment concentration

pro1047297le using a matrix formulation Fig 6 shows typical adsorption

isotherms plotted as surface coverage (in equivalent monolayers)

versus polymer volume fraction ϕ in bulk solution (ϕ was taken to

vary between 0 and 10minus3 which is the normal experimental range)

The results in Fig 6 show the effect of increasing the chain length r

and effect of solvency (athermal solvent with χ=0 and theta solvent

with χ=05) The adsorption energy χs was taken to be the same and

equal to 1kT When r = 1 θ is very small and the adsorption increases linearly

with increase of ϕ (Henrys type isotherm) On the other hand when

r = 10 the isotherm deviates much from a straight line and approaches

a Langmuirian type However when r ge20 high af 1047297nity isotherms are

obtained This implies that the 1047297rst added polymer chains are

completely adsorbed resulting in extremely low polymer concentra-

tion in solution (approaching zero) This explains the irreversibility of

adsorption of polymeric surfactants with r N100 The adsorption

isotherms with r =100 and above are typical of those observed

experimentally for most polymers that are not too polydisperse ie

showing a steep rise followed by a nearly horizontal plateau (which

only increases few percent per decade increase of ϕ) In these dilute

solutions the effect of solvency is most clearly seen with poor

solvents giving the highest adsorbed amounts In good solvents θ is

much smaller and levels off for long chains to attain an adsorption

plateau which is essentially independent of molecular weight

Another point that emerges from the SF theory is the difference in

shape between the experimental and theoretical adsorption isotherms

in thelow concentration region Theexperimentalisotherms areusuallyrounded whereas those predicted fromtheory are1047298at This is accounted

for in terms of the molecular weight distribution (polydispersity) which

is encountered with most practical polymers This effect has been

explained by CohenndashStuart et al [17] With polydisperse polymers the

larger molecular weight fractions adsorb preferentially over the smaller

ones At low polymer concentrations nearly all polymer fractions are

adsorbed leaving a small fraction of the polymer with the lowest

molecular weights in solution As the polymer concentration is

increased the higher molecular weight fractions displace the lower

ones on thesurfacewhichare nowreleasedin solution thusshiftingthe

molecular weight distribution in solution to lower values This process

continues with further increase in polymer concentration leading to

fractionation whereby the higher molecular weight fractions are

adsorbed at the expense of the lower fractions which are released to

the bulk solution However in very concentrated solutions monomers

adsorb preferentially with respect to polymers and short chains with

respect to larger ones This is due to the fact that in this region the

conformational entropy term predominates the free energy disfavour-

ing the adsorption of long chains

According to the SF theory the bound fraction p is high at low

concentration and relatively independent of molecular weight when

r N20 However with increase in surface coverage andor molecular

weight p tends to decrease indicating the formation of larger loops

and tails

The structure of the adsorbed layer is described in terms of the

segment density distribution As an illustration Fig 7 shows some

calculations using the SF theory for loops and tails with r =1000

ϕ= 10minus6 and χ=05 In this example 38 of the segments are in

trains 555 in loops and 65 in tails This theory demonstrates theimportance of tails which dominate the total distribution in the outer

region

3 Experimental techniques for studying polymeric surfactant

adsorption

As mentioned above for full characterization of polymeric sur-

factant adsorption one needs to determine three parameters (i) The

adsorbed amount Γ (mg mminus2 or mol mminus2) as a function of equilibrium

concentration C eq ie the adsorption isotherm (ii) The fraction of

segments in direct contact with the surface p (number of segments in

Fig 5 Schematic representation of a polymer molecule adsorbing on a 1047298at surface mdash

quasi-crystalline lattice with segments 1047297lling layers that are parallel to the surface

(random mixing of segments and solvent molecules in each layer is assumed)

Fig 6 Adsorption isotherms for oligomers and polymers in the dilute region based on

the SF theory Fig 7 Loop tail and total segment pro1047297le according to the SF theory

286 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 719

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 819

The microelectrophoresis technique is based on measurement of

the electrophoretic mobility u of the particles in the presence and

absence of the polymer layer From u one can calculate the zeta

potential ζ using the Huckel equation (which is applicable for small

particles and extended double layers ie κ Rbb1 where κ is the

DebyendashHuckel parameter that is related to the salt concentration

u = 23eeon η

eth15THORN

where ε is the relative permittivity of the medium and ε o is the

permittivity of free space

By measuring ζ of the particles with and without the adsorbed

polymer layer one can obtain the hydrodynamic thickness δh For

accurate measurements one should carry the measurements at various

electrolyte concentrations and extrapolate the results to the plateau

value Several automatic instruments are available for measurement of

the electrophoretic mobility Malvern zeta sizerndashCoulter Delsa sizer and

Broekahven instrument All these instruments are easy to use and the

measurement can be carried out within few minutes

4 Examples of the adsorption isotherms of nonionic polymericsurfactants

Fig 9 shows the adsorption isotherms for PEO with different

molecular weights on PS (at room temperature It can be seen that the

amount adsorbed in mgm-2 increases with increase in the polymer

molecular weight Fig 10 shows the variation of the hydrodynamic

thickness δh with molecular weight M δh shows a linear increase with

log M δh increases with n the number of segments in the chain

according to

δhasympn08 eth16THORN

Fig 11 shows the adsorption isotherms of PVA with various

molecular weights on PS latex (at 25 degC) [19] The polymers were

obtained by fractionation of a commercial sample of PVA with an

average molecular weight of 45000 The polymer also contained 12

vinyl acetate groups As with PEO the amount of adsorption increases

with increase in M The isotherms are also of the high af 1047297nity type Γ at the plateau increases linearly with M 12

The hydrodynamic thickness was determined using PCS and theresults are given below

M 67000 43000 28000 17000 8000

δhnm 255 197 140 98 33

δh seems to increase linearly with increase in the molecular weight

The effect of solvency on adsorption was investigated by increasing

the temperature (the PVA molecules are less soluble at higher

temperature) or addition of electrolyte (KCl) [20] The results are

shown in Figs 12 and 13 for M =65100 As can be seen from Fig 12

increase of temperature results in reduction of solvency of the

mediumfor the chain (due to break down of hydrogen bonds) and this

results in an increase in the amount adsorbed Addition of KCl (which

reduces the solvency of the medium for the chain) results in anincrease in adsorption (as predicted by theory)

The adsorption of block and graft copolymers is more complex

since the intimate structure of the chain determines the extent of

adsorption [18] Randomcopolymers adsorbin an intermediate way to

that of the corresponding homopolymers Blockcopolymers retain the

adsorption preference of the individual blocks The hydrophilic block

(eg PEO) the buoy (previously referred to as the A chain) extends

away from the particle surface into the bulk solution whereas the

hydrophobic anchor block (previously referred to as the B chain) (eg

PS or PPO) provides 1047297rm attachment to the surface Fig 14 shows the

theoretical prediction of diblock copolymer adsorption according to

Fig 9 Adsorption isotherms for PEO on PS

Fig 10 Hydrodynamic thickness of PEO on PS as a function of the molecular weight

Fig 11 Adsorption isotherms of PVA with different molecular weights on polystyrene

latex at 25 degC

Fig 12 In1047298uence of temperature on adsorption

288 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 919

the Scheutjens and Fleer theory The surface density σ is plotted

versus the fraction of anchor segmentsν A The adsorption depends on

the anchorbuoy composition

The amount of adsorption is higher than for homopolymers and

the adsorbed layer thickness is more extended and dense Fig 15

shows the theoretical prediction for the adsorbed layer thickness δwhich is plotted as a function of ν A For a triblock copolymer AndashBndashA

with two buoy chains (A) and one anchor chain (B) the behaviour is

similar to that of diblock copolymers This is shown in Fig16 for PEOndash

PPOndashPEO block (Pluronic)

5 Interaction between particles containing adsorbed polymeric

surfactant layers steric stabilization

When two particles each with a radius R and containing an

adsorbed polymer layer with a hydrodynamic thickness δh approach

each other to a surfacendashsurface separation distance h that is smaller

than 2 δh the polymer layers interact with each other resulting in two

main situations [21] (i) The polymer chains may overlap with each

other (ii) The polymer layer may undergo some compression In both

cases there will be an increase in the local segment density of the

polymer chains in the interaction region This is schematicallyillustrated in Fig 17 The real situation is perhaps in between the

above two cases ie the polymer chains may undergo some

interpenetration and some compression

Provided the dangling chains (the A chains in AndashB AndashBndashA block or

BAn graft copolymers) are in a good solvent this local increase in

segment density in the interaction zone will result in strong repulsion

as a result of two main effects (i) Increase in the osmotic pressure in

the overlap region as a result of the unfavourable mixing of the

polymer chains when these are in good solvent conditions This is

referred to as osmotic repulsion or mixing interaction and it is

described by a free energy of interaction Gmix (ii) Reduction of the

con1047297gurational entropy of the chains in the interaction zone this

entropy reduction results from the decrease in the volume available

for thechains when these areeither overlapped or compressed This is

referred to as volume restriction interaction entropic or elastic

interaction and it is described by a free energy of interaction Gel

Combination of Gmix and Gel is usually referred to as the steric

interaction free energy Gs ie

Gs = Gmix + Gel eth17THORN

The sign of Gmix depends on the solvency of the medium for the

chains If in a good solvent ie the FloryndashHuggins interaction

parameter χ is less than 05 then Gmix is positive and the mixing

interaction leads to repulsion (see below) In contrast if χN05 (ie the

chains are in a poor solvent condition) Gmix is negative and the

mixing interaction becomes attractive Gel is always positive and

hence in some cases one can produce stable dispersions in a relatively

poor solvent (enhanced steric stabilisation)

51 Mixing interaction Gmix

This results from the unfavourable mixing of the polymer chainswhen these are in a good solvent conditions This is schematically

shown in Fig 18

Consider two spherical particles with the same radius and each

containing an adsorbed polymer layer with thickness δ Before

overlap one can de1047297ne in each polymer layer a chemical potential

for the solvent μ iα and a volume fraction for the polymer in the

layerϕ2 In the overlap region (volume element dV ) the chemical

potential of the solvent is reduced to μ i β This results from the increase

in polymer segment concentration in this overlap region

In the overlap region the chemical potential of the polymer chains

is now higher than in the rest of the layer (with no overlap) This

Fig 13 In1047298uence of addition of KCl on adsorption

Fig 14 Prediction of Adsorption of diblock copolymer

Fig15 Theoretical predictions of the adsorbed layer thickness for a diblock copolymer

Fig 16 Adsorbed amount (mg mminus2) versus fraction of anchor segment for an AndashBndashA

triblock copolymer (PEOndash

PPOndash

PEO)

289T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1019

amounts to an increase in the osmotic pressure in the overlap region

as a result solvent will diffuse from thebulk to the overlap region thus

separating the particles and hence a strong repulsive energy arises

from this effect The above repulsive energy can be calculated by

considering the free energy of mixing of two polymer solutions as for

example treated by Flory and Krigbaum [22] The free energy of

mixing is given by two terms (i) An entropy term that depends on the

volume fraction of polymer and solvent (ii)An energy term that is

determined by the FloryndashHuggins interaction parameter χ

Using the above theory one can derive an expression for the free

energy of mixing of two polymer layers (assuming a uniform segment

density distribution in each layer) surrounding two spherical particles

as a function of the separation distance h between the particles The

expression for Gmix is

Gmix

kT =

2V 22V 1

m2

2

1

2minusχ

δminus

h

2

2

3R + 2δ + h

2

eth18THORN

k is the Boltzmann constant T is the absolute temperature V 2 is the

molar volume of polymer V 1 is the molar volume of solvent and ν 2 is

the number of polymer chains per unit area

The sign of Gmix depends on the value of the FloryndashHuggins

interaction parameter χ if χb05 Gmix is positive and the interaction

is repulsive if χN05 Gmix is negative and the interaction is attractiveif χ=05 Gmix =0 and this de1047297nes the θ-condition

52 Elastic interaction Gel

Thisarises fromthe lossin con1047297gurational entropy of thechains on

the approach of a second particle As a result of this approach the

volume available for the chains becomes restricted resulting in loss of

the number of con1047297gurations This can be illustrated by considering a

simple molecule represented by a rod that rotates freely in a

hemisphere across a surface (Fig 19) When the two surfaces are

separated byan in1047297nitedistanceinfin thenumber of con1047297gurations of the

rod is Ω(infin) which is proportional to the volume of the hemisphere

When a secondparticleapproaches to a distance h suchthatit cuts the

hemisphere (loosing some volume) the volume available to the chains

is reduced and the number of con1047297gurations become Ω(h) which is

less than Ω(infin) For two 1047298at plates Gel is given by the following

expression

Gel

kT = 2m2 ln

X heth THORN

X infineth THORN

= 2m2Rel heth THORN eth19THORN

where Rel

(h) is a geometric function whose form depends on the

segment density distribution It should be stressed that Gel is always

positive and could play a major role in steric stabilisation It becomes

very strong when the separation distance between the particles

becomes comparable to the adsorbed layer thickness δ

Combination of Gmix and Gel with GA (the van der Waals attractive

energy)gives the total free energy of interaction GT (assuming there is

no contribution from any residual electrostatic interaction) ie

GT = Gmix + Gel + GA eth20THORN

A schematic representation of the variation of Gmix Gel GA and GT

with surfacendashsurface separation distance h is shown in Fig 20

Gmix increases very sharply with decrease of h when hb2δ Gel

increases very sharply with decrease of h when hbδ GT versus h

shows a minimum Gmin at separation distances comparable to 2δ

When h b2δ GT shows a rapid increase with decrease in h

The depth of the minimum depends on the Hamaker constant A

the particle radius R and adsorbed layer thickness δ Gmin increases

with increase of A and R At a given A and R Gmin increases with

decrease in δ (ie with decrease of the molecular weight M w of the

stabiliser This is illustrated in Fig 21 which shows the energyndash

distance curves as a function of δR The larger the value of δR the

smaller the value of Gmin In this case the system may approach

thermodynamic stability as is the case with nano-dispersions

6 Emulsions stabilized by polymeric surfactants

The most effective method for emulsion stabilization is to use

polymeric surfactants that stronglyadsorb at the OWor WO interface

Fig 17 Schematic representation of the interaction between particles containing adsorbed polymer layers

Fig 18 Schematic representation of polymer layer overlap

Fig 19 Schematic representation of con1047297gurational entropy loss on approach of a

second particle

290 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1119

and produce effective steric stabilization against strong 1047298occulation

coalescence and Ostwald ripening [23]

As mentioned above a graft copolymer of the ABn type was

synthesized by grafting several alkyl groups on an inulin (polyfruc-

tose) chain The polymeric surfactant (INUTECregSP1) consists of a

linear polyfructose chain (the stabilizing A chain) and several alkyl

groups (the B chains) that provide multi-anchor attachment to the oil

droplets This polymeric surfactant produces enhanced steric stabili-

zation both in water and high electrolyte concentrations as will be

discussed later

For water-in-oil (WO) emulsions an AndashBndashA block copolymer of

poly (12-hydroxystearic acid) (PHS) (the A chains) and poly (ethylene

oxide) (PEO) (the B chain) PHSndashPEOndashPHS is commercially available

(Arlacel P135 UNIQEMA) The PEO chain (that is soluble in the water

droplets) forms the anchor chain whereas the PHS chains form the

stabilizing chains PHS is highly soluble in most hydrocarbon solvents

and is strongly solvated by its molecules The structure of the PHSndash

PEOndashPHS block copolymer is schematically shown in Fig 22The conformation of the polymeric surfactant at the WO interface

is schematically shown in Fig 23

Emulsions of Isopar Mwater and cyclomethiconewater were

prepared using INUTECregSP1 5050 (vv) OW emulsions were

prepared and the emulsi1047297er concentration was varied from 025 to 2

(wv) based on the oil phase 05 (wv) emulsi1047297erwas suf 1047297cient for

stabilization of these 5050 (vv) emulsions [23]

The emulsions were stored at room temperature and 50 degC and

optical micrographs were taken at intervals of time (for a year) in

order to check the stability Emulsions prepared in water were very

stableshowingno change in droplet size distributionover more than a

year period and this indicated absence of coalescence Any weak

1047298occulation that occurred was reversible and the emulsion could be

redispersed by gentle shaking Fig 24 shows an optical micrograph fora dilute 5050 (vv) emulsion that was stored for 15 and 14 weeks at

50 degC

No change in droplet size was observed after storage for more than

1 year at 50 degC indicating absence of coalescence The same result was

obtained when using different oils Emulsions were also stable against

coalescence in the presence of high electrolyte concentrations (up to

4 mol dmminus3 or ~25 NaCl

The above stability in high electrolyte concentrations is not

observed with polymeric surfactants based on polethylene oxide

The high stability observed using INUTECregSP1 is related to its

strong hydration both in water and in electrolyte solutions The

hydration of inulin (the backbone of HMI) could be assessed using

cloud point measurements A comparison was also made with PEO

with two molecular weights namely 4000 and 20000

Solutionsof PEO 4000 and 20000 showeda systematic decrease of

cloud point with increase in NaCl or MgSO4 concentration In contrast

inulin showed no cloud point up to 4 mol dmminus3 NaCl and up to 1 mol

dmminus3 MgSO4

The above results explain the difference between PEO and inulin

With PEO the chains show dehydration when the NaCl concentration

is increased above 2 mol dmminus3 or 05 mol dmminus3 MgSO4 The inulin

chains remain hydrated at much higher electrolyte concentrations It

seems that the linear polyfructose chains remain strongly hydrated athigh temperature and high electrolyte concentrations

The high emulsion stability obtained whenusingINUTECregSP1 can be

accounted for by the following factors (i) The multi-point attachment

of the polymer by several alkyl chains that are grafted on the backbone

(ii) The strong hydration of the polyfructose ldquoloopsrdquo both in water and

high electrolyte concentrations (χ remains below 05 under these

conditions) (iii) Thehigh volumefraction (concentration) of the loops at

the interface (iv) Enhanced steric stabilization this is the case with

multi-point attachment which produces strong elastic interaction

Evidence for the high stability of the liquid 1047297lm between emulsion

droplets when using INUTECregSP1 was obtained by Exerowa et al [24]

using disjoining pressure measurements This is illustrated in Fig 25

which shows a plot of disjoining pressure versus separation distance

between two emulsion droplets at various electrolyte concentrations

The results show that by increasing the capillary pressure a stable

Newton Black Film (NBF) is obtained at a 1047297lm thickness of sim7 nm

The lack of rupture of the 1047297lm at the highest pressure applied of

45times104 Pa indicate the high stability of the 1047297lm in water and in high

electrolyte concentrations (up to 20 mol dmminus3 NaCl)

The lack of rupture of the NBF up to the highest pressure applied

namely 45times104 Pa clearly indicatesthe high stability of the liquid1047297lm

in the presence of high NaCl concentrations (up to 2 mol dm minus3) This

result is consistent with the high emulsion stability obtained at high

electrolyte concentrations and high temperature Emulsions of Isopar

M-in-water are very stable under such conditions and this could be

accounted for by the high stability of the NBF The droplet size of

5050 OW emulsions prepared using 2 INUTEC regSP1 is in the region

of 1ndash10 μ m This corresponds to a capillary pressure of ~3times104 Pa for

the 1 μ m drops and ~3times103 Pa for the 10 μ m drops These capillarypressures are lower than those to which the NBF have been sub-

jected to and this clearly indicates the high stability obtained against

coalescence in these emulsions

61 WO emulsions stabilized with PHS ndashPEOndashPHS block copolymer

WO emulsions (the oil being Isopar M) were prepared using PHSndash

PEOndashPHS block copolymer at high water volume fractions (N07) The

emulsions have a narrow droplet size distribution with a z -average

radius of 183 nm [25] They also remained 1047298uid up to high water

volume fractions (N06) This could be illustrated from viscosityndash

volume fraction curves as is shown in Fig 26

Fig 20 Energyndashdistance curves for sterically stabilized systems

Fig 21 Variation of Gmin with δR

291T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1219

The effective volume fraction ϕeff of the emulsions (the core

droplets plus the adsorbed layer) could be calculated from the relative

viscosity and using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth21THORN

Where η r is the relative viscosity ϕp is the maximum packing fraction(sim07) and [ η ] is the intrinsic viscosity that is equal to 25 for hard-

spheres

The calculations based on Eq (22) are shown in Fig 27 (square

symbols) From the effective volume fraction ϕeff and the core volume

fraction ϕ the adsorbed layer thickness could be calculated This was

found to be in the region of 10 nm at ϕ =04 and it decreased with

increase in ϕ

The WO emulsions prepared using the PHSndashPEOndashPHS block

copolymer remained stable both at room temperature and 50 degC

This is consistent with the structure of the block copolymer the B

chain (PEO) is soluble in water and it forms a very strong anchor at the

WO interface The PHS chains (the A chains) provide effective steric

stabilization since the chains are highly soluble in Isopar M and are

strongly solvated by its molecules

7 Suspensions stabilised using polymeric surfactants

There are generally two procedures for preparation of solidliquid

dispersions

(i) Condensation methods build-up of particles from mole-

cular units ie nucleation and growth A special procedure

is the preparation of latexes by emulsion or dispersion

polymerization

(ii) Dispersion methods in this case one starts with preformed

large particles or crystals which are dispersed in the liquid by

using a surfactant (wetting agent) with subsequent breaking up

of the large particles by milling (comminution) to achieve the

desirable particle size distribution A dispersing agent (usuallya

polymeric surfactant) is used for the dispersion process and

subsequent stabilization of the resulting suspension

There are generally two procedures for preparation of latexes

(i) Emulsion polymerization the monomers that are essentially

insoluble in the aqueous medium are emulsi1047297ed using a

surfactant and an initiator is added while heating the system

to produce the polymer particles that are stabilized electro-

statically (when using ionic surfactants) or sterically (when

using non-ionic surfactants)

(ii) Dispersion polymerization the monomers are dissolved in a

solvent in which the resulting polymer particlesare insoluble A

protective colloid (normally a block or graft copolymer) is

added to prevent 1047298occulation of the resulting polymers

particles that are produced on addition of an initiator This

method is usually applied for the production of non-aqueous

latex dispersions and is sometimes referred to as Non-Aqueous

Dispersion Polymerization (NAD)

Surfactants play a crucial role in the process of latex preparation

since they determine the stabilizing ef 1047297ciency and the effectiveness of

the surface active agent ultimately determines the number of particles

Fig 22 Schematic representation of the structure of PHSndashPEOndashPHS block copolymer

Fig 23 Conformation of PHSndash

PEOndash

PHS polymeric surfactant at the WO interface

292 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1319

and their size The effectiveness of any surface active agent in

stabilizing the particles is the dominant factor and the number of

micelles formed is relatively unimportant In the NAD process themonomer normally an acrylic is dissolved in a non-aqueous solvent

normally an aliphatic hydrocarbon and an oil soluble initiator and a

stabilizer (to protect the resulting particles from1047298occulation) is added

to the reaction mixture The most successful stabilizers used in NAD

are block and graft copolymers Preformed graft stabilizers based

on poly(12-hydroxy stearic acid) (PHS) are simple to prepare and

effective in NAD polymerization

Dispersion methods are used for the preparation of suspensions of

preformed particles The role of surfactants (or polymers) in the

dispersion process can be analyzed in terms of the three processesinvolved (i) Wetting of the powder by the liquid (ii) Breaking of the

aggregates and agglomerates (iii) Comminution of the resulting

particles and their subsequent stabilization All these processes are

affected by surfactants or polymers which adsorb on the powder

surface thus aiding the wetting of the powder break-up of the

aggregates and agglomerates and subsequent reduction of particle

size by wet milling

71 Polymeric surfactants in emulsion polymerization

Recently the graft copolymer of hydrophobically modi1047297ed inulin

(INUTECreg SP1) has been used in emulsion polymerization of styrene

methyl methacrylate butyl acrylate and several other monomers [26]All lattices were prepared by emulsion polymerisation using potas-

sium persulphate as initiator The z -average particle size was

determined by photon correlation spectroscopy (PCS) and electron

micrographs were also taken

Fig 24 Optical micrographs of OW emulsions stabilized with INUTEC regSP1 stored at

50 degC for 15 weeks (a) and 14 weeks (b)

Fig 25 Variation of disjoining pressure with equivalent 1047297lm thickness at various NaCl concentrations

Fig 26 Viscosityndashvolume fraction for WO emulsion stabilized with PHSndashPEOndashPHS

block copolymer experimental data calculated using Eq (22)

293T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 219

6 Emulsions stabilized by polymeric surfactants 292

61 WO emulsions stabilized with PHSndashPEOndashPHS block copolymer 293

7 Suspensions stabilised using polymeric surfactants 295

71 Polymeric surfactants in emulsion polymerization 296

711 Dispersion polymerization 299

72 Polymeric surfactants for stabilisation of preformed latex dispersions

73 Use of polymeric surfactants for preparation and stabilisation of nano-emulsions

References

1 Introduction

Polymeric surfactants are essential materials for preparation of

many disperse systems of which we mention dyestuffs paper

coatings inks agrochemicals pharmaceuticals personal care pro-

ducts ceramics and detergents [1] One of the most important

applications of polymeric surfactants is in the preparation of oil-in-

water (OW) and water-in-oil (WO) emulsions as well as solidliquid

dispersions [23] In this case the hydrophobic portion of the

surfactant molecule should adsorb ldquostronglyrdquo at the OW or becomes

dissolved in the oil phase leaving the hydrophilic components in the

aqueous medium whereby they become strongly solvated by the

water molecules

The other major application of surfactants is for the preparation of

solidliquid dispersions (usually referred to as suspensions) There are

generally two methods for preparation of suspensions referred to as

condensation and dispersions methods In the 1047297rst case one starts

with molecular units and build-up the particles by a process of

nucleation and growth [4] A typical example is the preparation of

polymer lattices In this case the monomer (such as styrene or

methylmethacrylate) is emulsi1047297ed in water using an anionic or non-

ionic surfactant (such as sodium dodecyl sulphate or alcohol

ethoxylate) An initiator such as potassium persulphate is added and

when the temperature of the system is increased initiation occurs

resulting in the formation of the latex (polystyrene or polymethyl-

methacrylate) In the dispersion methods preformed particles

(usually powders) are dispersed in an aqueous solution containing asurfactant The latter is essential for adequate wetting of the powder

(both external and internal surfaces of the powder aggregates and

agglomerates must be wetted) [1] This is followed by dispersion of

the powder using high speed stirrers and 1047297nally the dispersion is

ldquomilledrdquo to reduce the particle size to the appropriate range

For stabilization of emulsionsand suspensions against 1047298occulation

coalescence and Ostwald ripening the following criteria must be

satis1047297ed (i) Complete coverage of the droplets or particles by the

surfactant Any bare patches may result in 1047298occulation as a result of

van der Waals attraction or bridging (ii) Strong adsorption (or

lsquorsquoanchoringrsquorsquo) of the surfactant molecule to the surface of droplet or

particle (iii) Strong solvation (hydration) of the stabilizing chain to

provide effective steric stabilization (iv) Reasonably thick adsorbed

layer to prevent weak 1047298occulation [1]Most of the above criteria for stability are best served by using a

polymeric surfactant In particular molecules of the AndashB AndashBndashA

blocks and BAn (or ABn) grafts (see below) are the most ef 1047297cient for

stabilisation of emulsions and suspensions In this case the B chain

(referred to as the ldquoanchoringrdquo chain) is chosen to be highly insoluble

in the medium and with a high af 1047297nity to the surface in the case of

suspensions or soluble in the oil in the case of emulsions The A chain

is chosen to be highly soluble in the medium and strongly solvated by

its molecules These block and graft copolymers are ideal for

preparation of concentrated emulsions and suspensions which are

needed in many industrial applications

In this overview I will start with a section on the general

classi1047297cation of polymeric surfactants This is followed by a discussion

of their solution properties The next section will be devoted to the

adsorption of polymeric surfactants at the solidliquid interface

whereby a summary will be given to some of the theoretical

treatments and the methods that may be applied for studying

polymeric surfactant adsorption and its conformation at the SL

interface The same principles may be applied to the adsorption of

polymeric surfactants at the LL interface although theoretical

treatments of this problem are not as developed as those for the SL

interface Thenext section will be devoted to theprinciples involved in

stabilisation of emulsions and suspensions using polymeric surfac-

tants ie the general theoryof steric stabilisation Two sections will be

devoted to describe the use of polymeric surfactants for stabilisation

of suspensions and emulsions The use of polymeric surfactants for

stabilization of nano-emulsions against Ostwald ripening will also be

described

11 General classi 1047297cation of polymeric surfactants

Perhaps the simplest type of a polymeric surfactant is a homo-

polymer that is formed from the same repeating units such as poly

(ethylene oxide) or poly(vinyl pyrrolidone) These homopolymers

have little surface activity at the OW interface since the homo-

polymer segments (ethylene oxide or vinylpyrrolidone) are highly

water soluble and have little af 1047297nity to the interface However such

homopolymers may adsorb signi1047297cantly at the SL interface Even if

the adsorption energy per monomer segment to the surface is small

(fraction of kT where k is the Boltzmann constant and T is the

absolute temperature) the total adsorption energy per molecule maybe suf 1047297cient to overcome the unfavourable entropy loss of the

molecule at the SL interface

Clearly homopolymers are not the most suitable emulsi1047297ers or

dispersants A small variant is to use polymers that contain speci1047297c

groups that have high af 1047297nity to the surface This is exempli1047297ed by

partially hydrolysed poly(vinyl acetate) (PVAc) technically referred to

as poly(vinyl alcohol) (PVA) The polymer is prepared by partial

hydrolysis of PVAc leaving some residual vinyl acetate groups Most

commercially available PVA molecules contain 4ndash12 acetate groups

These acetate groups which are hydrophobic give the molecule its

amphipathic character On a hydrophobic surface such as polystyrene

the polymer adsorbs with preferential attachment of the acetate

groups on the surface leaving the more hydrophilic vinyl alcohol

segments dangling in the aqueous medium These partially hydro-lysed PVA molecules also exhibit surface activity at the OW interface

[5]

The most convenient polymeric surfactants are those of the block

and graft copolymer type A block copolymer is a linear arrangement

of blocks of variable monomer composition The nomenclature for a

diblock is poly-A-block-poly-B and for a triblock is poly-A-block-poly-

B-poly-A One of the most widely used triblock polymeric surfactants

are the ldquoPluronicsrdquo (BASF Germany) or ldquoSynperonic PErdquo (ICI UK)

which consists of two poly-A blocks of poly(ethylene oxide) (PEO) and

one block of poly(propylene oxide) (PPO) Several chain lengths of PEO

and PPO are available More recently triblocks of PPOndashPEOndashPPO

(inverse Pluronics) became available for some speci1047297c applications

The above polymeric triblocks can be applied as emulsi1047297ers or

dispersants whereby the assumption is made that the hydrophobicPPO

291

292

293

294

296

296

297

300

282 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 319

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 419

Clearly when χb12 B2 is positive and mixing is non-ideal leading to

positive deviation (repulsion) this occurs when the polymer chains

are in ldquogoodrdquo solvent conditions In contrast when χN12 B2 is

negative and mixing is non-ideal leading to negative deviation

(attraction) this occurs when the polymer chains are in ldquopoorrdquo

solvent conditions (precipitation of the polymer may occur under

these conditions) Since the polymer solvency depends on tempera-

ture one can also de1047297ne a theta temperature θ at which χ=12

The function [(12)minus

χ] can also be expressed in terms of twomixing parameters an enthalpy parameter κ 1 and an entropy

parameter ψ1 ie

1

2minusχ

= κ 1minusψ1 eth8THORN

The θ-temperature can also be de1047297ned in terms of κ 1 and ψ1

θ = κ 1T

ψ1

eth9THORN

Alternatively one can write

1

2

minusχ = ψ1 1minusθ

T eth10THORN

Although the FloryndashHuggins theory is sound in principle several

experimental results cannot be accounted for For example it was

found that the χ parameter depends on the polymer concentration in

solution Most serious is the fact that many polymer solutions (such as

PEO) show phase separation on heating when the theory predictsthat

it should happen only on cooling

The solution properties of copolymers are much more complicated

This is due to the fact that the two copolymer components A and B

behave differently in different solvents Only when the two compo-

nents are both soluble in the same solvent then they exhibit similar

solution properties This is the case for example for a non-polar

copolymer in a non-polar solvent It should also be emphasised that

the FloryndashHuggins theory was developed for ideal linear polymers

Indeed with branched polymers consisting of high monomer density

(eg star branched polymers) the θ-temperature depends on the

lengthof the armsand isin general lower thanthatof a linearpolymer

with the same molecular weight

Another complication arises from speci1047297c interaction with the

solvent eg hydrogen bonding between polymer and solvent

molecules (eg with PEO and PVA in water) Also aggregation insolution (lack of complete dissolution) may present another problem

One of the most useful parameters for characterising the

conformation of a polymer in solution is the root mean-square (rms)

end to end length br 2N12 which represents a con1047297guration character r

as the distance from one end group to the other of a chain molecule

Another useful parameter is the radius of gyration bs2N

12 which is a

measure of the effective size of a polymer molecule (it is the root

mean-square distance of the elements of the chain from its centre of

gravity)

For linear polymers

bs2N

1=2 = br 2N1=2

61=2 eth11THORN

The radius of gyration of a polymer in solution can be determined

from light scattering measurements

As mentioned above dilute solutions of copolymers is solvents that

are good for both components exhibit similar behaviour to hompo-

lymer chains However in a selective solvent whereby the medium is

a good solvent for one component say A and a poor solvent for the

second component B the very different solvent af 1047297nities to the two

components will have a large effect on the conformation of the

isolated chain This results in formation of aggregates involving

several macromolecules in dilute solutions (low Critical Aggregation

Concentration CAC) It is believed that the polymeric aggregates are

Fig 3 Various conformations of macromolecules on a plane surface

284 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 519

spherical [5] The CAC of these block and graft copolymers is usually

very low

Several methods may be applied to obtain the aggregate size and

shape of block and graft copolymers of which light scattering small

angle X-ray and neutron scattering are probably the most direct

Dynamic light scattering (photon correlation spectroscopy) can also

be applied to obtain the hydrodynamic radius of the aggregate This

technique is relatively easy to perform when compared with static

light scattering since it does not require special preparation of thesample

13 Adsorption and conformation of polymeric surfactants at interfaces

Understanding the adsorption and conformation of polymeric

surfactants at interfaces is key to knowing how these molecules act as

stabilizers Most basic ideas on adsorption and conformation of

polymers have been developed for the solidliquid interface [10ndash17]

The process of polymer adsorption is fairly complicated mdash it

involves PolymerSurface Interaction PolymerSolvent Interaction

SurfaceSolvent Interaction The con1047297guration (conformation) of the

polymer at the solidliquid interface The polymersurface interaction

is described in terms of adsorption energy per segment χs The

polymersolvent interaction is described in terms of the Floryndash

Huggins interaction parameter χ The polymer con1047297guration is

described by sequences of Trains segments in direct contact with

the surface Loops segments in between the trains that extend into

solution Tails ends of the molecules that also extend into solution A

schematic representation of the various polymer con1047297gurations is

given in Fig 3

For homopolymers eg poly(ethylene oxide) (PEO) or polyvinyl

pyrrolidone (PVP) a train-loop-tail con1047297guration is the case For

adsorption to occur a minimum energy of adsorption per segment χs

is required When a polymer molecule adsorbs on a surface it looses

con1047297gurational entropy and this must be compensated by an

adsorption energy χs per segment This is schematically shown in

Fig 4 where the adsorbed amount Γ is plotted versus χs The

minimum value of χs can be very small (b01kT ) since a large number

of segments per molecule are adsorbed For a polymer with say 100segments and 10 of these are in trains the adsorption energy per

molecule now reaches 1kT (with χs =01kT ) For 1000 segments the

adsorption energy per molecule is now 10kT

Homopolymers are not the most suitable for stabilisation of

dispersions For strong adsorption one needs the molecule to be

ldquoinsolublerdquo in the medium and has strong af 1047297nity (ldquoanchoringrdquo) to the

surface For stabilisation one needs the molecule to be highly soluble

in themedium andstronglysolvated by itsmolecules mdash This requires a

FloryndashHuggins interaction parameter less than 05 The above

opposing effects can be resolved by introducing ldquoshortrdquo blocks in the

moleculewhichare insolublein themedium andhavea strong af 1047297nity

to the surface as illustrated in Fig 3b Example Partially hydrolysed

polyvinyl acetate (88 hydrolysed ie with 12 acetate groups)

usually referred to as polyvinyl alcohol (PVA)

The above requirements are better satis1047297ed using AndashB AndashBndashA and

BAn graft copolymers B is chosento be highly insolublein themedium

and it should have high af 1047297nity to the surface This is essential to

ensure strong ldquoanchoringrdquo to the surface (irreversible adsorption) A is

chosen to be highly soluble in the medium and strongly solvated by

its molecules The FloryndashHuggins χ parameter can be applied in this

case For a polymer in a good solvent χ has to be lower than 05 the

smaller the χ value the better the solvent for the polymer chains

Examples of B for a hydrophobic particles in aqueous media are

polystyrene polymethylmethacrylate Examples of A in aqueous

media are polyethylene oxide polyacrylic acid polyvinyl pyrollidone

and polysaccharides For non-aqueous media such as hydrocarbons

the A chain(s) could be poly(12-hydroxystearic acid)

For full description of polymer adsorption one needs to obtain

information on the following (i) The amount of polymer adsorbed Γ (in mg or moles) per unit area of the particles It is essential to know

thesurface area of theparticles in thesuspensionNitrogenadsorption

on the powder surface may give such information (by application of

the BET equation) provided there will be no change in area on

dispersing the particles in the medium For many practical systems a

change in surface area may occur on dispersing the powder in which

case one has to use dye adsorption to measure the surface area (some

assumptions have to be made in this case) (ii)The fraction of segments

in direct contact with thesurface ie the fraction of segmentsin trains

p ( p = (Number of segments in direct contact with the surface) Total

Number) (iii)The distribution of segments in loops and tails ρ(z)

which extend in several layersfrom the surface ρ(z) is usually dif 1047297cult

to obtain experimentally although recently application of small angle

neutron scattering could obtain such information An alternative anduseful parameter for assessing ldquosteric stabilisationrdquo is the hydro-

dynamic thickness δh (thickness of the adsorbed or grafted polymer

layer plus any contribution from the hydration layer) Several methods

can be applied to measure δh as will be discussed below

2 Theories of polymer adsorption

Two main approaches have been developed to treat the problem of

polymer adsorption (i) Random Walk approach This is based on

Florys treatment of the polymer chain in solution the surface was

considered as a re1047298ecting barrier (ii) Statistical mechanical approach

Thepolymer con1047297gurationwas treated as being made of three types of

structures trains loops and tails each having a different energy state

The random walk approach is an unrealistic model to the problemof polymer adsorption since the polymer interacts in a speci1047297c

manner with the surface and the solvent The statistical mechanical

approach is a more realistic model for the problem of polymer

adsorption since it takes into account the various interactions

involved A useful model for treating polymer adsorption and

con1047297guration was suggested by Scheutjens and Fleer (SF theory)

[14ndash16] that is referred to as the step weighted random walk approach

which is summarized below

The SF theory starts without any assumption for the segment

density distribution The partition functions were derived for the

mixture of free and adsorbed polymer molecules as well as for the

solvent molecules All chain conformations were described as step

weighted random walks on a quasi-crystalline lattice which extends in

parallel layers from the surface mdash

this is schematically shown in Fig 5Fig 4 Variation of adsorption amount Γ with adsorption energy per segment χs

285T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 619

The partition function is written in terms of a number of con1047297gura-

tions These were treated as connectedsequencesof segments In each

layer random mixing between segments and solvent molecules was

assumed (mean 1047297eld approximation) Each step in the random walk

was assigned a weighting factor pi

pi consists of three contributions (i) An adsorption energyχs (which

exists only for the segments that are near the surface) (ii) Con1047297gura-

tional entropy of mixing (that exists in each layer) (iii) Segment-solvent

interaction parameterχ (the FloryndashHuggins interaction parameter note

that χ=0 for an athermal solvent χ=05 for a θ-solvent)

From the weighting factors the statistical weight of any chain

conformation can be calculated for a given segment concentration

pro1047297le using a matrix formulation Fig 6 shows typical adsorption

isotherms plotted as surface coverage (in equivalent monolayers)

versus polymer volume fraction ϕ in bulk solution (ϕ was taken to

vary between 0 and 10minus3 which is the normal experimental range)

The results in Fig 6 show the effect of increasing the chain length r

and effect of solvency (athermal solvent with χ=0 and theta solvent

with χ=05) The adsorption energy χs was taken to be the same and

equal to 1kT When r = 1 θ is very small and the adsorption increases linearly

with increase of ϕ (Henrys type isotherm) On the other hand when

r = 10 the isotherm deviates much from a straight line and approaches

a Langmuirian type However when r ge20 high af 1047297nity isotherms are

obtained This implies that the 1047297rst added polymer chains are

completely adsorbed resulting in extremely low polymer concentra-

tion in solution (approaching zero) This explains the irreversibility of

adsorption of polymeric surfactants with r N100 The adsorption

isotherms with r =100 and above are typical of those observed

experimentally for most polymers that are not too polydisperse ie

showing a steep rise followed by a nearly horizontal plateau (which

only increases few percent per decade increase of ϕ) In these dilute

solutions the effect of solvency is most clearly seen with poor

solvents giving the highest adsorbed amounts In good solvents θ is

much smaller and levels off for long chains to attain an adsorption

plateau which is essentially independent of molecular weight

Another point that emerges from the SF theory is the difference in

shape between the experimental and theoretical adsorption isotherms

in thelow concentration region Theexperimentalisotherms areusuallyrounded whereas those predicted fromtheory are1047298at This is accounted

for in terms of the molecular weight distribution (polydispersity) which

is encountered with most practical polymers This effect has been

explained by CohenndashStuart et al [17] With polydisperse polymers the

larger molecular weight fractions adsorb preferentially over the smaller

ones At low polymer concentrations nearly all polymer fractions are

adsorbed leaving a small fraction of the polymer with the lowest

molecular weights in solution As the polymer concentration is

increased the higher molecular weight fractions displace the lower

ones on thesurfacewhichare nowreleasedin solution thusshiftingthe

molecular weight distribution in solution to lower values This process

continues with further increase in polymer concentration leading to

fractionation whereby the higher molecular weight fractions are

adsorbed at the expense of the lower fractions which are released to

the bulk solution However in very concentrated solutions monomers

adsorb preferentially with respect to polymers and short chains with

respect to larger ones This is due to the fact that in this region the

conformational entropy term predominates the free energy disfavour-

ing the adsorption of long chains

According to the SF theory the bound fraction p is high at low

concentration and relatively independent of molecular weight when

r N20 However with increase in surface coverage andor molecular

weight p tends to decrease indicating the formation of larger loops

and tails

The structure of the adsorbed layer is described in terms of the

segment density distribution As an illustration Fig 7 shows some

calculations using the SF theory for loops and tails with r =1000

ϕ= 10minus6 and χ=05 In this example 38 of the segments are in

trains 555 in loops and 65 in tails This theory demonstrates theimportance of tails which dominate the total distribution in the outer

region

3 Experimental techniques for studying polymeric surfactant

adsorption

As mentioned above for full characterization of polymeric sur-

factant adsorption one needs to determine three parameters (i) The

adsorbed amount Γ (mg mminus2 or mol mminus2) as a function of equilibrium

concentration C eq ie the adsorption isotherm (ii) The fraction of

segments in direct contact with the surface p (number of segments in

Fig 5 Schematic representation of a polymer molecule adsorbing on a 1047298at surface mdash

quasi-crystalline lattice with segments 1047297lling layers that are parallel to the surface

(random mixing of segments and solvent molecules in each layer is assumed)

Fig 6 Adsorption isotherms for oligomers and polymers in the dilute region based on

the SF theory Fig 7 Loop tail and total segment pro1047297le according to the SF theory

286 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 719

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 819

The microelectrophoresis technique is based on measurement of

the electrophoretic mobility u of the particles in the presence and

absence of the polymer layer From u one can calculate the zeta

potential ζ using the Huckel equation (which is applicable for small

particles and extended double layers ie κ Rbb1 where κ is the

DebyendashHuckel parameter that is related to the salt concentration

u = 23eeon η

eth15THORN

where ε is the relative permittivity of the medium and ε o is the

permittivity of free space

By measuring ζ of the particles with and without the adsorbed

polymer layer one can obtain the hydrodynamic thickness δh For

accurate measurements one should carry the measurements at various

electrolyte concentrations and extrapolate the results to the plateau

value Several automatic instruments are available for measurement of

the electrophoretic mobility Malvern zeta sizerndashCoulter Delsa sizer and

Broekahven instrument All these instruments are easy to use and the

measurement can be carried out within few minutes

4 Examples of the adsorption isotherms of nonionic polymericsurfactants

Fig 9 shows the adsorption isotherms for PEO with different

molecular weights on PS (at room temperature It can be seen that the

amount adsorbed in mgm-2 increases with increase in the polymer

molecular weight Fig 10 shows the variation of the hydrodynamic

thickness δh with molecular weight M δh shows a linear increase with

log M δh increases with n the number of segments in the chain

according to

δhasympn08 eth16THORN

Fig 11 shows the adsorption isotherms of PVA with various

molecular weights on PS latex (at 25 degC) [19] The polymers were

obtained by fractionation of a commercial sample of PVA with an

average molecular weight of 45000 The polymer also contained 12

vinyl acetate groups As with PEO the amount of adsorption increases

with increase in M The isotherms are also of the high af 1047297nity type Γ at the plateau increases linearly with M 12

The hydrodynamic thickness was determined using PCS and theresults are given below

M 67000 43000 28000 17000 8000

δhnm 255 197 140 98 33

δh seems to increase linearly with increase in the molecular weight

The effect of solvency on adsorption was investigated by increasing

the temperature (the PVA molecules are less soluble at higher

temperature) or addition of electrolyte (KCl) [20] The results are

shown in Figs 12 and 13 for M =65100 As can be seen from Fig 12

increase of temperature results in reduction of solvency of the

mediumfor the chain (due to break down of hydrogen bonds) and this

results in an increase in the amount adsorbed Addition of KCl (which

reduces the solvency of the medium for the chain) results in anincrease in adsorption (as predicted by theory)

The adsorption of block and graft copolymers is more complex

since the intimate structure of the chain determines the extent of

adsorption [18] Randomcopolymers adsorbin an intermediate way to

that of the corresponding homopolymers Blockcopolymers retain the

adsorption preference of the individual blocks The hydrophilic block

(eg PEO) the buoy (previously referred to as the A chain) extends

away from the particle surface into the bulk solution whereas the

hydrophobic anchor block (previously referred to as the B chain) (eg

PS or PPO) provides 1047297rm attachment to the surface Fig 14 shows the

theoretical prediction of diblock copolymer adsorption according to

Fig 9 Adsorption isotherms for PEO on PS

Fig 10 Hydrodynamic thickness of PEO on PS as a function of the molecular weight

Fig 11 Adsorption isotherms of PVA with different molecular weights on polystyrene

latex at 25 degC

Fig 12 In1047298uence of temperature on adsorption

288 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 919

the Scheutjens and Fleer theory The surface density σ is plotted

versus the fraction of anchor segmentsν A The adsorption depends on

the anchorbuoy composition

The amount of adsorption is higher than for homopolymers and

the adsorbed layer thickness is more extended and dense Fig 15

shows the theoretical prediction for the adsorbed layer thickness δwhich is plotted as a function of ν A For a triblock copolymer AndashBndashA

with two buoy chains (A) and one anchor chain (B) the behaviour is

similar to that of diblock copolymers This is shown in Fig16 for PEOndash

PPOndashPEO block (Pluronic)

5 Interaction between particles containing adsorbed polymeric

surfactant layers steric stabilization

When two particles each with a radius R and containing an

adsorbed polymer layer with a hydrodynamic thickness δh approach

each other to a surfacendashsurface separation distance h that is smaller

than 2 δh the polymer layers interact with each other resulting in two

main situations [21] (i) The polymer chains may overlap with each

other (ii) The polymer layer may undergo some compression In both

cases there will be an increase in the local segment density of the

polymer chains in the interaction region This is schematicallyillustrated in Fig 17 The real situation is perhaps in between the

above two cases ie the polymer chains may undergo some

interpenetration and some compression

Provided the dangling chains (the A chains in AndashB AndashBndashA block or

BAn graft copolymers) are in a good solvent this local increase in

segment density in the interaction zone will result in strong repulsion

as a result of two main effects (i) Increase in the osmotic pressure in

the overlap region as a result of the unfavourable mixing of the

polymer chains when these are in good solvent conditions This is

referred to as osmotic repulsion or mixing interaction and it is

described by a free energy of interaction Gmix (ii) Reduction of the

con1047297gurational entropy of the chains in the interaction zone this

entropy reduction results from the decrease in the volume available

for thechains when these areeither overlapped or compressed This is

referred to as volume restriction interaction entropic or elastic

interaction and it is described by a free energy of interaction Gel

Combination of Gmix and Gel is usually referred to as the steric

interaction free energy Gs ie

Gs = Gmix + Gel eth17THORN

The sign of Gmix depends on the solvency of the medium for the

chains If in a good solvent ie the FloryndashHuggins interaction

parameter χ is less than 05 then Gmix is positive and the mixing

interaction leads to repulsion (see below) In contrast if χN05 (ie the

chains are in a poor solvent condition) Gmix is negative and the

mixing interaction becomes attractive Gel is always positive and

hence in some cases one can produce stable dispersions in a relatively

poor solvent (enhanced steric stabilisation)

51 Mixing interaction Gmix

This results from the unfavourable mixing of the polymer chainswhen these are in a good solvent conditions This is schematically

shown in Fig 18

Consider two spherical particles with the same radius and each

containing an adsorbed polymer layer with thickness δ Before

overlap one can de1047297ne in each polymer layer a chemical potential

for the solvent μ iα and a volume fraction for the polymer in the

layerϕ2 In the overlap region (volume element dV ) the chemical

potential of the solvent is reduced to μ i β This results from the increase

in polymer segment concentration in this overlap region

In the overlap region the chemical potential of the polymer chains

is now higher than in the rest of the layer (with no overlap) This

Fig 13 In1047298uence of addition of KCl on adsorption

Fig 14 Prediction of Adsorption of diblock copolymer

Fig15 Theoretical predictions of the adsorbed layer thickness for a diblock copolymer

Fig 16 Adsorbed amount (mg mminus2) versus fraction of anchor segment for an AndashBndashA

triblock copolymer (PEOndash

PPOndash

PEO)

289T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1019

amounts to an increase in the osmotic pressure in the overlap region

as a result solvent will diffuse from thebulk to the overlap region thus

separating the particles and hence a strong repulsive energy arises

from this effect The above repulsive energy can be calculated by

considering the free energy of mixing of two polymer solutions as for

example treated by Flory and Krigbaum [22] The free energy of

mixing is given by two terms (i) An entropy term that depends on the

volume fraction of polymer and solvent (ii)An energy term that is

determined by the FloryndashHuggins interaction parameter χ

Using the above theory one can derive an expression for the free

energy of mixing of two polymer layers (assuming a uniform segment

density distribution in each layer) surrounding two spherical particles

as a function of the separation distance h between the particles The

expression for Gmix is

Gmix

kT =

2V 22V 1

m2

2

1

2minusχ

δminus

h

2

2

3R + 2δ + h

2

eth18THORN

k is the Boltzmann constant T is the absolute temperature V 2 is the

molar volume of polymer V 1 is the molar volume of solvent and ν 2 is

the number of polymer chains per unit area

The sign of Gmix depends on the value of the FloryndashHuggins

interaction parameter χ if χb05 Gmix is positive and the interaction

is repulsive if χN05 Gmix is negative and the interaction is attractiveif χ=05 Gmix =0 and this de1047297nes the θ-condition

52 Elastic interaction Gel

Thisarises fromthe lossin con1047297gurational entropy of thechains on

the approach of a second particle As a result of this approach the

volume available for the chains becomes restricted resulting in loss of

the number of con1047297gurations This can be illustrated by considering a

simple molecule represented by a rod that rotates freely in a

hemisphere across a surface (Fig 19) When the two surfaces are

separated byan in1047297nitedistanceinfin thenumber of con1047297gurations of the

rod is Ω(infin) which is proportional to the volume of the hemisphere

When a secondparticleapproaches to a distance h suchthatit cuts the

hemisphere (loosing some volume) the volume available to the chains

is reduced and the number of con1047297gurations become Ω(h) which is

less than Ω(infin) For two 1047298at plates Gel is given by the following

expression

Gel

kT = 2m2 ln

X heth THORN

X infineth THORN

= 2m2Rel heth THORN eth19THORN

where Rel

(h) is a geometric function whose form depends on the

segment density distribution It should be stressed that Gel is always

positive and could play a major role in steric stabilisation It becomes

very strong when the separation distance between the particles

becomes comparable to the adsorbed layer thickness δ

Combination of Gmix and Gel with GA (the van der Waals attractive

energy)gives the total free energy of interaction GT (assuming there is

no contribution from any residual electrostatic interaction) ie

GT = Gmix + Gel + GA eth20THORN

A schematic representation of the variation of Gmix Gel GA and GT

with surfacendashsurface separation distance h is shown in Fig 20

Gmix increases very sharply with decrease of h when hb2δ Gel

increases very sharply with decrease of h when hbδ GT versus h

shows a minimum Gmin at separation distances comparable to 2δ

When h b2δ GT shows a rapid increase with decrease in h

The depth of the minimum depends on the Hamaker constant A

the particle radius R and adsorbed layer thickness δ Gmin increases

with increase of A and R At a given A and R Gmin increases with

decrease in δ (ie with decrease of the molecular weight M w of the

stabiliser This is illustrated in Fig 21 which shows the energyndash

distance curves as a function of δR The larger the value of δR the

smaller the value of Gmin In this case the system may approach

thermodynamic stability as is the case with nano-dispersions

6 Emulsions stabilized by polymeric surfactants

The most effective method for emulsion stabilization is to use

polymeric surfactants that stronglyadsorb at the OWor WO interface

Fig 17 Schematic representation of the interaction between particles containing adsorbed polymer layers

Fig 18 Schematic representation of polymer layer overlap

Fig 19 Schematic representation of con1047297gurational entropy loss on approach of a

second particle

290 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1119

and produce effective steric stabilization against strong 1047298occulation

coalescence and Ostwald ripening [23]

As mentioned above a graft copolymer of the ABn type was

synthesized by grafting several alkyl groups on an inulin (polyfruc-

tose) chain The polymeric surfactant (INUTECregSP1) consists of a

linear polyfructose chain (the stabilizing A chain) and several alkyl

groups (the B chains) that provide multi-anchor attachment to the oil

droplets This polymeric surfactant produces enhanced steric stabili-

zation both in water and high electrolyte concentrations as will be

discussed later

For water-in-oil (WO) emulsions an AndashBndashA block copolymer of

poly (12-hydroxystearic acid) (PHS) (the A chains) and poly (ethylene

oxide) (PEO) (the B chain) PHSndashPEOndashPHS is commercially available

(Arlacel P135 UNIQEMA) The PEO chain (that is soluble in the water

droplets) forms the anchor chain whereas the PHS chains form the

stabilizing chains PHS is highly soluble in most hydrocarbon solvents

and is strongly solvated by its molecules The structure of the PHSndash

PEOndashPHS block copolymer is schematically shown in Fig 22The conformation of the polymeric surfactant at the WO interface

is schematically shown in Fig 23

Emulsions of Isopar Mwater and cyclomethiconewater were

prepared using INUTECregSP1 5050 (vv) OW emulsions were

prepared and the emulsi1047297er concentration was varied from 025 to 2

(wv) based on the oil phase 05 (wv) emulsi1047297erwas suf 1047297cient for

stabilization of these 5050 (vv) emulsions [23]

The emulsions were stored at room temperature and 50 degC and

optical micrographs were taken at intervals of time (for a year) in

order to check the stability Emulsions prepared in water were very

stableshowingno change in droplet size distributionover more than a

year period and this indicated absence of coalescence Any weak

1047298occulation that occurred was reversible and the emulsion could be

redispersed by gentle shaking Fig 24 shows an optical micrograph fora dilute 5050 (vv) emulsion that was stored for 15 and 14 weeks at

50 degC

No change in droplet size was observed after storage for more than

1 year at 50 degC indicating absence of coalescence The same result was

obtained when using different oils Emulsions were also stable against

coalescence in the presence of high electrolyte concentrations (up to

4 mol dmminus3 or ~25 NaCl

The above stability in high electrolyte concentrations is not

observed with polymeric surfactants based on polethylene oxide

The high stability observed using INUTECregSP1 is related to its

strong hydration both in water and in electrolyte solutions The

hydration of inulin (the backbone of HMI) could be assessed using

cloud point measurements A comparison was also made with PEO

with two molecular weights namely 4000 and 20000

Solutionsof PEO 4000 and 20000 showeda systematic decrease of

cloud point with increase in NaCl or MgSO4 concentration In contrast

inulin showed no cloud point up to 4 mol dmminus3 NaCl and up to 1 mol

dmminus3 MgSO4

The above results explain the difference between PEO and inulin

With PEO the chains show dehydration when the NaCl concentration

is increased above 2 mol dmminus3 or 05 mol dmminus3 MgSO4 The inulin

chains remain hydrated at much higher electrolyte concentrations It

seems that the linear polyfructose chains remain strongly hydrated athigh temperature and high electrolyte concentrations

The high emulsion stability obtained whenusingINUTECregSP1 can be

accounted for by the following factors (i) The multi-point attachment

of the polymer by several alkyl chains that are grafted on the backbone

(ii) The strong hydration of the polyfructose ldquoloopsrdquo both in water and

high electrolyte concentrations (χ remains below 05 under these

conditions) (iii) Thehigh volumefraction (concentration) of the loops at

the interface (iv) Enhanced steric stabilization this is the case with

multi-point attachment which produces strong elastic interaction

Evidence for the high stability of the liquid 1047297lm between emulsion

droplets when using INUTECregSP1 was obtained by Exerowa et al [24]

using disjoining pressure measurements This is illustrated in Fig 25

which shows a plot of disjoining pressure versus separation distance

between two emulsion droplets at various electrolyte concentrations

The results show that by increasing the capillary pressure a stable

Newton Black Film (NBF) is obtained at a 1047297lm thickness of sim7 nm

The lack of rupture of the 1047297lm at the highest pressure applied of

45times104 Pa indicate the high stability of the 1047297lm in water and in high

electrolyte concentrations (up to 20 mol dmminus3 NaCl)

The lack of rupture of the NBF up to the highest pressure applied

namely 45times104 Pa clearly indicatesthe high stability of the liquid1047297lm

in the presence of high NaCl concentrations (up to 2 mol dm minus3) This

result is consistent with the high emulsion stability obtained at high

electrolyte concentrations and high temperature Emulsions of Isopar

M-in-water are very stable under such conditions and this could be

accounted for by the high stability of the NBF The droplet size of

5050 OW emulsions prepared using 2 INUTEC regSP1 is in the region

of 1ndash10 μ m This corresponds to a capillary pressure of ~3times104 Pa for

the 1 μ m drops and ~3times103 Pa for the 10 μ m drops These capillarypressures are lower than those to which the NBF have been sub-

jected to and this clearly indicates the high stability obtained against

coalescence in these emulsions

61 WO emulsions stabilized with PHS ndashPEOndashPHS block copolymer

WO emulsions (the oil being Isopar M) were prepared using PHSndash

PEOndashPHS block copolymer at high water volume fractions (N07) The

emulsions have a narrow droplet size distribution with a z -average

radius of 183 nm [25] They also remained 1047298uid up to high water

volume fractions (N06) This could be illustrated from viscosityndash

volume fraction curves as is shown in Fig 26

Fig 20 Energyndashdistance curves for sterically stabilized systems

Fig 21 Variation of Gmin with δR

291T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1219

The effective volume fraction ϕeff of the emulsions (the core

droplets plus the adsorbed layer) could be calculated from the relative

viscosity and using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth21THORN

Where η r is the relative viscosity ϕp is the maximum packing fraction(sim07) and [ η ] is the intrinsic viscosity that is equal to 25 for hard-

spheres

The calculations based on Eq (22) are shown in Fig 27 (square

symbols) From the effective volume fraction ϕeff and the core volume

fraction ϕ the adsorbed layer thickness could be calculated This was

found to be in the region of 10 nm at ϕ =04 and it decreased with

increase in ϕ

The WO emulsions prepared using the PHSndashPEOndashPHS block

copolymer remained stable both at room temperature and 50 degC

This is consistent with the structure of the block copolymer the B

chain (PEO) is soluble in water and it forms a very strong anchor at the

WO interface The PHS chains (the A chains) provide effective steric

stabilization since the chains are highly soluble in Isopar M and are

strongly solvated by its molecules

7 Suspensions stabilised using polymeric surfactants

There are generally two procedures for preparation of solidliquid

dispersions

(i) Condensation methods build-up of particles from mole-

cular units ie nucleation and growth A special procedure

is the preparation of latexes by emulsion or dispersion

polymerization

(ii) Dispersion methods in this case one starts with preformed

large particles or crystals which are dispersed in the liquid by

using a surfactant (wetting agent) with subsequent breaking up

of the large particles by milling (comminution) to achieve the

desirable particle size distribution A dispersing agent (usuallya

polymeric surfactant) is used for the dispersion process and

subsequent stabilization of the resulting suspension

There are generally two procedures for preparation of latexes

(i) Emulsion polymerization the monomers that are essentially

insoluble in the aqueous medium are emulsi1047297ed using a

surfactant and an initiator is added while heating the system

to produce the polymer particles that are stabilized electro-

statically (when using ionic surfactants) or sterically (when

using non-ionic surfactants)

(ii) Dispersion polymerization the monomers are dissolved in a

solvent in which the resulting polymer particlesare insoluble A

protective colloid (normally a block or graft copolymer) is

added to prevent 1047298occulation of the resulting polymers

particles that are produced on addition of an initiator This

method is usually applied for the production of non-aqueous

latex dispersions and is sometimes referred to as Non-Aqueous

Dispersion Polymerization (NAD)

Surfactants play a crucial role in the process of latex preparation

since they determine the stabilizing ef 1047297ciency and the effectiveness of

the surface active agent ultimately determines the number of particles

Fig 22 Schematic representation of the structure of PHSndashPEOndashPHS block copolymer

Fig 23 Conformation of PHSndash

PEOndash

PHS polymeric surfactant at the WO interface

292 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1319

and their size The effectiveness of any surface active agent in

stabilizing the particles is the dominant factor and the number of

micelles formed is relatively unimportant In the NAD process themonomer normally an acrylic is dissolved in a non-aqueous solvent

normally an aliphatic hydrocarbon and an oil soluble initiator and a

stabilizer (to protect the resulting particles from1047298occulation) is added

to the reaction mixture The most successful stabilizers used in NAD

are block and graft copolymers Preformed graft stabilizers based

on poly(12-hydroxy stearic acid) (PHS) are simple to prepare and

effective in NAD polymerization

Dispersion methods are used for the preparation of suspensions of

preformed particles The role of surfactants (or polymers) in the

dispersion process can be analyzed in terms of the three processesinvolved (i) Wetting of the powder by the liquid (ii) Breaking of the

aggregates and agglomerates (iii) Comminution of the resulting

particles and their subsequent stabilization All these processes are

affected by surfactants or polymers which adsorb on the powder

surface thus aiding the wetting of the powder break-up of the

aggregates and agglomerates and subsequent reduction of particle

size by wet milling

71 Polymeric surfactants in emulsion polymerization

Recently the graft copolymer of hydrophobically modi1047297ed inulin

(INUTECreg SP1) has been used in emulsion polymerization of styrene

methyl methacrylate butyl acrylate and several other monomers [26]All lattices were prepared by emulsion polymerisation using potas-

sium persulphate as initiator The z -average particle size was

determined by photon correlation spectroscopy (PCS) and electron

micrographs were also taken

Fig 24 Optical micrographs of OW emulsions stabilized with INUTEC regSP1 stored at

50 degC for 15 weeks (a) and 14 weeks (b)

Fig 25 Variation of disjoining pressure with equivalent 1047297lm thickness at various NaCl concentrations

Fig 26 Viscosityndashvolume fraction for WO emulsion stabilized with PHSndashPEOndashPHS

block copolymer experimental data calculated using Eq (22)

293T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 319

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 419

Clearly when χb12 B2 is positive and mixing is non-ideal leading to

positive deviation (repulsion) this occurs when the polymer chains

are in ldquogoodrdquo solvent conditions In contrast when χN12 B2 is

negative and mixing is non-ideal leading to negative deviation

(attraction) this occurs when the polymer chains are in ldquopoorrdquo

solvent conditions (precipitation of the polymer may occur under

these conditions) Since the polymer solvency depends on tempera-

ture one can also de1047297ne a theta temperature θ at which χ=12

The function [(12)minus

χ] can also be expressed in terms of twomixing parameters an enthalpy parameter κ 1 and an entropy

parameter ψ1 ie

1

2minusχ

= κ 1minusψ1 eth8THORN

The θ-temperature can also be de1047297ned in terms of κ 1 and ψ1

θ = κ 1T

ψ1

eth9THORN

Alternatively one can write

1

2

minusχ = ψ1 1minusθ

T eth10THORN

Although the FloryndashHuggins theory is sound in principle several

experimental results cannot be accounted for For example it was

found that the χ parameter depends on the polymer concentration in

solution Most serious is the fact that many polymer solutions (such as

PEO) show phase separation on heating when the theory predictsthat

it should happen only on cooling

The solution properties of copolymers are much more complicated

This is due to the fact that the two copolymer components A and B

behave differently in different solvents Only when the two compo-

nents are both soluble in the same solvent then they exhibit similar

solution properties This is the case for example for a non-polar

copolymer in a non-polar solvent It should also be emphasised that

the FloryndashHuggins theory was developed for ideal linear polymers

Indeed with branched polymers consisting of high monomer density

(eg star branched polymers) the θ-temperature depends on the

lengthof the armsand isin general lower thanthatof a linearpolymer

with the same molecular weight

Another complication arises from speci1047297c interaction with the

solvent eg hydrogen bonding between polymer and solvent

molecules (eg with PEO and PVA in water) Also aggregation insolution (lack of complete dissolution) may present another problem

One of the most useful parameters for characterising the

conformation of a polymer in solution is the root mean-square (rms)

end to end length br 2N12 which represents a con1047297guration character r

as the distance from one end group to the other of a chain molecule

Another useful parameter is the radius of gyration bs2N

12 which is a

measure of the effective size of a polymer molecule (it is the root

mean-square distance of the elements of the chain from its centre of

gravity)

For linear polymers

bs2N

1=2 = br 2N1=2

61=2 eth11THORN

The radius of gyration of a polymer in solution can be determined

from light scattering measurements

As mentioned above dilute solutions of copolymers is solvents that

are good for both components exhibit similar behaviour to hompo-

lymer chains However in a selective solvent whereby the medium is

a good solvent for one component say A and a poor solvent for the

second component B the very different solvent af 1047297nities to the two

components will have a large effect on the conformation of the

isolated chain This results in formation of aggregates involving

several macromolecules in dilute solutions (low Critical Aggregation

Concentration CAC) It is believed that the polymeric aggregates are

Fig 3 Various conformations of macromolecules on a plane surface

284 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 519

spherical [5] The CAC of these block and graft copolymers is usually

very low

Several methods may be applied to obtain the aggregate size and

shape of block and graft copolymers of which light scattering small

angle X-ray and neutron scattering are probably the most direct

Dynamic light scattering (photon correlation spectroscopy) can also

be applied to obtain the hydrodynamic radius of the aggregate This

technique is relatively easy to perform when compared with static

light scattering since it does not require special preparation of thesample

13 Adsorption and conformation of polymeric surfactants at interfaces

Understanding the adsorption and conformation of polymeric

surfactants at interfaces is key to knowing how these molecules act as

stabilizers Most basic ideas on adsorption and conformation of

polymers have been developed for the solidliquid interface [10ndash17]

The process of polymer adsorption is fairly complicated mdash it

involves PolymerSurface Interaction PolymerSolvent Interaction

SurfaceSolvent Interaction The con1047297guration (conformation) of the

polymer at the solidliquid interface The polymersurface interaction

is described in terms of adsorption energy per segment χs The

polymersolvent interaction is described in terms of the Floryndash

Huggins interaction parameter χ The polymer con1047297guration is

described by sequences of Trains segments in direct contact with

the surface Loops segments in between the trains that extend into

solution Tails ends of the molecules that also extend into solution A

schematic representation of the various polymer con1047297gurations is

given in Fig 3

For homopolymers eg poly(ethylene oxide) (PEO) or polyvinyl

pyrrolidone (PVP) a train-loop-tail con1047297guration is the case For

adsorption to occur a minimum energy of adsorption per segment χs

is required When a polymer molecule adsorbs on a surface it looses

con1047297gurational entropy and this must be compensated by an

adsorption energy χs per segment This is schematically shown in

Fig 4 where the adsorbed amount Γ is plotted versus χs The

minimum value of χs can be very small (b01kT ) since a large number

of segments per molecule are adsorbed For a polymer with say 100segments and 10 of these are in trains the adsorption energy per

molecule now reaches 1kT (with χs =01kT ) For 1000 segments the

adsorption energy per molecule is now 10kT

Homopolymers are not the most suitable for stabilisation of

dispersions For strong adsorption one needs the molecule to be

ldquoinsolublerdquo in the medium and has strong af 1047297nity (ldquoanchoringrdquo) to the

surface For stabilisation one needs the molecule to be highly soluble

in themedium andstronglysolvated by itsmolecules mdash This requires a

FloryndashHuggins interaction parameter less than 05 The above

opposing effects can be resolved by introducing ldquoshortrdquo blocks in the

moleculewhichare insolublein themedium andhavea strong af 1047297nity

to the surface as illustrated in Fig 3b Example Partially hydrolysed

polyvinyl acetate (88 hydrolysed ie with 12 acetate groups)

usually referred to as polyvinyl alcohol (PVA)

The above requirements are better satis1047297ed using AndashB AndashBndashA and

BAn graft copolymers B is chosento be highly insolublein themedium

and it should have high af 1047297nity to the surface This is essential to

ensure strong ldquoanchoringrdquo to the surface (irreversible adsorption) A is

chosen to be highly soluble in the medium and strongly solvated by

its molecules The FloryndashHuggins χ parameter can be applied in this

case For a polymer in a good solvent χ has to be lower than 05 the

smaller the χ value the better the solvent for the polymer chains

Examples of B for a hydrophobic particles in aqueous media are

polystyrene polymethylmethacrylate Examples of A in aqueous

media are polyethylene oxide polyacrylic acid polyvinyl pyrollidone

and polysaccharides For non-aqueous media such as hydrocarbons

the A chain(s) could be poly(12-hydroxystearic acid)

For full description of polymer adsorption one needs to obtain

information on the following (i) The amount of polymer adsorbed Γ (in mg or moles) per unit area of the particles It is essential to know

thesurface area of theparticles in thesuspensionNitrogenadsorption

on the powder surface may give such information (by application of

the BET equation) provided there will be no change in area on

dispersing the particles in the medium For many practical systems a

change in surface area may occur on dispersing the powder in which

case one has to use dye adsorption to measure the surface area (some

assumptions have to be made in this case) (ii)The fraction of segments

in direct contact with thesurface ie the fraction of segmentsin trains

p ( p = (Number of segments in direct contact with the surface) Total

Number) (iii)The distribution of segments in loops and tails ρ(z)

which extend in several layersfrom the surface ρ(z) is usually dif 1047297cult

to obtain experimentally although recently application of small angle

neutron scattering could obtain such information An alternative anduseful parameter for assessing ldquosteric stabilisationrdquo is the hydro-

dynamic thickness δh (thickness of the adsorbed or grafted polymer

layer plus any contribution from the hydration layer) Several methods

can be applied to measure δh as will be discussed below

2 Theories of polymer adsorption

Two main approaches have been developed to treat the problem of

polymer adsorption (i) Random Walk approach This is based on

Florys treatment of the polymer chain in solution the surface was

considered as a re1047298ecting barrier (ii) Statistical mechanical approach

Thepolymer con1047297gurationwas treated as being made of three types of

structures trains loops and tails each having a different energy state

The random walk approach is an unrealistic model to the problemof polymer adsorption since the polymer interacts in a speci1047297c

manner with the surface and the solvent The statistical mechanical

approach is a more realistic model for the problem of polymer

adsorption since it takes into account the various interactions

involved A useful model for treating polymer adsorption and

con1047297guration was suggested by Scheutjens and Fleer (SF theory)

[14ndash16] that is referred to as the step weighted random walk approach

which is summarized below

The SF theory starts without any assumption for the segment

density distribution The partition functions were derived for the

mixture of free and adsorbed polymer molecules as well as for the

solvent molecules All chain conformations were described as step

weighted random walks on a quasi-crystalline lattice which extends in

parallel layers from the surface mdash

this is schematically shown in Fig 5Fig 4 Variation of adsorption amount Γ with adsorption energy per segment χs

285T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 619

The partition function is written in terms of a number of con1047297gura-

tions These were treated as connectedsequencesof segments In each

layer random mixing between segments and solvent molecules was

assumed (mean 1047297eld approximation) Each step in the random walk

was assigned a weighting factor pi

pi consists of three contributions (i) An adsorption energyχs (which

exists only for the segments that are near the surface) (ii) Con1047297gura-

tional entropy of mixing (that exists in each layer) (iii) Segment-solvent

interaction parameterχ (the FloryndashHuggins interaction parameter note

that χ=0 for an athermal solvent χ=05 for a θ-solvent)

From the weighting factors the statistical weight of any chain

conformation can be calculated for a given segment concentration

pro1047297le using a matrix formulation Fig 6 shows typical adsorption

isotherms plotted as surface coverage (in equivalent monolayers)

versus polymer volume fraction ϕ in bulk solution (ϕ was taken to

vary between 0 and 10minus3 which is the normal experimental range)

The results in Fig 6 show the effect of increasing the chain length r

and effect of solvency (athermal solvent with χ=0 and theta solvent

with χ=05) The adsorption energy χs was taken to be the same and

equal to 1kT When r = 1 θ is very small and the adsorption increases linearly

with increase of ϕ (Henrys type isotherm) On the other hand when

r = 10 the isotherm deviates much from a straight line and approaches

a Langmuirian type However when r ge20 high af 1047297nity isotherms are

obtained This implies that the 1047297rst added polymer chains are

completely adsorbed resulting in extremely low polymer concentra-

tion in solution (approaching zero) This explains the irreversibility of

adsorption of polymeric surfactants with r N100 The adsorption

isotherms with r =100 and above are typical of those observed

experimentally for most polymers that are not too polydisperse ie

showing a steep rise followed by a nearly horizontal plateau (which

only increases few percent per decade increase of ϕ) In these dilute

solutions the effect of solvency is most clearly seen with poor

solvents giving the highest adsorbed amounts In good solvents θ is

much smaller and levels off for long chains to attain an adsorption

plateau which is essentially independent of molecular weight

Another point that emerges from the SF theory is the difference in

shape between the experimental and theoretical adsorption isotherms

in thelow concentration region Theexperimentalisotherms areusuallyrounded whereas those predicted fromtheory are1047298at This is accounted

for in terms of the molecular weight distribution (polydispersity) which

is encountered with most practical polymers This effect has been

explained by CohenndashStuart et al [17] With polydisperse polymers the

larger molecular weight fractions adsorb preferentially over the smaller

ones At low polymer concentrations nearly all polymer fractions are

adsorbed leaving a small fraction of the polymer with the lowest

molecular weights in solution As the polymer concentration is

increased the higher molecular weight fractions displace the lower

ones on thesurfacewhichare nowreleasedin solution thusshiftingthe

molecular weight distribution in solution to lower values This process

continues with further increase in polymer concentration leading to

fractionation whereby the higher molecular weight fractions are

adsorbed at the expense of the lower fractions which are released to

the bulk solution However in very concentrated solutions monomers

adsorb preferentially with respect to polymers and short chains with

respect to larger ones This is due to the fact that in this region the

conformational entropy term predominates the free energy disfavour-

ing the adsorption of long chains

According to the SF theory the bound fraction p is high at low

concentration and relatively independent of molecular weight when

r N20 However with increase in surface coverage andor molecular

weight p tends to decrease indicating the formation of larger loops

and tails

The structure of the adsorbed layer is described in terms of the

segment density distribution As an illustration Fig 7 shows some

calculations using the SF theory for loops and tails with r =1000

ϕ= 10minus6 and χ=05 In this example 38 of the segments are in

trains 555 in loops and 65 in tails This theory demonstrates theimportance of tails which dominate the total distribution in the outer

region

3 Experimental techniques for studying polymeric surfactant

adsorption

As mentioned above for full characterization of polymeric sur-

factant adsorption one needs to determine three parameters (i) The

adsorbed amount Γ (mg mminus2 or mol mminus2) as a function of equilibrium

concentration C eq ie the adsorption isotherm (ii) The fraction of

segments in direct contact with the surface p (number of segments in

Fig 5 Schematic representation of a polymer molecule adsorbing on a 1047298at surface mdash

quasi-crystalline lattice with segments 1047297lling layers that are parallel to the surface

(random mixing of segments and solvent molecules in each layer is assumed)

Fig 6 Adsorption isotherms for oligomers and polymers in the dilute region based on

the SF theory Fig 7 Loop tail and total segment pro1047297le according to the SF theory

286 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 719

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 819

The microelectrophoresis technique is based on measurement of

the electrophoretic mobility u of the particles in the presence and

absence of the polymer layer From u one can calculate the zeta

potential ζ using the Huckel equation (which is applicable for small

particles and extended double layers ie κ Rbb1 where κ is the

DebyendashHuckel parameter that is related to the salt concentration

u = 23eeon η

eth15THORN

where ε is the relative permittivity of the medium and ε o is the

permittivity of free space

By measuring ζ of the particles with and without the adsorbed

polymer layer one can obtain the hydrodynamic thickness δh For

accurate measurements one should carry the measurements at various

electrolyte concentrations and extrapolate the results to the plateau

value Several automatic instruments are available for measurement of

the electrophoretic mobility Malvern zeta sizerndashCoulter Delsa sizer and

Broekahven instrument All these instruments are easy to use and the

measurement can be carried out within few minutes

4 Examples of the adsorption isotherms of nonionic polymericsurfactants

Fig 9 shows the adsorption isotherms for PEO with different

molecular weights on PS (at room temperature It can be seen that the

amount adsorbed in mgm-2 increases with increase in the polymer

molecular weight Fig 10 shows the variation of the hydrodynamic

thickness δh with molecular weight M δh shows a linear increase with

log M δh increases with n the number of segments in the chain

according to

δhasympn08 eth16THORN

Fig 11 shows the adsorption isotherms of PVA with various

molecular weights on PS latex (at 25 degC) [19] The polymers were

obtained by fractionation of a commercial sample of PVA with an

average molecular weight of 45000 The polymer also contained 12

vinyl acetate groups As with PEO the amount of adsorption increases

with increase in M The isotherms are also of the high af 1047297nity type Γ at the plateau increases linearly with M 12

The hydrodynamic thickness was determined using PCS and theresults are given below

M 67000 43000 28000 17000 8000

δhnm 255 197 140 98 33

δh seems to increase linearly with increase in the molecular weight

The effect of solvency on adsorption was investigated by increasing

the temperature (the PVA molecules are less soluble at higher

temperature) or addition of electrolyte (KCl) [20] The results are

shown in Figs 12 and 13 for M =65100 As can be seen from Fig 12

increase of temperature results in reduction of solvency of the

mediumfor the chain (due to break down of hydrogen bonds) and this

results in an increase in the amount adsorbed Addition of KCl (which

reduces the solvency of the medium for the chain) results in anincrease in adsorption (as predicted by theory)

The adsorption of block and graft copolymers is more complex

since the intimate structure of the chain determines the extent of

adsorption [18] Randomcopolymers adsorbin an intermediate way to

that of the corresponding homopolymers Blockcopolymers retain the

adsorption preference of the individual blocks The hydrophilic block

(eg PEO) the buoy (previously referred to as the A chain) extends

away from the particle surface into the bulk solution whereas the

hydrophobic anchor block (previously referred to as the B chain) (eg

PS or PPO) provides 1047297rm attachment to the surface Fig 14 shows the

theoretical prediction of diblock copolymer adsorption according to

Fig 9 Adsorption isotherms for PEO on PS

Fig 10 Hydrodynamic thickness of PEO on PS as a function of the molecular weight

Fig 11 Adsorption isotherms of PVA with different molecular weights on polystyrene

latex at 25 degC

Fig 12 In1047298uence of temperature on adsorption

288 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 919

the Scheutjens and Fleer theory The surface density σ is plotted

versus the fraction of anchor segmentsν A The adsorption depends on

the anchorbuoy composition

The amount of adsorption is higher than for homopolymers and

the adsorbed layer thickness is more extended and dense Fig 15

shows the theoretical prediction for the adsorbed layer thickness δwhich is plotted as a function of ν A For a triblock copolymer AndashBndashA

with two buoy chains (A) and one anchor chain (B) the behaviour is

similar to that of diblock copolymers This is shown in Fig16 for PEOndash

PPOndashPEO block (Pluronic)

5 Interaction between particles containing adsorbed polymeric

surfactant layers steric stabilization

When two particles each with a radius R and containing an

adsorbed polymer layer with a hydrodynamic thickness δh approach

each other to a surfacendashsurface separation distance h that is smaller

than 2 δh the polymer layers interact with each other resulting in two

main situations [21] (i) The polymer chains may overlap with each

other (ii) The polymer layer may undergo some compression In both

cases there will be an increase in the local segment density of the

polymer chains in the interaction region This is schematicallyillustrated in Fig 17 The real situation is perhaps in between the

above two cases ie the polymer chains may undergo some

interpenetration and some compression

Provided the dangling chains (the A chains in AndashB AndashBndashA block or

BAn graft copolymers) are in a good solvent this local increase in

segment density in the interaction zone will result in strong repulsion

as a result of two main effects (i) Increase in the osmotic pressure in

the overlap region as a result of the unfavourable mixing of the

polymer chains when these are in good solvent conditions This is

referred to as osmotic repulsion or mixing interaction and it is

described by a free energy of interaction Gmix (ii) Reduction of the

con1047297gurational entropy of the chains in the interaction zone this

entropy reduction results from the decrease in the volume available

for thechains when these areeither overlapped or compressed This is

referred to as volume restriction interaction entropic or elastic

interaction and it is described by a free energy of interaction Gel

Combination of Gmix and Gel is usually referred to as the steric

interaction free energy Gs ie

Gs = Gmix + Gel eth17THORN

The sign of Gmix depends on the solvency of the medium for the

chains If in a good solvent ie the FloryndashHuggins interaction

parameter χ is less than 05 then Gmix is positive and the mixing

interaction leads to repulsion (see below) In contrast if χN05 (ie the

chains are in a poor solvent condition) Gmix is negative and the

mixing interaction becomes attractive Gel is always positive and

hence in some cases one can produce stable dispersions in a relatively

poor solvent (enhanced steric stabilisation)

51 Mixing interaction Gmix

This results from the unfavourable mixing of the polymer chainswhen these are in a good solvent conditions This is schematically

shown in Fig 18

Consider two spherical particles with the same radius and each

containing an adsorbed polymer layer with thickness δ Before

overlap one can de1047297ne in each polymer layer a chemical potential

for the solvent μ iα and a volume fraction for the polymer in the

layerϕ2 In the overlap region (volume element dV ) the chemical

potential of the solvent is reduced to μ i β This results from the increase

in polymer segment concentration in this overlap region

In the overlap region the chemical potential of the polymer chains

is now higher than in the rest of the layer (with no overlap) This

Fig 13 In1047298uence of addition of KCl on adsorption

Fig 14 Prediction of Adsorption of diblock copolymer

Fig15 Theoretical predictions of the adsorbed layer thickness for a diblock copolymer

Fig 16 Adsorbed amount (mg mminus2) versus fraction of anchor segment for an AndashBndashA

triblock copolymer (PEOndash

PPOndash

PEO)

289T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1019

amounts to an increase in the osmotic pressure in the overlap region

as a result solvent will diffuse from thebulk to the overlap region thus

separating the particles and hence a strong repulsive energy arises

from this effect The above repulsive energy can be calculated by

considering the free energy of mixing of two polymer solutions as for

example treated by Flory and Krigbaum [22] The free energy of

mixing is given by two terms (i) An entropy term that depends on the

volume fraction of polymer and solvent (ii)An energy term that is

determined by the FloryndashHuggins interaction parameter χ

Using the above theory one can derive an expression for the free

energy of mixing of two polymer layers (assuming a uniform segment

density distribution in each layer) surrounding two spherical particles

as a function of the separation distance h between the particles The

expression for Gmix is

Gmix

kT =

2V 22V 1

m2

2

1

2minusχ

δminus

h

2

2

3R + 2δ + h

2

eth18THORN

k is the Boltzmann constant T is the absolute temperature V 2 is the

molar volume of polymer V 1 is the molar volume of solvent and ν 2 is

the number of polymer chains per unit area

The sign of Gmix depends on the value of the FloryndashHuggins

interaction parameter χ if χb05 Gmix is positive and the interaction

is repulsive if χN05 Gmix is negative and the interaction is attractiveif χ=05 Gmix =0 and this de1047297nes the θ-condition

52 Elastic interaction Gel

Thisarises fromthe lossin con1047297gurational entropy of thechains on

the approach of a second particle As a result of this approach the

volume available for the chains becomes restricted resulting in loss of

the number of con1047297gurations This can be illustrated by considering a

simple molecule represented by a rod that rotates freely in a

hemisphere across a surface (Fig 19) When the two surfaces are

separated byan in1047297nitedistanceinfin thenumber of con1047297gurations of the

rod is Ω(infin) which is proportional to the volume of the hemisphere

When a secondparticleapproaches to a distance h suchthatit cuts the

hemisphere (loosing some volume) the volume available to the chains

is reduced and the number of con1047297gurations become Ω(h) which is

less than Ω(infin) For two 1047298at plates Gel is given by the following

expression

Gel

kT = 2m2 ln

X heth THORN

X infineth THORN

= 2m2Rel heth THORN eth19THORN

where Rel

(h) is a geometric function whose form depends on the

segment density distribution It should be stressed that Gel is always

positive and could play a major role in steric stabilisation It becomes

very strong when the separation distance between the particles

becomes comparable to the adsorbed layer thickness δ

Combination of Gmix and Gel with GA (the van der Waals attractive

energy)gives the total free energy of interaction GT (assuming there is

no contribution from any residual electrostatic interaction) ie

GT = Gmix + Gel + GA eth20THORN

A schematic representation of the variation of Gmix Gel GA and GT

with surfacendashsurface separation distance h is shown in Fig 20

Gmix increases very sharply with decrease of h when hb2δ Gel

increases very sharply with decrease of h when hbδ GT versus h

shows a minimum Gmin at separation distances comparable to 2δ

When h b2δ GT shows a rapid increase with decrease in h

The depth of the minimum depends on the Hamaker constant A

the particle radius R and adsorbed layer thickness δ Gmin increases

with increase of A and R At a given A and R Gmin increases with

decrease in δ (ie with decrease of the molecular weight M w of the

stabiliser This is illustrated in Fig 21 which shows the energyndash

distance curves as a function of δR The larger the value of δR the

smaller the value of Gmin In this case the system may approach

thermodynamic stability as is the case with nano-dispersions

6 Emulsions stabilized by polymeric surfactants

The most effective method for emulsion stabilization is to use

polymeric surfactants that stronglyadsorb at the OWor WO interface

Fig 17 Schematic representation of the interaction between particles containing adsorbed polymer layers

Fig 18 Schematic representation of polymer layer overlap

Fig 19 Schematic representation of con1047297gurational entropy loss on approach of a

second particle

290 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1119

and produce effective steric stabilization against strong 1047298occulation

coalescence and Ostwald ripening [23]

As mentioned above a graft copolymer of the ABn type was

synthesized by grafting several alkyl groups on an inulin (polyfruc-

tose) chain The polymeric surfactant (INUTECregSP1) consists of a

linear polyfructose chain (the stabilizing A chain) and several alkyl

groups (the B chains) that provide multi-anchor attachment to the oil

droplets This polymeric surfactant produces enhanced steric stabili-

zation both in water and high electrolyte concentrations as will be

discussed later

For water-in-oil (WO) emulsions an AndashBndashA block copolymer of

poly (12-hydroxystearic acid) (PHS) (the A chains) and poly (ethylene

oxide) (PEO) (the B chain) PHSndashPEOndashPHS is commercially available

(Arlacel P135 UNIQEMA) The PEO chain (that is soluble in the water

droplets) forms the anchor chain whereas the PHS chains form the

stabilizing chains PHS is highly soluble in most hydrocarbon solvents

and is strongly solvated by its molecules The structure of the PHSndash

PEOndashPHS block copolymer is schematically shown in Fig 22The conformation of the polymeric surfactant at the WO interface

is schematically shown in Fig 23

Emulsions of Isopar Mwater and cyclomethiconewater were

prepared using INUTECregSP1 5050 (vv) OW emulsions were

prepared and the emulsi1047297er concentration was varied from 025 to 2

(wv) based on the oil phase 05 (wv) emulsi1047297erwas suf 1047297cient for

stabilization of these 5050 (vv) emulsions [23]

The emulsions were stored at room temperature and 50 degC and

optical micrographs were taken at intervals of time (for a year) in

order to check the stability Emulsions prepared in water were very

stableshowingno change in droplet size distributionover more than a

year period and this indicated absence of coalescence Any weak

1047298occulation that occurred was reversible and the emulsion could be

redispersed by gentle shaking Fig 24 shows an optical micrograph fora dilute 5050 (vv) emulsion that was stored for 15 and 14 weeks at

50 degC

No change in droplet size was observed after storage for more than

1 year at 50 degC indicating absence of coalescence The same result was

obtained when using different oils Emulsions were also stable against

coalescence in the presence of high electrolyte concentrations (up to

4 mol dmminus3 or ~25 NaCl

The above stability in high electrolyte concentrations is not

observed with polymeric surfactants based on polethylene oxide

The high stability observed using INUTECregSP1 is related to its

strong hydration both in water and in electrolyte solutions The

hydration of inulin (the backbone of HMI) could be assessed using

cloud point measurements A comparison was also made with PEO

with two molecular weights namely 4000 and 20000

Solutionsof PEO 4000 and 20000 showeda systematic decrease of

cloud point with increase in NaCl or MgSO4 concentration In contrast

inulin showed no cloud point up to 4 mol dmminus3 NaCl and up to 1 mol

dmminus3 MgSO4

The above results explain the difference between PEO and inulin

With PEO the chains show dehydration when the NaCl concentration

is increased above 2 mol dmminus3 or 05 mol dmminus3 MgSO4 The inulin

chains remain hydrated at much higher electrolyte concentrations It

seems that the linear polyfructose chains remain strongly hydrated athigh temperature and high electrolyte concentrations

The high emulsion stability obtained whenusingINUTECregSP1 can be

accounted for by the following factors (i) The multi-point attachment

of the polymer by several alkyl chains that are grafted on the backbone

(ii) The strong hydration of the polyfructose ldquoloopsrdquo both in water and

high electrolyte concentrations (χ remains below 05 under these

conditions) (iii) Thehigh volumefraction (concentration) of the loops at

the interface (iv) Enhanced steric stabilization this is the case with

multi-point attachment which produces strong elastic interaction

Evidence for the high stability of the liquid 1047297lm between emulsion

droplets when using INUTECregSP1 was obtained by Exerowa et al [24]

using disjoining pressure measurements This is illustrated in Fig 25

which shows a plot of disjoining pressure versus separation distance

between two emulsion droplets at various electrolyte concentrations

The results show that by increasing the capillary pressure a stable

Newton Black Film (NBF) is obtained at a 1047297lm thickness of sim7 nm

The lack of rupture of the 1047297lm at the highest pressure applied of

45times104 Pa indicate the high stability of the 1047297lm in water and in high

electrolyte concentrations (up to 20 mol dmminus3 NaCl)

The lack of rupture of the NBF up to the highest pressure applied

namely 45times104 Pa clearly indicatesthe high stability of the liquid1047297lm

in the presence of high NaCl concentrations (up to 2 mol dm minus3) This

result is consistent with the high emulsion stability obtained at high

electrolyte concentrations and high temperature Emulsions of Isopar

M-in-water are very stable under such conditions and this could be

accounted for by the high stability of the NBF The droplet size of

5050 OW emulsions prepared using 2 INUTEC regSP1 is in the region

of 1ndash10 μ m This corresponds to a capillary pressure of ~3times104 Pa for

the 1 μ m drops and ~3times103 Pa for the 10 μ m drops These capillarypressures are lower than those to which the NBF have been sub-

jected to and this clearly indicates the high stability obtained against

coalescence in these emulsions

61 WO emulsions stabilized with PHS ndashPEOndashPHS block copolymer

WO emulsions (the oil being Isopar M) were prepared using PHSndash

PEOndashPHS block copolymer at high water volume fractions (N07) The

emulsions have a narrow droplet size distribution with a z -average

radius of 183 nm [25] They also remained 1047298uid up to high water

volume fractions (N06) This could be illustrated from viscosityndash

volume fraction curves as is shown in Fig 26

Fig 20 Energyndashdistance curves for sterically stabilized systems

Fig 21 Variation of Gmin with δR

291T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1219

The effective volume fraction ϕeff of the emulsions (the core

droplets plus the adsorbed layer) could be calculated from the relative

viscosity and using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth21THORN

Where η r is the relative viscosity ϕp is the maximum packing fraction(sim07) and [ η ] is the intrinsic viscosity that is equal to 25 for hard-

spheres

The calculations based on Eq (22) are shown in Fig 27 (square

symbols) From the effective volume fraction ϕeff and the core volume

fraction ϕ the adsorbed layer thickness could be calculated This was

found to be in the region of 10 nm at ϕ =04 and it decreased with

increase in ϕ

The WO emulsions prepared using the PHSndashPEOndashPHS block

copolymer remained stable both at room temperature and 50 degC

This is consistent with the structure of the block copolymer the B

chain (PEO) is soluble in water and it forms a very strong anchor at the

WO interface The PHS chains (the A chains) provide effective steric

stabilization since the chains are highly soluble in Isopar M and are

strongly solvated by its molecules

7 Suspensions stabilised using polymeric surfactants

There are generally two procedures for preparation of solidliquid

dispersions

(i) Condensation methods build-up of particles from mole-

cular units ie nucleation and growth A special procedure

is the preparation of latexes by emulsion or dispersion

polymerization

(ii) Dispersion methods in this case one starts with preformed

large particles or crystals which are dispersed in the liquid by

using a surfactant (wetting agent) with subsequent breaking up

of the large particles by milling (comminution) to achieve the

desirable particle size distribution A dispersing agent (usuallya

polymeric surfactant) is used for the dispersion process and

subsequent stabilization of the resulting suspension

There are generally two procedures for preparation of latexes

(i) Emulsion polymerization the monomers that are essentially

insoluble in the aqueous medium are emulsi1047297ed using a

surfactant and an initiator is added while heating the system

to produce the polymer particles that are stabilized electro-

statically (when using ionic surfactants) or sterically (when

using non-ionic surfactants)

(ii) Dispersion polymerization the monomers are dissolved in a

solvent in which the resulting polymer particlesare insoluble A

protective colloid (normally a block or graft copolymer) is

added to prevent 1047298occulation of the resulting polymers

particles that are produced on addition of an initiator This

method is usually applied for the production of non-aqueous

latex dispersions and is sometimes referred to as Non-Aqueous

Dispersion Polymerization (NAD)

Surfactants play a crucial role in the process of latex preparation

since they determine the stabilizing ef 1047297ciency and the effectiveness of

the surface active agent ultimately determines the number of particles

Fig 22 Schematic representation of the structure of PHSndashPEOndashPHS block copolymer

Fig 23 Conformation of PHSndash

PEOndash

PHS polymeric surfactant at the WO interface

292 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1319

and their size The effectiveness of any surface active agent in

stabilizing the particles is the dominant factor and the number of

micelles formed is relatively unimportant In the NAD process themonomer normally an acrylic is dissolved in a non-aqueous solvent

normally an aliphatic hydrocarbon and an oil soluble initiator and a

stabilizer (to protect the resulting particles from1047298occulation) is added

to the reaction mixture The most successful stabilizers used in NAD

are block and graft copolymers Preformed graft stabilizers based

on poly(12-hydroxy stearic acid) (PHS) are simple to prepare and

effective in NAD polymerization

Dispersion methods are used for the preparation of suspensions of

preformed particles The role of surfactants (or polymers) in the

dispersion process can be analyzed in terms of the three processesinvolved (i) Wetting of the powder by the liquid (ii) Breaking of the

aggregates and agglomerates (iii) Comminution of the resulting

particles and their subsequent stabilization All these processes are

affected by surfactants or polymers which adsorb on the powder

surface thus aiding the wetting of the powder break-up of the

aggregates and agglomerates and subsequent reduction of particle

size by wet milling

71 Polymeric surfactants in emulsion polymerization

Recently the graft copolymer of hydrophobically modi1047297ed inulin

(INUTECreg SP1) has been used in emulsion polymerization of styrene

methyl methacrylate butyl acrylate and several other monomers [26]All lattices were prepared by emulsion polymerisation using potas-

sium persulphate as initiator The z -average particle size was

determined by photon correlation spectroscopy (PCS) and electron

micrographs were also taken

Fig 24 Optical micrographs of OW emulsions stabilized with INUTEC regSP1 stored at

50 degC for 15 weeks (a) and 14 weeks (b)

Fig 25 Variation of disjoining pressure with equivalent 1047297lm thickness at various NaCl concentrations

Fig 26 Viscosityndashvolume fraction for WO emulsion stabilized with PHSndashPEOndashPHS

block copolymer experimental data calculated using Eq (22)

293T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 419

Clearly when χb12 B2 is positive and mixing is non-ideal leading to

positive deviation (repulsion) this occurs when the polymer chains

are in ldquogoodrdquo solvent conditions In contrast when χN12 B2 is

negative and mixing is non-ideal leading to negative deviation

(attraction) this occurs when the polymer chains are in ldquopoorrdquo

solvent conditions (precipitation of the polymer may occur under

these conditions) Since the polymer solvency depends on tempera-

ture one can also de1047297ne a theta temperature θ at which χ=12

The function [(12)minus

χ] can also be expressed in terms of twomixing parameters an enthalpy parameter κ 1 and an entropy

parameter ψ1 ie

1

2minusχ

= κ 1minusψ1 eth8THORN

The θ-temperature can also be de1047297ned in terms of κ 1 and ψ1

θ = κ 1T

ψ1

eth9THORN

Alternatively one can write

1

2

minusχ = ψ1 1minusθ

T eth10THORN

Although the FloryndashHuggins theory is sound in principle several

experimental results cannot be accounted for For example it was

found that the χ parameter depends on the polymer concentration in

solution Most serious is the fact that many polymer solutions (such as

PEO) show phase separation on heating when the theory predictsthat

it should happen only on cooling

The solution properties of copolymers are much more complicated

This is due to the fact that the two copolymer components A and B

behave differently in different solvents Only when the two compo-

nents are both soluble in the same solvent then they exhibit similar

solution properties This is the case for example for a non-polar

copolymer in a non-polar solvent It should also be emphasised that

the FloryndashHuggins theory was developed for ideal linear polymers

Indeed with branched polymers consisting of high monomer density

(eg star branched polymers) the θ-temperature depends on the

lengthof the armsand isin general lower thanthatof a linearpolymer

with the same molecular weight

Another complication arises from speci1047297c interaction with the

solvent eg hydrogen bonding between polymer and solvent

molecules (eg with PEO and PVA in water) Also aggregation insolution (lack of complete dissolution) may present another problem

One of the most useful parameters for characterising the

conformation of a polymer in solution is the root mean-square (rms)

end to end length br 2N12 which represents a con1047297guration character r

as the distance from one end group to the other of a chain molecule

Another useful parameter is the radius of gyration bs2N

12 which is a

measure of the effective size of a polymer molecule (it is the root

mean-square distance of the elements of the chain from its centre of

gravity)

For linear polymers

bs2N

1=2 = br 2N1=2

61=2 eth11THORN

The radius of gyration of a polymer in solution can be determined

from light scattering measurements

As mentioned above dilute solutions of copolymers is solvents that

are good for both components exhibit similar behaviour to hompo-

lymer chains However in a selective solvent whereby the medium is

a good solvent for one component say A and a poor solvent for the

second component B the very different solvent af 1047297nities to the two

components will have a large effect on the conformation of the

isolated chain This results in formation of aggregates involving

several macromolecules in dilute solutions (low Critical Aggregation

Concentration CAC) It is believed that the polymeric aggregates are

Fig 3 Various conformations of macromolecules on a plane surface

284 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 519

spherical [5] The CAC of these block and graft copolymers is usually

very low

Several methods may be applied to obtain the aggregate size and

shape of block and graft copolymers of which light scattering small

angle X-ray and neutron scattering are probably the most direct

Dynamic light scattering (photon correlation spectroscopy) can also

be applied to obtain the hydrodynamic radius of the aggregate This

technique is relatively easy to perform when compared with static

light scattering since it does not require special preparation of thesample

13 Adsorption and conformation of polymeric surfactants at interfaces

Understanding the adsorption and conformation of polymeric

surfactants at interfaces is key to knowing how these molecules act as

stabilizers Most basic ideas on adsorption and conformation of

polymers have been developed for the solidliquid interface [10ndash17]

The process of polymer adsorption is fairly complicated mdash it

involves PolymerSurface Interaction PolymerSolvent Interaction

SurfaceSolvent Interaction The con1047297guration (conformation) of the

polymer at the solidliquid interface The polymersurface interaction

is described in terms of adsorption energy per segment χs The

polymersolvent interaction is described in terms of the Floryndash

Huggins interaction parameter χ The polymer con1047297guration is

described by sequences of Trains segments in direct contact with

the surface Loops segments in between the trains that extend into

solution Tails ends of the molecules that also extend into solution A

schematic representation of the various polymer con1047297gurations is

given in Fig 3

For homopolymers eg poly(ethylene oxide) (PEO) or polyvinyl

pyrrolidone (PVP) a train-loop-tail con1047297guration is the case For

adsorption to occur a minimum energy of adsorption per segment χs

is required When a polymer molecule adsorbs on a surface it looses

con1047297gurational entropy and this must be compensated by an

adsorption energy χs per segment This is schematically shown in

Fig 4 where the adsorbed amount Γ is plotted versus χs The

minimum value of χs can be very small (b01kT ) since a large number

of segments per molecule are adsorbed For a polymer with say 100segments and 10 of these are in trains the adsorption energy per

molecule now reaches 1kT (with χs =01kT ) For 1000 segments the

adsorption energy per molecule is now 10kT

Homopolymers are not the most suitable for stabilisation of

dispersions For strong adsorption one needs the molecule to be

ldquoinsolublerdquo in the medium and has strong af 1047297nity (ldquoanchoringrdquo) to the

surface For stabilisation one needs the molecule to be highly soluble

in themedium andstronglysolvated by itsmolecules mdash This requires a

FloryndashHuggins interaction parameter less than 05 The above

opposing effects can be resolved by introducing ldquoshortrdquo blocks in the

moleculewhichare insolublein themedium andhavea strong af 1047297nity

to the surface as illustrated in Fig 3b Example Partially hydrolysed

polyvinyl acetate (88 hydrolysed ie with 12 acetate groups)

usually referred to as polyvinyl alcohol (PVA)

The above requirements are better satis1047297ed using AndashB AndashBndashA and

BAn graft copolymers B is chosento be highly insolublein themedium

and it should have high af 1047297nity to the surface This is essential to

ensure strong ldquoanchoringrdquo to the surface (irreversible adsorption) A is

chosen to be highly soluble in the medium and strongly solvated by

its molecules The FloryndashHuggins χ parameter can be applied in this

case For a polymer in a good solvent χ has to be lower than 05 the

smaller the χ value the better the solvent for the polymer chains

Examples of B for a hydrophobic particles in aqueous media are

polystyrene polymethylmethacrylate Examples of A in aqueous

media are polyethylene oxide polyacrylic acid polyvinyl pyrollidone

and polysaccharides For non-aqueous media such as hydrocarbons

the A chain(s) could be poly(12-hydroxystearic acid)

For full description of polymer adsorption one needs to obtain

information on the following (i) The amount of polymer adsorbed Γ (in mg or moles) per unit area of the particles It is essential to know

thesurface area of theparticles in thesuspensionNitrogenadsorption

on the powder surface may give such information (by application of

the BET equation) provided there will be no change in area on

dispersing the particles in the medium For many practical systems a

change in surface area may occur on dispersing the powder in which

case one has to use dye adsorption to measure the surface area (some

assumptions have to be made in this case) (ii)The fraction of segments

in direct contact with thesurface ie the fraction of segmentsin trains

p ( p = (Number of segments in direct contact with the surface) Total

Number) (iii)The distribution of segments in loops and tails ρ(z)

which extend in several layersfrom the surface ρ(z) is usually dif 1047297cult

to obtain experimentally although recently application of small angle

neutron scattering could obtain such information An alternative anduseful parameter for assessing ldquosteric stabilisationrdquo is the hydro-

dynamic thickness δh (thickness of the adsorbed or grafted polymer

layer plus any contribution from the hydration layer) Several methods

can be applied to measure δh as will be discussed below

2 Theories of polymer adsorption

Two main approaches have been developed to treat the problem of

polymer adsorption (i) Random Walk approach This is based on

Florys treatment of the polymer chain in solution the surface was

considered as a re1047298ecting barrier (ii) Statistical mechanical approach

Thepolymer con1047297gurationwas treated as being made of three types of

structures trains loops and tails each having a different energy state

The random walk approach is an unrealistic model to the problemof polymer adsorption since the polymer interacts in a speci1047297c

manner with the surface and the solvent The statistical mechanical

approach is a more realistic model for the problem of polymer

adsorption since it takes into account the various interactions

involved A useful model for treating polymer adsorption and

con1047297guration was suggested by Scheutjens and Fleer (SF theory)

[14ndash16] that is referred to as the step weighted random walk approach

which is summarized below

The SF theory starts without any assumption for the segment

density distribution The partition functions were derived for the

mixture of free and adsorbed polymer molecules as well as for the

solvent molecules All chain conformations were described as step

weighted random walks on a quasi-crystalline lattice which extends in

parallel layers from the surface mdash

this is schematically shown in Fig 5Fig 4 Variation of adsorption amount Γ with adsorption energy per segment χs

285T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 619

The partition function is written in terms of a number of con1047297gura-

tions These were treated as connectedsequencesof segments In each

layer random mixing between segments and solvent molecules was

assumed (mean 1047297eld approximation) Each step in the random walk

was assigned a weighting factor pi

pi consists of three contributions (i) An adsorption energyχs (which

exists only for the segments that are near the surface) (ii) Con1047297gura-

tional entropy of mixing (that exists in each layer) (iii) Segment-solvent

interaction parameterχ (the FloryndashHuggins interaction parameter note

that χ=0 for an athermal solvent χ=05 for a θ-solvent)

From the weighting factors the statistical weight of any chain

conformation can be calculated for a given segment concentration

pro1047297le using a matrix formulation Fig 6 shows typical adsorption

isotherms plotted as surface coverage (in equivalent monolayers)

versus polymer volume fraction ϕ in bulk solution (ϕ was taken to

vary between 0 and 10minus3 which is the normal experimental range)

The results in Fig 6 show the effect of increasing the chain length r

and effect of solvency (athermal solvent with χ=0 and theta solvent

with χ=05) The adsorption energy χs was taken to be the same and

equal to 1kT When r = 1 θ is very small and the adsorption increases linearly

with increase of ϕ (Henrys type isotherm) On the other hand when

r = 10 the isotherm deviates much from a straight line and approaches

a Langmuirian type However when r ge20 high af 1047297nity isotherms are

obtained This implies that the 1047297rst added polymer chains are

completely adsorbed resulting in extremely low polymer concentra-

tion in solution (approaching zero) This explains the irreversibility of

adsorption of polymeric surfactants with r N100 The adsorption

isotherms with r =100 and above are typical of those observed

experimentally for most polymers that are not too polydisperse ie

showing a steep rise followed by a nearly horizontal plateau (which

only increases few percent per decade increase of ϕ) In these dilute

solutions the effect of solvency is most clearly seen with poor

solvents giving the highest adsorbed amounts In good solvents θ is

much smaller and levels off for long chains to attain an adsorption

plateau which is essentially independent of molecular weight

Another point that emerges from the SF theory is the difference in

shape between the experimental and theoretical adsorption isotherms

in thelow concentration region Theexperimentalisotherms areusuallyrounded whereas those predicted fromtheory are1047298at This is accounted

for in terms of the molecular weight distribution (polydispersity) which

is encountered with most practical polymers This effect has been

explained by CohenndashStuart et al [17] With polydisperse polymers the

larger molecular weight fractions adsorb preferentially over the smaller

ones At low polymer concentrations nearly all polymer fractions are

adsorbed leaving a small fraction of the polymer with the lowest

molecular weights in solution As the polymer concentration is

increased the higher molecular weight fractions displace the lower

ones on thesurfacewhichare nowreleasedin solution thusshiftingthe

molecular weight distribution in solution to lower values This process

continues with further increase in polymer concentration leading to

fractionation whereby the higher molecular weight fractions are

adsorbed at the expense of the lower fractions which are released to

the bulk solution However in very concentrated solutions monomers

adsorb preferentially with respect to polymers and short chains with

respect to larger ones This is due to the fact that in this region the

conformational entropy term predominates the free energy disfavour-

ing the adsorption of long chains

According to the SF theory the bound fraction p is high at low

concentration and relatively independent of molecular weight when

r N20 However with increase in surface coverage andor molecular

weight p tends to decrease indicating the formation of larger loops

and tails

The structure of the adsorbed layer is described in terms of the

segment density distribution As an illustration Fig 7 shows some

calculations using the SF theory for loops and tails with r =1000

ϕ= 10minus6 and χ=05 In this example 38 of the segments are in

trains 555 in loops and 65 in tails This theory demonstrates theimportance of tails which dominate the total distribution in the outer

region

3 Experimental techniques for studying polymeric surfactant

adsorption

As mentioned above for full characterization of polymeric sur-

factant adsorption one needs to determine three parameters (i) The

adsorbed amount Γ (mg mminus2 or mol mminus2) as a function of equilibrium

concentration C eq ie the adsorption isotherm (ii) The fraction of

segments in direct contact with the surface p (number of segments in

Fig 5 Schematic representation of a polymer molecule adsorbing on a 1047298at surface mdash

quasi-crystalline lattice with segments 1047297lling layers that are parallel to the surface

(random mixing of segments and solvent molecules in each layer is assumed)

Fig 6 Adsorption isotherms for oligomers and polymers in the dilute region based on

the SF theory Fig 7 Loop tail and total segment pro1047297le according to the SF theory

286 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 719

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 819

The microelectrophoresis technique is based on measurement of

the electrophoretic mobility u of the particles in the presence and

absence of the polymer layer From u one can calculate the zeta

potential ζ using the Huckel equation (which is applicable for small

particles and extended double layers ie κ Rbb1 where κ is the

DebyendashHuckel parameter that is related to the salt concentration

u = 23eeon η

eth15THORN

where ε is the relative permittivity of the medium and ε o is the

permittivity of free space

By measuring ζ of the particles with and without the adsorbed

polymer layer one can obtain the hydrodynamic thickness δh For

accurate measurements one should carry the measurements at various

electrolyte concentrations and extrapolate the results to the plateau

value Several automatic instruments are available for measurement of

the electrophoretic mobility Malvern zeta sizerndashCoulter Delsa sizer and

Broekahven instrument All these instruments are easy to use and the

measurement can be carried out within few minutes

4 Examples of the adsorption isotherms of nonionic polymericsurfactants

Fig 9 shows the adsorption isotherms for PEO with different

molecular weights on PS (at room temperature It can be seen that the

amount adsorbed in mgm-2 increases with increase in the polymer

molecular weight Fig 10 shows the variation of the hydrodynamic

thickness δh with molecular weight M δh shows a linear increase with

log M δh increases with n the number of segments in the chain

according to

δhasympn08 eth16THORN

Fig 11 shows the adsorption isotherms of PVA with various

molecular weights on PS latex (at 25 degC) [19] The polymers were

obtained by fractionation of a commercial sample of PVA with an

average molecular weight of 45000 The polymer also contained 12

vinyl acetate groups As with PEO the amount of adsorption increases

with increase in M The isotherms are also of the high af 1047297nity type Γ at the plateau increases linearly with M 12

The hydrodynamic thickness was determined using PCS and theresults are given below

M 67000 43000 28000 17000 8000

δhnm 255 197 140 98 33

δh seems to increase linearly with increase in the molecular weight

The effect of solvency on adsorption was investigated by increasing

the temperature (the PVA molecules are less soluble at higher

temperature) or addition of electrolyte (KCl) [20] The results are

shown in Figs 12 and 13 for M =65100 As can be seen from Fig 12

increase of temperature results in reduction of solvency of the

mediumfor the chain (due to break down of hydrogen bonds) and this

results in an increase in the amount adsorbed Addition of KCl (which

reduces the solvency of the medium for the chain) results in anincrease in adsorption (as predicted by theory)

The adsorption of block and graft copolymers is more complex

since the intimate structure of the chain determines the extent of

adsorption [18] Randomcopolymers adsorbin an intermediate way to

that of the corresponding homopolymers Blockcopolymers retain the

adsorption preference of the individual blocks The hydrophilic block

(eg PEO) the buoy (previously referred to as the A chain) extends

away from the particle surface into the bulk solution whereas the

hydrophobic anchor block (previously referred to as the B chain) (eg

PS or PPO) provides 1047297rm attachment to the surface Fig 14 shows the

theoretical prediction of diblock copolymer adsorption according to

Fig 9 Adsorption isotherms for PEO on PS

Fig 10 Hydrodynamic thickness of PEO on PS as a function of the molecular weight

Fig 11 Adsorption isotherms of PVA with different molecular weights on polystyrene

latex at 25 degC

Fig 12 In1047298uence of temperature on adsorption

288 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 919

the Scheutjens and Fleer theory The surface density σ is plotted

versus the fraction of anchor segmentsν A The adsorption depends on

the anchorbuoy composition

The amount of adsorption is higher than for homopolymers and

the adsorbed layer thickness is more extended and dense Fig 15

shows the theoretical prediction for the adsorbed layer thickness δwhich is plotted as a function of ν A For a triblock copolymer AndashBndashA

with two buoy chains (A) and one anchor chain (B) the behaviour is

similar to that of diblock copolymers This is shown in Fig16 for PEOndash

PPOndashPEO block (Pluronic)

5 Interaction between particles containing adsorbed polymeric

surfactant layers steric stabilization

When two particles each with a radius R and containing an

adsorbed polymer layer with a hydrodynamic thickness δh approach

each other to a surfacendashsurface separation distance h that is smaller

than 2 δh the polymer layers interact with each other resulting in two

main situations [21] (i) The polymer chains may overlap with each

other (ii) The polymer layer may undergo some compression In both

cases there will be an increase in the local segment density of the

polymer chains in the interaction region This is schematicallyillustrated in Fig 17 The real situation is perhaps in between the

above two cases ie the polymer chains may undergo some

interpenetration and some compression

Provided the dangling chains (the A chains in AndashB AndashBndashA block or

BAn graft copolymers) are in a good solvent this local increase in

segment density in the interaction zone will result in strong repulsion

as a result of two main effects (i) Increase in the osmotic pressure in

the overlap region as a result of the unfavourable mixing of the

polymer chains when these are in good solvent conditions This is

referred to as osmotic repulsion or mixing interaction and it is

described by a free energy of interaction Gmix (ii) Reduction of the

con1047297gurational entropy of the chains in the interaction zone this

entropy reduction results from the decrease in the volume available

for thechains when these areeither overlapped or compressed This is

referred to as volume restriction interaction entropic or elastic

interaction and it is described by a free energy of interaction Gel

Combination of Gmix and Gel is usually referred to as the steric

interaction free energy Gs ie

Gs = Gmix + Gel eth17THORN

The sign of Gmix depends on the solvency of the medium for the

chains If in a good solvent ie the FloryndashHuggins interaction

parameter χ is less than 05 then Gmix is positive and the mixing

interaction leads to repulsion (see below) In contrast if χN05 (ie the

chains are in a poor solvent condition) Gmix is negative and the

mixing interaction becomes attractive Gel is always positive and

hence in some cases one can produce stable dispersions in a relatively

poor solvent (enhanced steric stabilisation)

51 Mixing interaction Gmix

This results from the unfavourable mixing of the polymer chainswhen these are in a good solvent conditions This is schematically

shown in Fig 18

Consider two spherical particles with the same radius and each

containing an adsorbed polymer layer with thickness δ Before

overlap one can de1047297ne in each polymer layer a chemical potential

for the solvent μ iα and a volume fraction for the polymer in the

layerϕ2 In the overlap region (volume element dV ) the chemical

potential of the solvent is reduced to μ i β This results from the increase

in polymer segment concentration in this overlap region

In the overlap region the chemical potential of the polymer chains

is now higher than in the rest of the layer (with no overlap) This

Fig 13 In1047298uence of addition of KCl on adsorption

Fig 14 Prediction of Adsorption of diblock copolymer

Fig15 Theoretical predictions of the adsorbed layer thickness for a diblock copolymer

Fig 16 Adsorbed amount (mg mminus2) versus fraction of anchor segment for an AndashBndashA

triblock copolymer (PEOndash

PPOndash

PEO)

289T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1019

amounts to an increase in the osmotic pressure in the overlap region

as a result solvent will diffuse from thebulk to the overlap region thus

separating the particles and hence a strong repulsive energy arises

from this effect The above repulsive energy can be calculated by

considering the free energy of mixing of two polymer solutions as for

example treated by Flory and Krigbaum [22] The free energy of

mixing is given by two terms (i) An entropy term that depends on the

volume fraction of polymer and solvent (ii)An energy term that is

determined by the FloryndashHuggins interaction parameter χ

Using the above theory one can derive an expression for the free

energy of mixing of two polymer layers (assuming a uniform segment

density distribution in each layer) surrounding two spherical particles

as a function of the separation distance h between the particles The

expression for Gmix is

Gmix

kT =

2V 22V 1

m2

2

1

2minusχ

δminus

h

2

2

3R + 2δ + h

2

eth18THORN

k is the Boltzmann constant T is the absolute temperature V 2 is the

molar volume of polymer V 1 is the molar volume of solvent and ν 2 is

the number of polymer chains per unit area

The sign of Gmix depends on the value of the FloryndashHuggins

interaction parameter χ if χb05 Gmix is positive and the interaction

is repulsive if χN05 Gmix is negative and the interaction is attractiveif χ=05 Gmix =0 and this de1047297nes the θ-condition

52 Elastic interaction Gel

Thisarises fromthe lossin con1047297gurational entropy of thechains on

the approach of a second particle As a result of this approach the

volume available for the chains becomes restricted resulting in loss of

the number of con1047297gurations This can be illustrated by considering a

simple molecule represented by a rod that rotates freely in a

hemisphere across a surface (Fig 19) When the two surfaces are

separated byan in1047297nitedistanceinfin thenumber of con1047297gurations of the

rod is Ω(infin) which is proportional to the volume of the hemisphere

When a secondparticleapproaches to a distance h suchthatit cuts the

hemisphere (loosing some volume) the volume available to the chains

is reduced and the number of con1047297gurations become Ω(h) which is

less than Ω(infin) For two 1047298at plates Gel is given by the following

expression

Gel

kT = 2m2 ln

X heth THORN

X infineth THORN

= 2m2Rel heth THORN eth19THORN

where Rel

(h) is a geometric function whose form depends on the

segment density distribution It should be stressed that Gel is always

positive and could play a major role in steric stabilisation It becomes

very strong when the separation distance between the particles

becomes comparable to the adsorbed layer thickness δ

Combination of Gmix and Gel with GA (the van der Waals attractive

energy)gives the total free energy of interaction GT (assuming there is

no contribution from any residual electrostatic interaction) ie

GT = Gmix + Gel + GA eth20THORN

A schematic representation of the variation of Gmix Gel GA and GT

with surfacendashsurface separation distance h is shown in Fig 20

Gmix increases very sharply with decrease of h when hb2δ Gel

increases very sharply with decrease of h when hbδ GT versus h

shows a minimum Gmin at separation distances comparable to 2δ

When h b2δ GT shows a rapid increase with decrease in h

The depth of the minimum depends on the Hamaker constant A

the particle radius R and adsorbed layer thickness δ Gmin increases

with increase of A and R At a given A and R Gmin increases with

decrease in δ (ie with decrease of the molecular weight M w of the

stabiliser This is illustrated in Fig 21 which shows the energyndash

distance curves as a function of δR The larger the value of δR the

smaller the value of Gmin In this case the system may approach

thermodynamic stability as is the case with nano-dispersions

6 Emulsions stabilized by polymeric surfactants

The most effective method for emulsion stabilization is to use

polymeric surfactants that stronglyadsorb at the OWor WO interface

Fig 17 Schematic representation of the interaction between particles containing adsorbed polymer layers

Fig 18 Schematic representation of polymer layer overlap

Fig 19 Schematic representation of con1047297gurational entropy loss on approach of a

second particle

290 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1119

and produce effective steric stabilization against strong 1047298occulation

coalescence and Ostwald ripening [23]

As mentioned above a graft copolymer of the ABn type was

synthesized by grafting several alkyl groups on an inulin (polyfruc-

tose) chain The polymeric surfactant (INUTECregSP1) consists of a

linear polyfructose chain (the stabilizing A chain) and several alkyl

groups (the B chains) that provide multi-anchor attachment to the oil

droplets This polymeric surfactant produces enhanced steric stabili-

zation both in water and high electrolyte concentrations as will be

discussed later

For water-in-oil (WO) emulsions an AndashBndashA block copolymer of

poly (12-hydroxystearic acid) (PHS) (the A chains) and poly (ethylene

oxide) (PEO) (the B chain) PHSndashPEOndashPHS is commercially available

(Arlacel P135 UNIQEMA) The PEO chain (that is soluble in the water

droplets) forms the anchor chain whereas the PHS chains form the

stabilizing chains PHS is highly soluble in most hydrocarbon solvents

and is strongly solvated by its molecules The structure of the PHSndash

PEOndashPHS block copolymer is schematically shown in Fig 22The conformation of the polymeric surfactant at the WO interface

is schematically shown in Fig 23

Emulsions of Isopar Mwater and cyclomethiconewater were

prepared using INUTECregSP1 5050 (vv) OW emulsions were

prepared and the emulsi1047297er concentration was varied from 025 to 2

(wv) based on the oil phase 05 (wv) emulsi1047297erwas suf 1047297cient for

stabilization of these 5050 (vv) emulsions [23]

The emulsions were stored at room temperature and 50 degC and

optical micrographs were taken at intervals of time (for a year) in

order to check the stability Emulsions prepared in water were very

stableshowingno change in droplet size distributionover more than a

year period and this indicated absence of coalescence Any weak

1047298occulation that occurred was reversible and the emulsion could be

redispersed by gentle shaking Fig 24 shows an optical micrograph fora dilute 5050 (vv) emulsion that was stored for 15 and 14 weeks at

50 degC

No change in droplet size was observed after storage for more than

1 year at 50 degC indicating absence of coalescence The same result was

obtained when using different oils Emulsions were also stable against

coalescence in the presence of high electrolyte concentrations (up to

4 mol dmminus3 or ~25 NaCl

The above stability in high electrolyte concentrations is not

observed with polymeric surfactants based on polethylene oxide

The high stability observed using INUTECregSP1 is related to its

strong hydration both in water and in electrolyte solutions The

hydration of inulin (the backbone of HMI) could be assessed using

cloud point measurements A comparison was also made with PEO

with two molecular weights namely 4000 and 20000

Solutionsof PEO 4000 and 20000 showeda systematic decrease of

cloud point with increase in NaCl or MgSO4 concentration In contrast

inulin showed no cloud point up to 4 mol dmminus3 NaCl and up to 1 mol

dmminus3 MgSO4

The above results explain the difference between PEO and inulin

With PEO the chains show dehydration when the NaCl concentration

is increased above 2 mol dmminus3 or 05 mol dmminus3 MgSO4 The inulin

chains remain hydrated at much higher electrolyte concentrations It

seems that the linear polyfructose chains remain strongly hydrated athigh temperature and high electrolyte concentrations

The high emulsion stability obtained whenusingINUTECregSP1 can be

accounted for by the following factors (i) The multi-point attachment

of the polymer by several alkyl chains that are grafted on the backbone

(ii) The strong hydration of the polyfructose ldquoloopsrdquo both in water and

high electrolyte concentrations (χ remains below 05 under these

conditions) (iii) Thehigh volumefraction (concentration) of the loops at

the interface (iv) Enhanced steric stabilization this is the case with

multi-point attachment which produces strong elastic interaction

Evidence for the high stability of the liquid 1047297lm between emulsion

droplets when using INUTECregSP1 was obtained by Exerowa et al [24]

using disjoining pressure measurements This is illustrated in Fig 25

which shows a plot of disjoining pressure versus separation distance

between two emulsion droplets at various electrolyte concentrations

The results show that by increasing the capillary pressure a stable

Newton Black Film (NBF) is obtained at a 1047297lm thickness of sim7 nm

The lack of rupture of the 1047297lm at the highest pressure applied of

45times104 Pa indicate the high stability of the 1047297lm in water and in high

electrolyte concentrations (up to 20 mol dmminus3 NaCl)

The lack of rupture of the NBF up to the highest pressure applied

namely 45times104 Pa clearly indicatesthe high stability of the liquid1047297lm

in the presence of high NaCl concentrations (up to 2 mol dm minus3) This

result is consistent with the high emulsion stability obtained at high

electrolyte concentrations and high temperature Emulsions of Isopar

M-in-water are very stable under such conditions and this could be

accounted for by the high stability of the NBF The droplet size of

5050 OW emulsions prepared using 2 INUTEC regSP1 is in the region

of 1ndash10 μ m This corresponds to a capillary pressure of ~3times104 Pa for

the 1 μ m drops and ~3times103 Pa for the 10 μ m drops These capillarypressures are lower than those to which the NBF have been sub-

jected to and this clearly indicates the high stability obtained against

coalescence in these emulsions

61 WO emulsions stabilized with PHS ndashPEOndashPHS block copolymer

WO emulsions (the oil being Isopar M) were prepared using PHSndash

PEOndashPHS block copolymer at high water volume fractions (N07) The

emulsions have a narrow droplet size distribution with a z -average

radius of 183 nm [25] They also remained 1047298uid up to high water

volume fractions (N06) This could be illustrated from viscosityndash

volume fraction curves as is shown in Fig 26

Fig 20 Energyndashdistance curves for sterically stabilized systems

Fig 21 Variation of Gmin with δR

291T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1219

The effective volume fraction ϕeff of the emulsions (the core

droplets plus the adsorbed layer) could be calculated from the relative

viscosity and using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth21THORN

Where η r is the relative viscosity ϕp is the maximum packing fraction(sim07) and [ η ] is the intrinsic viscosity that is equal to 25 for hard-

spheres

The calculations based on Eq (22) are shown in Fig 27 (square

symbols) From the effective volume fraction ϕeff and the core volume

fraction ϕ the adsorbed layer thickness could be calculated This was

found to be in the region of 10 nm at ϕ =04 and it decreased with

increase in ϕ

The WO emulsions prepared using the PHSndashPEOndashPHS block

copolymer remained stable both at room temperature and 50 degC

This is consistent with the structure of the block copolymer the B

chain (PEO) is soluble in water and it forms a very strong anchor at the

WO interface The PHS chains (the A chains) provide effective steric

stabilization since the chains are highly soluble in Isopar M and are

strongly solvated by its molecules

7 Suspensions stabilised using polymeric surfactants

There are generally two procedures for preparation of solidliquid

dispersions

(i) Condensation methods build-up of particles from mole-

cular units ie nucleation and growth A special procedure

is the preparation of latexes by emulsion or dispersion

polymerization

(ii) Dispersion methods in this case one starts with preformed

large particles or crystals which are dispersed in the liquid by

using a surfactant (wetting agent) with subsequent breaking up

of the large particles by milling (comminution) to achieve the

desirable particle size distribution A dispersing agent (usuallya

polymeric surfactant) is used for the dispersion process and

subsequent stabilization of the resulting suspension

There are generally two procedures for preparation of latexes

(i) Emulsion polymerization the monomers that are essentially

insoluble in the aqueous medium are emulsi1047297ed using a

surfactant and an initiator is added while heating the system

to produce the polymer particles that are stabilized electro-

statically (when using ionic surfactants) or sterically (when

using non-ionic surfactants)

(ii) Dispersion polymerization the monomers are dissolved in a

solvent in which the resulting polymer particlesare insoluble A

protective colloid (normally a block or graft copolymer) is

added to prevent 1047298occulation of the resulting polymers

particles that are produced on addition of an initiator This

method is usually applied for the production of non-aqueous

latex dispersions and is sometimes referred to as Non-Aqueous

Dispersion Polymerization (NAD)

Surfactants play a crucial role in the process of latex preparation

since they determine the stabilizing ef 1047297ciency and the effectiveness of

the surface active agent ultimately determines the number of particles

Fig 22 Schematic representation of the structure of PHSndashPEOndashPHS block copolymer

Fig 23 Conformation of PHSndash

PEOndash

PHS polymeric surfactant at the WO interface

292 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1319

and their size The effectiveness of any surface active agent in

stabilizing the particles is the dominant factor and the number of

micelles formed is relatively unimportant In the NAD process themonomer normally an acrylic is dissolved in a non-aqueous solvent

normally an aliphatic hydrocarbon and an oil soluble initiator and a

stabilizer (to protect the resulting particles from1047298occulation) is added

to the reaction mixture The most successful stabilizers used in NAD

are block and graft copolymers Preformed graft stabilizers based

on poly(12-hydroxy stearic acid) (PHS) are simple to prepare and

effective in NAD polymerization

Dispersion methods are used for the preparation of suspensions of

preformed particles The role of surfactants (or polymers) in the

dispersion process can be analyzed in terms of the three processesinvolved (i) Wetting of the powder by the liquid (ii) Breaking of the

aggregates and agglomerates (iii) Comminution of the resulting

particles and their subsequent stabilization All these processes are

affected by surfactants or polymers which adsorb on the powder

surface thus aiding the wetting of the powder break-up of the

aggregates and agglomerates and subsequent reduction of particle

size by wet milling

71 Polymeric surfactants in emulsion polymerization

Recently the graft copolymer of hydrophobically modi1047297ed inulin

(INUTECreg SP1) has been used in emulsion polymerization of styrene

methyl methacrylate butyl acrylate and several other monomers [26]All lattices were prepared by emulsion polymerisation using potas-

sium persulphate as initiator The z -average particle size was

determined by photon correlation spectroscopy (PCS) and electron

micrographs were also taken

Fig 24 Optical micrographs of OW emulsions stabilized with INUTEC regSP1 stored at

50 degC for 15 weeks (a) and 14 weeks (b)

Fig 25 Variation of disjoining pressure with equivalent 1047297lm thickness at various NaCl concentrations

Fig 26 Viscosityndashvolume fraction for WO emulsion stabilized with PHSndashPEOndashPHS

block copolymer experimental data calculated using Eq (22)

293T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 519

spherical [5] The CAC of these block and graft copolymers is usually

very low

Several methods may be applied to obtain the aggregate size and

shape of block and graft copolymers of which light scattering small

angle X-ray and neutron scattering are probably the most direct

Dynamic light scattering (photon correlation spectroscopy) can also

be applied to obtain the hydrodynamic radius of the aggregate This

technique is relatively easy to perform when compared with static

light scattering since it does not require special preparation of thesample

13 Adsorption and conformation of polymeric surfactants at interfaces

Understanding the adsorption and conformation of polymeric

surfactants at interfaces is key to knowing how these molecules act as

stabilizers Most basic ideas on adsorption and conformation of

polymers have been developed for the solidliquid interface [10ndash17]

The process of polymer adsorption is fairly complicated mdash it

involves PolymerSurface Interaction PolymerSolvent Interaction

SurfaceSolvent Interaction The con1047297guration (conformation) of the

polymer at the solidliquid interface The polymersurface interaction

is described in terms of adsorption energy per segment χs The

polymersolvent interaction is described in terms of the Floryndash

Huggins interaction parameter χ The polymer con1047297guration is

described by sequences of Trains segments in direct contact with

the surface Loops segments in between the trains that extend into

solution Tails ends of the molecules that also extend into solution A

schematic representation of the various polymer con1047297gurations is

given in Fig 3

For homopolymers eg poly(ethylene oxide) (PEO) or polyvinyl

pyrrolidone (PVP) a train-loop-tail con1047297guration is the case For

adsorption to occur a minimum energy of adsorption per segment χs

is required When a polymer molecule adsorbs on a surface it looses

con1047297gurational entropy and this must be compensated by an

adsorption energy χs per segment This is schematically shown in

Fig 4 where the adsorbed amount Γ is plotted versus χs The

minimum value of χs can be very small (b01kT ) since a large number

of segments per molecule are adsorbed For a polymer with say 100segments and 10 of these are in trains the adsorption energy per

molecule now reaches 1kT (with χs =01kT ) For 1000 segments the

adsorption energy per molecule is now 10kT

Homopolymers are not the most suitable for stabilisation of

dispersions For strong adsorption one needs the molecule to be

ldquoinsolublerdquo in the medium and has strong af 1047297nity (ldquoanchoringrdquo) to the

surface For stabilisation one needs the molecule to be highly soluble

in themedium andstronglysolvated by itsmolecules mdash This requires a

FloryndashHuggins interaction parameter less than 05 The above

opposing effects can be resolved by introducing ldquoshortrdquo blocks in the

moleculewhichare insolublein themedium andhavea strong af 1047297nity

to the surface as illustrated in Fig 3b Example Partially hydrolysed

polyvinyl acetate (88 hydrolysed ie with 12 acetate groups)

usually referred to as polyvinyl alcohol (PVA)

The above requirements are better satis1047297ed using AndashB AndashBndashA and

BAn graft copolymers B is chosento be highly insolublein themedium

and it should have high af 1047297nity to the surface This is essential to

ensure strong ldquoanchoringrdquo to the surface (irreversible adsorption) A is

chosen to be highly soluble in the medium and strongly solvated by

its molecules The FloryndashHuggins χ parameter can be applied in this

case For a polymer in a good solvent χ has to be lower than 05 the

smaller the χ value the better the solvent for the polymer chains

Examples of B for a hydrophobic particles in aqueous media are

polystyrene polymethylmethacrylate Examples of A in aqueous

media are polyethylene oxide polyacrylic acid polyvinyl pyrollidone

and polysaccharides For non-aqueous media such as hydrocarbons

the A chain(s) could be poly(12-hydroxystearic acid)

For full description of polymer adsorption one needs to obtain

information on the following (i) The amount of polymer adsorbed Γ (in mg or moles) per unit area of the particles It is essential to know

thesurface area of theparticles in thesuspensionNitrogenadsorption

on the powder surface may give such information (by application of

the BET equation) provided there will be no change in area on

dispersing the particles in the medium For many practical systems a

change in surface area may occur on dispersing the powder in which

case one has to use dye adsorption to measure the surface area (some

assumptions have to be made in this case) (ii)The fraction of segments

in direct contact with thesurface ie the fraction of segmentsin trains

p ( p = (Number of segments in direct contact with the surface) Total

Number) (iii)The distribution of segments in loops and tails ρ(z)

which extend in several layersfrom the surface ρ(z) is usually dif 1047297cult

to obtain experimentally although recently application of small angle

neutron scattering could obtain such information An alternative anduseful parameter for assessing ldquosteric stabilisationrdquo is the hydro-

dynamic thickness δh (thickness of the adsorbed or grafted polymer

layer plus any contribution from the hydration layer) Several methods

can be applied to measure δh as will be discussed below

2 Theories of polymer adsorption

Two main approaches have been developed to treat the problem of

polymer adsorption (i) Random Walk approach This is based on

Florys treatment of the polymer chain in solution the surface was

considered as a re1047298ecting barrier (ii) Statistical mechanical approach

Thepolymer con1047297gurationwas treated as being made of three types of

structures trains loops and tails each having a different energy state

The random walk approach is an unrealistic model to the problemof polymer adsorption since the polymer interacts in a speci1047297c

manner with the surface and the solvent The statistical mechanical

approach is a more realistic model for the problem of polymer

adsorption since it takes into account the various interactions

involved A useful model for treating polymer adsorption and

con1047297guration was suggested by Scheutjens and Fleer (SF theory)

[14ndash16] that is referred to as the step weighted random walk approach

which is summarized below

The SF theory starts without any assumption for the segment

density distribution The partition functions were derived for the

mixture of free and adsorbed polymer molecules as well as for the

solvent molecules All chain conformations were described as step

weighted random walks on a quasi-crystalline lattice which extends in

parallel layers from the surface mdash

this is schematically shown in Fig 5Fig 4 Variation of adsorption amount Γ with adsorption energy per segment χs

285T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 619

The partition function is written in terms of a number of con1047297gura-

tions These were treated as connectedsequencesof segments In each

layer random mixing between segments and solvent molecules was

assumed (mean 1047297eld approximation) Each step in the random walk

was assigned a weighting factor pi

pi consists of three contributions (i) An adsorption energyχs (which

exists only for the segments that are near the surface) (ii) Con1047297gura-

tional entropy of mixing (that exists in each layer) (iii) Segment-solvent

interaction parameterχ (the FloryndashHuggins interaction parameter note

that χ=0 for an athermal solvent χ=05 for a θ-solvent)

From the weighting factors the statistical weight of any chain

conformation can be calculated for a given segment concentration

pro1047297le using a matrix formulation Fig 6 shows typical adsorption

isotherms plotted as surface coverage (in equivalent monolayers)

versus polymer volume fraction ϕ in bulk solution (ϕ was taken to

vary between 0 and 10minus3 which is the normal experimental range)

The results in Fig 6 show the effect of increasing the chain length r

and effect of solvency (athermal solvent with χ=0 and theta solvent

with χ=05) The adsorption energy χs was taken to be the same and

equal to 1kT When r = 1 θ is very small and the adsorption increases linearly

with increase of ϕ (Henrys type isotherm) On the other hand when

r = 10 the isotherm deviates much from a straight line and approaches

a Langmuirian type However when r ge20 high af 1047297nity isotherms are

obtained This implies that the 1047297rst added polymer chains are

completely adsorbed resulting in extremely low polymer concentra-

tion in solution (approaching zero) This explains the irreversibility of

adsorption of polymeric surfactants with r N100 The adsorption

isotherms with r =100 and above are typical of those observed

experimentally for most polymers that are not too polydisperse ie

showing a steep rise followed by a nearly horizontal plateau (which

only increases few percent per decade increase of ϕ) In these dilute

solutions the effect of solvency is most clearly seen with poor

solvents giving the highest adsorbed amounts In good solvents θ is

much smaller and levels off for long chains to attain an adsorption

plateau which is essentially independent of molecular weight

Another point that emerges from the SF theory is the difference in

shape between the experimental and theoretical adsorption isotherms

in thelow concentration region Theexperimentalisotherms areusuallyrounded whereas those predicted fromtheory are1047298at This is accounted

for in terms of the molecular weight distribution (polydispersity) which

is encountered with most practical polymers This effect has been

explained by CohenndashStuart et al [17] With polydisperse polymers the

larger molecular weight fractions adsorb preferentially over the smaller

ones At low polymer concentrations nearly all polymer fractions are

adsorbed leaving a small fraction of the polymer with the lowest

molecular weights in solution As the polymer concentration is

increased the higher molecular weight fractions displace the lower

ones on thesurfacewhichare nowreleasedin solution thusshiftingthe

molecular weight distribution in solution to lower values This process

continues with further increase in polymer concentration leading to

fractionation whereby the higher molecular weight fractions are

adsorbed at the expense of the lower fractions which are released to

the bulk solution However in very concentrated solutions monomers

adsorb preferentially with respect to polymers and short chains with

respect to larger ones This is due to the fact that in this region the

conformational entropy term predominates the free energy disfavour-

ing the adsorption of long chains

According to the SF theory the bound fraction p is high at low

concentration and relatively independent of molecular weight when

r N20 However with increase in surface coverage andor molecular

weight p tends to decrease indicating the formation of larger loops

and tails

The structure of the adsorbed layer is described in terms of the

segment density distribution As an illustration Fig 7 shows some

calculations using the SF theory for loops and tails with r =1000

ϕ= 10minus6 and χ=05 In this example 38 of the segments are in

trains 555 in loops and 65 in tails This theory demonstrates theimportance of tails which dominate the total distribution in the outer

region

3 Experimental techniques for studying polymeric surfactant

adsorption

As mentioned above for full characterization of polymeric sur-

factant adsorption one needs to determine three parameters (i) The

adsorbed amount Γ (mg mminus2 or mol mminus2) as a function of equilibrium

concentration C eq ie the adsorption isotherm (ii) The fraction of

segments in direct contact with the surface p (number of segments in

Fig 5 Schematic representation of a polymer molecule adsorbing on a 1047298at surface mdash

quasi-crystalline lattice with segments 1047297lling layers that are parallel to the surface

(random mixing of segments and solvent molecules in each layer is assumed)

Fig 6 Adsorption isotherms for oligomers and polymers in the dilute region based on

the SF theory Fig 7 Loop tail and total segment pro1047297le according to the SF theory

286 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 719

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 819

The microelectrophoresis technique is based on measurement of

the electrophoretic mobility u of the particles in the presence and

absence of the polymer layer From u one can calculate the zeta

potential ζ using the Huckel equation (which is applicable for small

particles and extended double layers ie κ Rbb1 where κ is the

DebyendashHuckel parameter that is related to the salt concentration

u = 23eeon η

eth15THORN

where ε is the relative permittivity of the medium and ε o is the

permittivity of free space

By measuring ζ of the particles with and without the adsorbed

polymer layer one can obtain the hydrodynamic thickness δh For

accurate measurements one should carry the measurements at various

electrolyte concentrations and extrapolate the results to the plateau

value Several automatic instruments are available for measurement of

the electrophoretic mobility Malvern zeta sizerndashCoulter Delsa sizer and

Broekahven instrument All these instruments are easy to use and the

measurement can be carried out within few minutes

4 Examples of the adsorption isotherms of nonionic polymericsurfactants

Fig 9 shows the adsorption isotherms for PEO with different

molecular weights on PS (at room temperature It can be seen that the

amount adsorbed in mgm-2 increases with increase in the polymer

molecular weight Fig 10 shows the variation of the hydrodynamic

thickness δh with molecular weight M δh shows a linear increase with

log M δh increases with n the number of segments in the chain

according to

δhasympn08 eth16THORN

Fig 11 shows the adsorption isotherms of PVA with various

molecular weights on PS latex (at 25 degC) [19] The polymers were

obtained by fractionation of a commercial sample of PVA with an

average molecular weight of 45000 The polymer also contained 12

vinyl acetate groups As with PEO the amount of adsorption increases

with increase in M The isotherms are also of the high af 1047297nity type Γ at the plateau increases linearly with M 12

The hydrodynamic thickness was determined using PCS and theresults are given below

M 67000 43000 28000 17000 8000

δhnm 255 197 140 98 33

δh seems to increase linearly with increase in the molecular weight

The effect of solvency on adsorption was investigated by increasing

the temperature (the PVA molecules are less soluble at higher

temperature) or addition of electrolyte (KCl) [20] The results are

shown in Figs 12 and 13 for M =65100 As can be seen from Fig 12

increase of temperature results in reduction of solvency of the

mediumfor the chain (due to break down of hydrogen bonds) and this

results in an increase in the amount adsorbed Addition of KCl (which

reduces the solvency of the medium for the chain) results in anincrease in adsorption (as predicted by theory)

The adsorption of block and graft copolymers is more complex

since the intimate structure of the chain determines the extent of

adsorption [18] Randomcopolymers adsorbin an intermediate way to

that of the corresponding homopolymers Blockcopolymers retain the

adsorption preference of the individual blocks The hydrophilic block

(eg PEO) the buoy (previously referred to as the A chain) extends

away from the particle surface into the bulk solution whereas the

hydrophobic anchor block (previously referred to as the B chain) (eg

PS or PPO) provides 1047297rm attachment to the surface Fig 14 shows the

theoretical prediction of diblock copolymer adsorption according to

Fig 9 Adsorption isotherms for PEO on PS

Fig 10 Hydrodynamic thickness of PEO on PS as a function of the molecular weight

Fig 11 Adsorption isotherms of PVA with different molecular weights on polystyrene

latex at 25 degC

Fig 12 In1047298uence of temperature on adsorption

288 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 919

the Scheutjens and Fleer theory The surface density σ is plotted

versus the fraction of anchor segmentsν A The adsorption depends on

the anchorbuoy composition

The amount of adsorption is higher than for homopolymers and

the adsorbed layer thickness is more extended and dense Fig 15

shows the theoretical prediction for the adsorbed layer thickness δwhich is plotted as a function of ν A For a triblock copolymer AndashBndashA

with two buoy chains (A) and one anchor chain (B) the behaviour is

similar to that of diblock copolymers This is shown in Fig16 for PEOndash

PPOndashPEO block (Pluronic)

5 Interaction between particles containing adsorbed polymeric

surfactant layers steric stabilization

When two particles each with a radius R and containing an

adsorbed polymer layer with a hydrodynamic thickness δh approach

each other to a surfacendashsurface separation distance h that is smaller

than 2 δh the polymer layers interact with each other resulting in two

main situations [21] (i) The polymer chains may overlap with each

other (ii) The polymer layer may undergo some compression In both

cases there will be an increase in the local segment density of the

polymer chains in the interaction region This is schematicallyillustrated in Fig 17 The real situation is perhaps in between the

above two cases ie the polymer chains may undergo some

interpenetration and some compression

Provided the dangling chains (the A chains in AndashB AndashBndashA block or

BAn graft copolymers) are in a good solvent this local increase in

segment density in the interaction zone will result in strong repulsion

as a result of two main effects (i) Increase in the osmotic pressure in

the overlap region as a result of the unfavourable mixing of the

polymer chains when these are in good solvent conditions This is

referred to as osmotic repulsion or mixing interaction and it is

described by a free energy of interaction Gmix (ii) Reduction of the

con1047297gurational entropy of the chains in the interaction zone this

entropy reduction results from the decrease in the volume available

for thechains when these areeither overlapped or compressed This is

referred to as volume restriction interaction entropic or elastic

interaction and it is described by a free energy of interaction Gel

Combination of Gmix and Gel is usually referred to as the steric

interaction free energy Gs ie

Gs = Gmix + Gel eth17THORN

The sign of Gmix depends on the solvency of the medium for the

chains If in a good solvent ie the FloryndashHuggins interaction

parameter χ is less than 05 then Gmix is positive and the mixing

interaction leads to repulsion (see below) In contrast if χN05 (ie the

chains are in a poor solvent condition) Gmix is negative and the

mixing interaction becomes attractive Gel is always positive and

hence in some cases one can produce stable dispersions in a relatively

poor solvent (enhanced steric stabilisation)

51 Mixing interaction Gmix

This results from the unfavourable mixing of the polymer chainswhen these are in a good solvent conditions This is schematically

shown in Fig 18

Consider two spherical particles with the same radius and each

containing an adsorbed polymer layer with thickness δ Before

overlap one can de1047297ne in each polymer layer a chemical potential

for the solvent μ iα and a volume fraction for the polymer in the

layerϕ2 In the overlap region (volume element dV ) the chemical

potential of the solvent is reduced to μ i β This results from the increase

in polymer segment concentration in this overlap region

In the overlap region the chemical potential of the polymer chains

is now higher than in the rest of the layer (with no overlap) This

Fig 13 In1047298uence of addition of KCl on adsorption

Fig 14 Prediction of Adsorption of diblock copolymer

Fig15 Theoretical predictions of the adsorbed layer thickness for a diblock copolymer

Fig 16 Adsorbed amount (mg mminus2) versus fraction of anchor segment for an AndashBndashA

triblock copolymer (PEOndash

PPOndash

PEO)

289T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1019

amounts to an increase in the osmotic pressure in the overlap region

as a result solvent will diffuse from thebulk to the overlap region thus

separating the particles and hence a strong repulsive energy arises

from this effect The above repulsive energy can be calculated by

considering the free energy of mixing of two polymer solutions as for

example treated by Flory and Krigbaum [22] The free energy of

mixing is given by two terms (i) An entropy term that depends on the

volume fraction of polymer and solvent (ii)An energy term that is

determined by the FloryndashHuggins interaction parameter χ

Using the above theory one can derive an expression for the free

energy of mixing of two polymer layers (assuming a uniform segment

density distribution in each layer) surrounding two spherical particles

as a function of the separation distance h between the particles The

expression for Gmix is

Gmix

kT =

2V 22V 1

m2

2

1

2minusχ

δminus

h

2

2

3R + 2δ + h

2

eth18THORN

k is the Boltzmann constant T is the absolute temperature V 2 is the

molar volume of polymer V 1 is the molar volume of solvent and ν 2 is

the number of polymer chains per unit area

The sign of Gmix depends on the value of the FloryndashHuggins

interaction parameter χ if χb05 Gmix is positive and the interaction

is repulsive if χN05 Gmix is negative and the interaction is attractiveif χ=05 Gmix =0 and this de1047297nes the θ-condition

52 Elastic interaction Gel

Thisarises fromthe lossin con1047297gurational entropy of thechains on

the approach of a second particle As a result of this approach the

volume available for the chains becomes restricted resulting in loss of

the number of con1047297gurations This can be illustrated by considering a

simple molecule represented by a rod that rotates freely in a

hemisphere across a surface (Fig 19) When the two surfaces are

separated byan in1047297nitedistanceinfin thenumber of con1047297gurations of the

rod is Ω(infin) which is proportional to the volume of the hemisphere

When a secondparticleapproaches to a distance h suchthatit cuts the

hemisphere (loosing some volume) the volume available to the chains

is reduced and the number of con1047297gurations become Ω(h) which is

less than Ω(infin) For two 1047298at plates Gel is given by the following

expression

Gel

kT = 2m2 ln

X heth THORN

X infineth THORN

= 2m2Rel heth THORN eth19THORN

where Rel

(h) is a geometric function whose form depends on the

segment density distribution It should be stressed that Gel is always

positive and could play a major role in steric stabilisation It becomes

very strong when the separation distance between the particles

becomes comparable to the adsorbed layer thickness δ

Combination of Gmix and Gel with GA (the van der Waals attractive

energy)gives the total free energy of interaction GT (assuming there is

no contribution from any residual electrostatic interaction) ie

GT = Gmix + Gel + GA eth20THORN

A schematic representation of the variation of Gmix Gel GA and GT

with surfacendashsurface separation distance h is shown in Fig 20

Gmix increases very sharply with decrease of h when hb2δ Gel

increases very sharply with decrease of h when hbδ GT versus h

shows a minimum Gmin at separation distances comparable to 2δ

When h b2δ GT shows a rapid increase with decrease in h

The depth of the minimum depends on the Hamaker constant A

the particle radius R and adsorbed layer thickness δ Gmin increases

with increase of A and R At a given A and R Gmin increases with

decrease in δ (ie with decrease of the molecular weight M w of the

stabiliser This is illustrated in Fig 21 which shows the energyndash

distance curves as a function of δR The larger the value of δR the

smaller the value of Gmin In this case the system may approach

thermodynamic stability as is the case with nano-dispersions

6 Emulsions stabilized by polymeric surfactants

The most effective method for emulsion stabilization is to use

polymeric surfactants that stronglyadsorb at the OWor WO interface

Fig 17 Schematic representation of the interaction between particles containing adsorbed polymer layers

Fig 18 Schematic representation of polymer layer overlap

Fig 19 Schematic representation of con1047297gurational entropy loss on approach of a

second particle

290 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1119

and produce effective steric stabilization against strong 1047298occulation

coalescence and Ostwald ripening [23]

As mentioned above a graft copolymer of the ABn type was

synthesized by grafting several alkyl groups on an inulin (polyfruc-

tose) chain The polymeric surfactant (INUTECregSP1) consists of a

linear polyfructose chain (the stabilizing A chain) and several alkyl

groups (the B chains) that provide multi-anchor attachment to the oil

droplets This polymeric surfactant produces enhanced steric stabili-

zation both in water and high electrolyte concentrations as will be

discussed later

For water-in-oil (WO) emulsions an AndashBndashA block copolymer of

poly (12-hydroxystearic acid) (PHS) (the A chains) and poly (ethylene

oxide) (PEO) (the B chain) PHSndashPEOndashPHS is commercially available

(Arlacel P135 UNIQEMA) The PEO chain (that is soluble in the water

droplets) forms the anchor chain whereas the PHS chains form the

stabilizing chains PHS is highly soluble in most hydrocarbon solvents

and is strongly solvated by its molecules The structure of the PHSndash

PEOndashPHS block copolymer is schematically shown in Fig 22The conformation of the polymeric surfactant at the WO interface

is schematically shown in Fig 23

Emulsions of Isopar Mwater and cyclomethiconewater were

prepared using INUTECregSP1 5050 (vv) OW emulsions were

prepared and the emulsi1047297er concentration was varied from 025 to 2

(wv) based on the oil phase 05 (wv) emulsi1047297erwas suf 1047297cient for

stabilization of these 5050 (vv) emulsions [23]

The emulsions were stored at room temperature and 50 degC and

optical micrographs were taken at intervals of time (for a year) in

order to check the stability Emulsions prepared in water were very

stableshowingno change in droplet size distributionover more than a

year period and this indicated absence of coalescence Any weak

1047298occulation that occurred was reversible and the emulsion could be

redispersed by gentle shaking Fig 24 shows an optical micrograph fora dilute 5050 (vv) emulsion that was stored for 15 and 14 weeks at

50 degC

No change in droplet size was observed after storage for more than

1 year at 50 degC indicating absence of coalescence The same result was

obtained when using different oils Emulsions were also stable against

coalescence in the presence of high electrolyte concentrations (up to

4 mol dmminus3 or ~25 NaCl

The above stability in high electrolyte concentrations is not

observed with polymeric surfactants based on polethylene oxide

The high stability observed using INUTECregSP1 is related to its

strong hydration both in water and in electrolyte solutions The

hydration of inulin (the backbone of HMI) could be assessed using

cloud point measurements A comparison was also made with PEO

with two molecular weights namely 4000 and 20000

Solutionsof PEO 4000 and 20000 showeda systematic decrease of

cloud point with increase in NaCl or MgSO4 concentration In contrast

inulin showed no cloud point up to 4 mol dmminus3 NaCl and up to 1 mol

dmminus3 MgSO4

The above results explain the difference between PEO and inulin

With PEO the chains show dehydration when the NaCl concentration

is increased above 2 mol dmminus3 or 05 mol dmminus3 MgSO4 The inulin

chains remain hydrated at much higher electrolyte concentrations It

seems that the linear polyfructose chains remain strongly hydrated athigh temperature and high electrolyte concentrations

The high emulsion stability obtained whenusingINUTECregSP1 can be

accounted for by the following factors (i) The multi-point attachment

of the polymer by several alkyl chains that are grafted on the backbone

(ii) The strong hydration of the polyfructose ldquoloopsrdquo both in water and

high electrolyte concentrations (χ remains below 05 under these

conditions) (iii) Thehigh volumefraction (concentration) of the loops at

the interface (iv) Enhanced steric stabilization this is the case with

multi-point attachment which produces strong elastic interaction

Evidence for the high stability of the liquid 1047297lm between emulsion

droplets when using INUTECregSP1 was obtained by Exerowa et al [24]

using disjoining pressure measurements This is illustrated in Fig 25

which shows a plot of disjoining pressure versus separation distance

between two emulsion droplets at various electrolyte concentrations

The results show that by increasing the capillary pressure a stable

Newton Black Film (NBF) is obtained at a 1047297lm thickness of sim7 nm

The lack of rupture of the 1047297lm at the highest pressure applied of

45times104 Pa indicate the high stability of the 1047297lm in water and in high

electrolyte concentrations (up to 20 mol dmminus3 NaCl)

The lack of rupture of the NBF up to the highest pressure applied

namely 45times104 Pa clearly indicatesthe high stability of the liquid1047297lm

in the presence of high NaCl concentrations (up to 2 mol dm minus3) This

result is consistent with the high emulsion stability obtained at high

electrolyte concentrations and high temperature Emulsions of Isopar

M-in-water are very stable under such conditions and this could be

accounted for by the high stability of the NBF The droplet size of

5050 OW emulsions prepared using 2 INUTEC regSP1 is in the region

of 1ndash10 μ m This corresponds to a capillary pressure of ~3times104 Pa for

the 1 μ m drops and ~3times103 Pa for the 10 μ m drops These capillarypressures are lower than those to which the NBF have been sub-

jected to and this clearly indicates the high stability obtained against

coalescence in these emulsions

61 WO emulsions stabilized with PHS ndashPEOndashPHS block copolymer

WO emulsions (the oil being Isopar M) were prepared using PHSndash

PEOndashPHS block copolymer at high water volume fractions (N07) The

emulsions have a narrow droplet size distribution with a z -average

radius of 183 nm [25] They also remained 1047298uid up to high water

volume fractions (N06) This could be illustrated from viscosityndash

volume fraction curves as is shown in Fig 26

Fig 20 Energyndashdistance curves for sterically stabilized systems

Fig 21 Variation of Gmin with δR

291T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1219

The effective volume fraction ϕeff of the emulsions (the core

droplets plus the adsorbed layer) could be calculated from the relative

viscosity and using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth21THORN

Where η r is the relative viscosity ϕp is the maximum packing fraction(sim07) and [ η ] is the intrinsic viscosity that is equal to 25 for hard-

spheres

The calculations based on Eq (22) are shown in Fig 27 (square

symbols) From the effective volume fraction ϕeff and the core volume

fraction ϕ the adsorbed layer thickness could be calculated This was

found to be in the region of 10 nm at ϕ =04 and it decreased with

increase in ϕ

The WO emulsions prepared using the PHSndashPEOndashPHS block

copolymer remained stable both at room temperature and 50 degC

This is consistent with the structure of the block copolymer the B

chain (PEO) is soluble in water and it forms a very strong anchor at the

WO interface The PHS chains (the A chains) provide effective steric

stabilization since the chains are highly soluble in Isopar M and are

strongly solvated by its molecules

7 Suspensions stabilised using polymeric surfactants

There are generally two procedures for preparation of solidliquid

dispersions

(i) Condensation methods build-up of particles from mole-

cular units ie nucleation and growth A special procedure

is the preparation of latexes by emulsion or dispersion

polymerization

(ii) Dispersion methods in this case one starts with preformed

large particles or crystals which are dispersed in the liquid by

using a surfactant (wetting agent) with subsequent breaking up

of the large particles by milling (comminution) to achieve the

desirable particle size distribution A dispersing agent (usuallya

polymeric surfactant) is used for the dispersion process and

subsequent stabilization of the resulting suspension

There are generally two procedures for preparation of latexes

(i) Emulsion polymerization the monomers that are essentially

insoluble in the aqueous medium are emulsi1047297ed using a

surfactant and an initiator is added while heating the system

to produce the polymer particles that are stabilized electro-

statically (when using ionic surfactants) or sterically (when

using non-ionic surfactants)

(ii) Dispersion polymerization the monomers are dissolved in a

solvent in which the resulting polymer particlesare insoluble A

protective colloid (normally a block or graft copolymer) is

added to prevent 1047298occulation of the resulting polymers

particles that are produced on addition of an initiator This

method is usually applied for the production of non-aqueous

latex dispersions and is sometimes referred to as Non-Aqueous

Dispersion Polymerization (NAD)

Surfactants play a crucial role in the process of latex preparation

since they determine the stabilizing ef 1047297ciency and the effectiveness of

the surface active agent ultimately determines the number of particles

Fig 22 Schematic representation of the structure of PHSndashPEOndashPHS block copolymer

Fig 23 Conformation of PHSndash

PEOndash

PHS polymeric surfactant at the WO interface

292 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1319

and their size The effectiveness of any surface active agent in

stabilizing the particles is the dominant factor and the number of

micelles formed is relatively unimportant In the NAD process themonomer normally an acrylic is dissolved in a non-aqueous solvent

normally an aliphatic hydrocarbon and an oil soluble initiator and a

stabilizer (to protect the resulting particles from1047298occulation) is added

to the reaction mixture The most successful stabilizers used in NAD

are block and graft copolymers Preformed graft stabilizers based

on poly(12-hydroxy stearic acid) (PHS) are simple to prepare and

effective in NAD polymerization

Dispersion methods are used for the preparation of suspensions of

preformed particles The role of surfactants (or polymers) in the

dispersion process can be analyzed in terms of the three processesinvolved (i) Wetting of the powder by the liquid (ii) Breaking of the

aggregates and agglomerates (iii) Comminution of the resulting

particles and their subsequent stabilization All these processes are

affected by surfactants or polymers which adsorb on the powder

surface thus aiding the wetting of the powder break-up of the

aggregates and agglomerates and subsequent reduction of particle

size by wet milling

71 Polymeric surfactants in emulsion polymerization

Recently the graft copolymer of hydrophobically modi1047297ed inulin

(INUTECreg SP1) has been used in emulsion polymerization of styrene

methyl methacrylate butyl acrylate and several other monomers [26]All lattices were prepared by emulsion polymerisation using potas-

sium persulphate as initiator The z -average particle size was

determined by photon correlation spectroscopy (PCS) and electron

micrographs were also taken

Fig 24 Optical micrographs of OW emulsions stabilized with INUTEC regSP1 stored at

50 degC for 15 weeks (a) and 14 weeks (b)

Fig 25 Variation of disjoining pressure with equivalent 1047297lm thickness at various NaCl concentrations

Fig 26 Viscosityndashvolume fraction for WO emulsion stabilized with PHSndashPEOndashPHS

block copolymer experimental data calculated using Eq (22)

293T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 619

The partition function is written in terms of a number of con1047297gura-

tions These were treated as connectedsequencesof segments In each

layer random mixing between segments and solvent molecules was

assumed (mean 1047297eld approximation) Each step in the random walk

was assigned a weighting factor pi

pi consists of three contributions (i) An adsorption energyχs (which

exists only for the segments that are near the surface) (ii) Con1047297gura-

tional entropy of mixing (that exists in each layer) (iii) Segment-solvent

interaction parameterχ (the FloryndashHuggins interaction parameter note

that χ=0 for an athermal solvent χ=05 for a θ-solvent)

From the weighting factors the statistical weight of any chain

conformation can be calculated for a given segment concentration

pro1047297le using a matrix formulation Fig 6 shows typical adsorption

isotherms plotted as surface coverage (in equivalent monolayers)

versus polymer volume fraction ϕ in bulk solution (ϕ was taken to

vary between 0 and 10minus3 which is the normal experimental range)

The results in Fig 6 show the effect of increasing the chain length r

and effect of solvency (athermal solvent with χ=0 and theta solvent

with χ=05) The adsorption energy χs was taken to be the same and

equal to 1kT When r = 1 θ is very small and the adsorption increases linearly

with increase of ϕ (Henrys type isotherm) On the other hand when

r = 10 the isotherm deviates much from a straight line and approaches

a Langmuirian type However when r ge20 high af 1047297nity isotherms are

obtained This implies that the 1047297rst added polymer chains are

completely adsorbed resulting in extremely low polymer concentra-

tion in solution (approaching zero) This explains the irreversibility of

adsorption of polymeric surfactants with r N100 The adsorption

isotherms with r =100 and above are typical of those observed

experimentally for most polymers that are not too polydisperse ie

showing a steep rise followed by a nearly horizontal plateau (which

only increases few percent per decade increase of ϕ) In these dilute

solutions the effect of solvency is most clearly seen with poor

solvents giving the highest adsorbed amounts In good solvents θ is

much smaller and levels off for long chains to attain an adsorption

plateau which is essentially independent of molecular weight

Another point that emerges from the SF theory is the difference in

shape between the experimental and theoretical adsorption isotherms

in thelow concentration region Theexperimentalisotherms areusuallyrounded whereas those predicted fromtheory are1047298at This is accounted

for in terms of the molecular weight distribution (polydispersity) which

is encountered with most practical polymers This effect has been

explained by CohenndashStuart et al [17] With polydisperse polymers the

larger molecular weight fractions adsorb preferentially over the smaller

ones At low polymer concentrations nearly all polymer fractions are

adsorbed leaving a small fraction of the polymer with the lowest

molecular weights in solution As the polymer concentration is

increased the higher molecular weight fractions displace the lower

ones on thesurfacewhichare nowreleasedin solution thusshiftingthe

molecular weight distribution in solution to lower values This process

continues with further increase in polymer concentration leading to

fractionation whereby the higher molecular weight fractions are

adsorbed at the expense of the lower fractions which are released to

the bulk solution However in very concentrated solutions monomers

adsorb preferentially with respect to polymers and short chains with

respect to larger ones This is due to the fact that in this region the

conformational entropy term predominates the free energy disfavour-

ing the adsorption of long chains

According to the SF theory the bound fraction p is high at low

concentration and relatively independent of molecular weight when

r N20 However with increase in surface coverage andor molecular

weight p tends to decrease indicating the formation of larger loops

and tails

The structure of the adsorbed layer is described in terms of the

segment density distribution As an illustration Fig 7 shows some

calculations using the SF theory for loops and tails with r =1000

ϕ= 10minus6 and χ=05 In this example 38 of the segments are in

trains 555 in loops and 65 in tails This theory demonstrates theimportance of tails which dominate the total distribution in the outer

region

3 Experimental techniques for studying polymeric surfactant

adsorption

As mentioned above for full characterization of polymeric sur-

factant adsorption one needs to determine three parameters (i) The

adsorbed amount Γ (mg mminus2 or mol mminus2) as a function of equilibrium

concentration C eq ie the adsorption isotherm (ii) The fraction of

segments in direct contact with the surface p (number of segments in

Fig 5 Schematic representation of a polymer molecule adsorbing on a 1047298at surface mdash

quasi-crystalline lattice with segments 1047297lling layers that are parallel to the surface

(random mixing of segments and solvent molecules in each layer is assumed)

Fig 6 Adsorption isotherms for oligomers and polymers in the dilute region based on

the SF theory Fig 7 Loop tail and total segment pro1047297le according to the SF theory

286 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 719

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 819

The microelectrophoresis technique is based on measurement of

the electrophoretic mobility u of the particles in the presence and

absence of the polymer layer From u one can calculate the zeta

potential ζ using the Huckel equation (which is applicable for small

particles and extended double layers ie κ Rbb1 where κ is the

DebyendashHuckel parameter that is related to the salt concentration

u = 23eeon η

eth15THORN

where ε is the relative permittivity of the medium and ε o is the

permittivity of free space

By measuring ζ of the particles with and without the adsorbed

polymer layer one can obtain the hydrodynamic thickness δh For

accurate measurements one should carry the measurements at various

electrolyte concentrations and extrapolate the results to the plateau

value Several automatic instruments are available for measurement of

the electrophoretic mobility Malvern zeta sizerndashCoulter Delsa sizer and

Broekahven instrument All these instruments are easy to use and the

measurement can be carried out within few minutes

4 Examples of the adsorption isotherms of nonionic polymericsurfactants

Fig 9 shows the adsorption isotherms for PEO with different

molecular weights on PS (at room temperature It can be seen that the

amount adsorbed in mgm-2 increases with increase in the polymer

molecular weight Fig 10 shows the variation of the hydrodynamic

thickness δh with molecular weight M δh shows a linear increase with

log M δh increases with n the number of segments in the chain

according to

δhasympn08 eth16THORN

Fig 11 shows the adsorption isotherms of PVA with various

molecular weights on PS latex (at 25 degC) [19] The polymers were

obtained by fractionation of a commercial sample of PVA with an

average molecular weight of 45000 The polymer also contained 12

vinyl acetate groups As with PEO the amount of adsorption increases

with increase in M The isotherms are also of the high af 1047297nity type Γ at the plateau increases linearly with M 12

The hydrodynamic thickness was determined using PCS and theresults are given below

M 67000 43000 28000 17000 8000

δhnm 255 197 140 98 33

δh seems to increase linearly with increase in the molecular weight

The effect of solvency on adsorption was investigated by increasing

the temperature (the PVA molecules are less soluble at higher

temperature) or addition of electrolyte (KCl) [20] The results are

shown in Figs 12 and 13 for M =65100 As can be seen from Fig 12

increase of temperature results in reduction of solvency of the

mediumfor the chain (due to break down of hydrogen bonds) and this

results in an increase in the amount adsorbed Addition of KCl (which

reduces the solvency of the medium for the chain) results in anincrease in adsorption (as predicted by theory)

The adsorption of block and graft copolymers is more complex

since the intimate structure of the chain determines the extent of

adsorption [18] Randomcopolymers adsorbin an intermediate way to

that of the corresponding homopolymers Blockcopolymers retain the

adsorption preference of the individual blocks The hydrophilic block

(eg PEO) the buoy (previously referred to as the A chain) extends

away from the particle surface into the bulk solution whereas the

hydrophobic anchor block (previously referred to as the B chain) (eg

PS or PPO) provides 1047297rm attachment to the surface Fig 14 shows the

theoretical prediction of diblock copolymer adsorption according to

Fig 9 Adsorption isotherms for PEO on PS

Fig 10 Hydrodynamic thickness of PEO on PS as a function of the molecular weight

Fig 11 Adsorption isotherms of PVA with different molecular weights on polystyrene

latex at 25 degC

Fig 12 In1047298uence of temperature on adsorption

288 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 919

the Scheutjens and Fleer theory The surface density σ is plotted

versus the fraction of anchor segmentsν A The adsorption depends on

the anchorbuoy composition

The amount of adsorption is higher than for homopolymers and

the adsorbed layer thickness is more extended and dense Fig 15

shows the theoretical prediction for the adsorbed layer thickness δwhich is plotted as a function of ν A For a triblock copolymer AndashBndashA

with two buoy chains (A) and one anchor chain (B) the behaviour is

similar to that of diblock copolymers This is shown in Fig16 for PEOndash

PPOndashPEO block (Pluronic)

5 Interaction between particles containing adsorbed polymeric

surfactant layers steric stabilization

When two particles each with a radius R and containing an

adsorbed polymer layer with a hydrodynamic thickness δh approach

each other to a surfacendashsurface separation distance h that is smaller

than 2 δh the polymer layers interact with each other resulting in two

main situations [21] (i) The polymer chains may overlap with each

other (ii) The polymer layer may undergo some compression In both

cases there will be an increase in the local segment density of the

polymer chains in the interaction region This is schematicallyillustrated in Fig 17 The real situation is perhaps in between the

above two cases ie the polymer chains may undergo some

interpenetration and some compression

Provided the dangling chains (the A chains in AndashB AndashBndashA block or

BAn graft copolymers) are in a good solvent this local increase in

segment density in the interaction zone will result in strong repulsion

as a result of two main effects (i) Increase in the osmotic pressure in

the overlap region as a result of the unfavourable mixing of the

polymer chains when these are in good solvent conditions This is

referred to as osmotic repulsion or mixing interaction and it is

described by a free energy of interaction Gmix (ii) Reduction of the

con1047297gurational entropy of the chains in the interaction zone this

entropy reduction results from the decrease in the volume available

for thechains when these areeither overlapped or compressed This is

referred to as volume restriction interaction entropic or elastic

interaction and it is described by a free energy of interaction Gel

Combination of Gmix and Gel is usually referred to as the steric

interaction free energy Gs ie

Gs = Gmix + Gel eth17THORN

The sign of Gmix depends on the solvency of the medium for the

chains If in a good solvent ie the FloryndashHuggins interaction

parameter χ is less than 05 then Gmix is positive and the mixing

interaction leads to repulsion (see below) In contrast if χN05 (ie the

chains are in a poor solvent condition) Gmix is negative and the

mixing interaction becomes attractive Gel is always positive and

hence in some cases one can produce stable dispersions in a relatively

poor solvent (enhanced steric stabilisation)

51 Mixing interaction Gmix

This results from the unfavourable mixing of the polymer chainswhen these are in a good solvent conditions This is schematically

shown in Fig 18

Consider two spherical particles with the same radius and each

containing an adsorbed polymer layer with thickness δ Before

overlap one can de1047297ne in each polymer layer a chemical potential

for the solvent μ iα and a volume fraction for the polymer in the

layerϕ2 In the overlap region (volume element dV ) the chemical

potential of the solvent is reduced to μ i β This results from the increase

in polymer segment concentration in this overlap region

In the overlap region the chemical potential of the polymer chains

is now higher than in the rest of the layer (with no overlap) This

Fig 13 In1047298uence of addition of KCl on adsorption

Fig 14 Prediction of Adsorption of diblock copolymer

Fig15 Theoretical predictions of the adsorbed layer thickness for a diblock copolymer

Fig 16 Adsorbed amount (mg mminus2) versus fraction of anchor segment for an AndashBndashA

triblock copolymer (PEOndash

PPOndash

PEO)

289T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1019

amounts to an increase in the osmotic pressure in the overlap region

as a result solvent will diffuse from thebulk to the overlap region thus

separating the particles and hence a strong repulsive energy arises

from this effect The above repulsive energy can be calculated by

considering the free energy of mixing of two polymer solutions as for

example treated by Flory and Krigbaum [22] The free energy of

mixing is given by two terms (i) An entropy term that depends on the

volume fraction of polymer and solvent (ii)An energy term that is

determined by the FloryndashHuggins interaction parameter χ

Using the above theory one can derive an expression for the free

energy of mixing of two polymer layers (assuming a uniform segment

density distribution in each layer) surrounding two spherical particles

as a function of the separation distance h between the particles The

expression for Gmix is

Gmix

kT =

2V 22V 1

m2

2

1

2minusχ

δminus

h

2

2

3R + 2δ + h

2

eth18THORN

k is the Boltzmann constant T is the absolute temperature V 2 is the

molar volume of polymer V 1 is the molar volume of solvent and ν 2 is

the number of polymer chains per unit area

The sign of Gmix depends on the value of the FloryndashHuggins

interaction parameter χ if χb05 Gmix is positive and the interaction

is repulsive if χN05 Gmix is negative and the interaction is attractiveif χ=05 Gmix =0 and this de1047297nes the θ-condition

52 Elastic interaction Gel

Thisarises fromthe lossin con1047297gurational entropy of thechains on

the approach of a second particle As a result of this approach the

volume available for the chains becomes restricted resulting in loss of

the number of con1047297gurations This can be illustrated by considering a

simple molecule represented by a rod that rotates freely in a

hemisphere across a surface (Fig 19) When the two surfaces are

separated byan in1047297nitedistanceinfin thenumber of con1047297gurations of the

rod is Ω(infin) which is proportional to the volume of the hemisphere

When a secondparticleapproaches to a distance h suchthatit cuts the

hemisphere (loosing some volume) the volume available to the chains

is reduced and the number of con1047297gurations become Ω(h) which is

less than Ω(infin) For two 1047298at plates Gel is given by the following

expression

Gel

kT = 2m2 ln

X heth THORN

X infineth THORN

= 2m2Rel heth THORN eth19THORN

where Rel

(h) is a geometric function whose form depends on the

segment density distribution It should be stressed that Gel is always

positive and could play a major role in steric stabilisation It becomes

very strong when the separation distance between the particles

becomes comparable to the adsorbed layer thickness δ

Combination of Gmix and Gel with GA (the van der Waals attractive

energy)gives the total free energy of interaction GT (assuming there is

no contribution from any residual electrostatic interaction) ie

GT = Gmix + Gel + GA eth20THORN

A schematic representation of the variation of Gmix Gel GA and GT

with surfacendashsurface separation distance h is shown in Fig 20

Gmix increases very sharply with decrease of h when hb2δ Gel

increases very sharply with decrease of h when hbδ GT versus h

shows a minimum Gmin at separation distances comparable to 2δ

When h b2δ GT shows a rapid increase with decrease in h

The depth of the minimum depends on the Hamaker constant A

the particle radius R and adsorbed layer thickness δ Gmin increases

with increase of A and R At a given A and R Gmin increases with

decrease in δ (ie with decrease of the molecular weight M w of the

stabiliser This is illustrated in Fig 21 which shows the energyndash

distance curves as a function of δR The larger the value of δR the

smaller the value of Gmin In this case the system may approach

thermodynamic stability as is the case with nano-dispersions

6 Emulsions stabilized by polymeric surfactants

The most effective method for emulsion stabilization is to use

polymeric surfactants that stronglyadsorb at the OWor WO interface

Fig 17 Schematic representation of the interaction between particles containing adsorbed polymer layers

Fig 18 Schematic representation of polymer layer overlap

Fig 19 Schematic representation of con1047297gurational entropy loss on approach of a

second particle

290 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1119

and produce effective steric stabilization against strong 1047298occulation

coalescence and Ostwald ripening [23]

As mentioned above a graft copolymer of the ABn type was

synthesized by grafting several alkyl groups on an inulin (polyfruc-

tose) chain The polymeric surfactant (INUTECregSP1) consists of a

linear polyfructose chain (the stabilizing A chain) and several alkyl

groups (the B chains) that provide multi-anchor attachment to the oil

droplets This polymeric surfactant produces enhanced steric stabili-

zation both in water and high electrolyte concentrations as will be

discussed later

For water-in-oil (WO) emulsions an AndashBndashA block copolymer of

poly (12-hydroxystearic acid) (PHS) (the A chains) and poly (ethylene

oxide) (PEO) (the B chain) PHSndashPEOndashPHS is commercially available

(Arlacel P135 UNIQEMA) The PEO chain (that is soluble in the water

droplets) forms the anchor chain whereas the PHS chains form the

stabilizing chains PHS is highly soluble in most hydrocarbon solvents

and is strongly solvated by its molecules The structure of the PHSndash

PEOndashPHS block copolymer is schematically shown in Fig 22The conformation of the polymeric surfactant at the WO interface

is schematically shown in Fig 23

Emulsions of Isopar Mwater and cyclomethiconewater were

prepared using INUTECregSP1 5050 (vv) OW emulsions were

prepared and the emulsi1047297er concentration was varied from 025 to 2

(wv) based on the oil phase 05 (wv) emulsi1047297erwas suf 1047297cient for

stabilization of these 5050 (vv) emulsions [23]

The emulsions were stored at room temperature and 50 degC and

optical micrographs were taken at intervals of time (for a year) in

order to check the stability Emulsions prepared in water were very

stableshowingno change in droplet size distributionover more than a

year period and this indicated absence of coalescence Any weak

1047298occulation that occurred was reversible and the emulsion could be

redispersed by gentle shaking Fig 24 shows an optical micrograph fora dilute 5050 (vv) emulsion that was stored for 15 and 14 weeks at

50 degC

No change in droplet size was observed after storage for more than

1 year at 50 degC indicating absence of coalescence The same result was

obtained when using different oils Emulsions were also stable against

coalescence in the presence of high electrolyte concentrations (up to

4 mol dmminus3 or ~25 NaCl

The above stability in high electrolyte concentrations is not

observed with polymeric surfactants based on polethylene oxide

The high stability observed using INUTECregSP1 is related to its

strong hydration both in water and in electrolyte solutions The

hydration of inulin (the backbone of HMI) could be assessed using

cloud point measurements A comparison was also made with PEO

with two molecular weights namely 4000 and 20000

Solutionsof PEO 4000 and 20000 showeda systematic decrease of

cloud point with increase in NaCl or MgSO4 concentration In contrast

inulin showed no cloud point up to 4 mol dmminus3 NaCl and up to 1 mol

dmminus3 MgSO4

The above results explain the difference between PEO and inulin

With PEO the chains show dehydration when the NaCl concentration

is increased above 2 mol dmminus3 or 05 mol dmminus3 MgSO4 The inulin

chains remain hydrated at much higher electrolyte concentrations It

seems that the linear polyfructose chains remain strongly hydrated athigh temperature and high electrolyte concentrations

The high emulsion stability obtained whenusingINUTECregSP1 can be

accounted for by the following factors (i) The multi-point attachment

of the polymer by several alkyl chains that are grafted on the backbone

(ii) The strong hydration of the polyfructose ldquoloopsrdquo both in water and

high electrolyte concentrations (χ remains below 05 under these

conditions) (iii) Thehigh volumefraction (concentration) of the loops at

the interface (iv) Enhanced steric stabilization this is the case with

multi-point attachment which produces strong elastic interaction

Evidence for the high stability of the liquid 1047297lm between emulsion

droplets when using INUTECregSP1 was obtained by Exerowa et al [24]

using disjoining pressure measurements This is illustrated in Fig 25

which shows a plot of disjoining pressure versus separation distance

between two emulsion droplets at various electrolyte concentrations

The results show that by increasing the capillary pressure a stable

Newton Black Film (NBF) is obtained at a 1047297lm thickness of sim7 nm

The lack of rupture of the 1047297lm at the highest pressure applied of

45times104 Pa indicate the high stability of the 1047297lm in water and in high

electrolyte concentrations (up to 20 mol dmminus3 NaCl)

The lack of rupture of the NBF up to the highest pressure applied

namely 45times104 Pa clearly indicatesthe high stability of the liquid1047297lm

in the presence of high NaCl concentrations (up to 2 mol dm minus3) This

result is consistent with the high emulsion stability obtained at high

electrolyte concentrations and high temperature Emulsions of Isopar

M-in-water are very stable under such conditions and this could be

accounted for by the high stability of the NBF The droplet size of

5050 OW emulsions prepared using 2 INUTEC regSP1 is in the region

of 1ndash10 μ m This corresponds to a capillary pressure of ~3times104 Pa for

the 1 μ m drops and ~3times103 Pa for the 10 μ m drops These capillarypressures are lower than those to which the NBF have been sub-

jected to and this clearly indicates the high stability obtained against

coalescence in these emulsions

61 WO emulsions stabilized with PHS ndashPEOndashPHS block copolymer

WO emulsions (the oil being Isopar M) were prepared using PHSndash

PEOndashPHS block copolymer at high water volume fractions (N07) The

emulsions have a narrow droplet size distribution with a z -average

radius of 183 nm [25] They also remained 1047298uid up to high water

volume fractions (N06) This could be illustrated from viscosityndash

volume fraction curves as is shown in Fig 26

Fig 20 Energyndashdistance curves for sterically stabilized systems

Fig 21 Variation of Gmin with δR

291T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1219

The effective volume fraction ϕeff of the emulsions (the core

droplets plus the adsorbed layer) could be calculated from the relative

viscosity and using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth21THORN

Where η r is the relative viscosity ϕp is the maximum packing fraction(sim07) and [ η ] is the intrinsic viscosity that is equal to 25 for hard-

spheres

The calculations based on Eq (22) are shown in Fig 27 (square

symbols) From the effective volume fraction ϕeff and the core volume

fraction ϕ the adsorbed layer thickness could be calculated This was

found to be in the region of 10 nm at ϕ =04 and it decreased with

increase in ϕ

The WO emulsions prepared using the PHSndashPEOndashPHS block

copolymer remained stable both at room temperature and 50 degC

This is consistent with the structure of the block copolymer the B

chain (PEO) is soluble in water and it forms a very strong anchor at the

WO interface The PHS chains (the A chains) provide effective steric

stabilization since the chains are highly soluble in Isopar M and are

strongly solvated by its molecules

7 Suspensions stabilised using polymeric surfactants

There are generally two procedures for preparation of solidliquid

dispersions

(i) Condensation methods build-up of particles from mole-

cular units ie nucleation and growth A special procedure

is the preparation of latexes by emulsion or dispersion

polymerization

(ii) Dispersion methods in this case one starts with preformed

large particles or crystals which are dispersed in the liquid by

using a surfactant (wetting agent) with subsequent breaking up

of the large particles by milling (comminution) to achieve the

desirable particle size distribution A dispersing agent (usuallya

polymeric surfactant) is used for the dispersion process and

subsequent stabilization of the resulting suspension

There are generally two procedures for preparation of latexes

(i) Emulsion polymerization the monomers that are essentially

insoluble in the aqueous medium are emulsi1047297ed using a

surfactant and an initiator is added while heating the system

to produce the polymer particles that are stabilized electro-

statically (when using ionic surfactants) or sterically (when

using non-ionic surfactants)

(ii) Dispersion polymerization the monomers are dissolved in a

solvent in which the resulting polymer particlesare insoluble A

protective colloid (normally a block or graft copolymer) is

added to prevent 1047298occulation of the resulting polymers

particles that are produced on addition of an initiator This

method is usually applied for the production of non-aqueous

latex dispersions and is sometimes referred to as Non-Aqueous

Dispersion Polymerization (NAD)

Surfactants play a crucial role in the process of latex preparation

since they determine the stabilizing ef 1047297ciency and the effectiveness of

the surface active agent ultimately determines the number of particles

Fig 22 Schematic representation of the structure of PHSndashPEOndashPHS block copolymer

Fig 23 Conformation of PHSndash

PEOndash

PHS polymeric surfactant at the WO interface

292 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1319

and their size The effectiveness of any surface active agent in

stabilizing the particles is the dominant factor and the number of

micelles formed is relatively unimportant In the NAD process themonomer normally an acrylic is dissolved in a non-aqueous solvent

normally an aliphatic hydrocarbon and an oil soluble initiator and a

stabilizer (to protect the resulting particles from1047298occulation) is added

to the reaction mixture The most successful stabilizers used in NAD

are block and graft copolymers Preformed graft stabilizers based

on poly(12-hydroxy stearic acid) (PHS) are simple to prepare and

effective in NAD polymerization

Dispersion methods are used for the preparation of suspensions of

preformed particles The role of surfactants (or polymers) in the

dispersion process can be analyzed in terms of the three processesinvolved (i) Wetting of the powder by the liquid (ii) Breaking of the

aggregates and agglomerates (iii) Comminution of the resulting

particles and their subsequent stabilization All these processes are

affected by surfactants or polymers which adsorb on the powder

surface thus aiding the wetting of the powder break-up of the

aggregates and agglomerates and subsequent reduction of particle

size by wet milling

71 Polymeric surfactants in emulsion polymerization

Recently the graft copolymer of hydrophobically modi1047297ed inulin

(INUTECreg SP1) has been used in emulsion polymerization of styrene

methyl methacrylate butyl acrylate and several other monomers [26]All lattices were prepared by emulsion polymerisation using potas-

sium persulphate as initiator The z -average particle size was

determined by photon correlation spectroscopy (PCS) and electron

micrographs were also taken

Fig 24 Optical micrographs of OW emulsions stabilized with INUTEC regSP1 stored at

50 degC for 15 weeks (a) and 14 weeks (b)

Fig 25 Variation of disjoining pressure with equivalent 1047297lm thickness at various NaCl concentrations

Fig 26 Viscosityndashvolume fraction for WO emulsion stabilized with PHSndashPEOndashPHS

block copolymer experimental data calculated using Eq (22)

293T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 719

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 819

The microelectrophoresis technique is based on measurement of

the electrophoretic mobility u of the particles in the presence and

absence of the polymer layer From u one can calculate the zeta

potential ζ using the Huckel equation (which is applicable for small

particles and extended double layers ie κ Rbb1 where κ is the

DebyendashHuckel parameter that is related to the salt concentration

u = 23eeon η

eth15THORN

where ε is the relative permittivity of the medium and ε o is the

permittivity of free space

By measuring ζ of the particles with and without the adsorbed

polymer layer one can obtain the hydrodynamic thickness δh For

accurate measurements one should carry the measurements at various

electrolyte concentrations and extrapolate the results to the plateau

value Several automatic instruments are available for measurement of

the electrophoretic mobility Malvern zeta sizerndashCoulter Delsa sizer and

Broekahven instrument All these instruments are easy to use and the

measurement can be carried out within few minutes

4 Examples of the adsorption isotherms of nonionic polymericsurfactants

Fig 9 shows the adsorption isotherms for PEO with different

molecular weights on PS (at room temperature It can be seen that the

amount adsorbed in mgm-2 increases with increase in the polymer

molecular weight Fig 10 shows the variation of the hydrodynamic

thickness δh with molecular weight M δh shows a linear increase with

log M δh increases with n the number of segments in the chain

according to

δhasympn08 eth16THORN

Fig 11 shows the adsorption isotherms of PVA with various

molecular weights on PS latex (at 25 degC) [19] The polymers were

obtained by fractionation of a commercial sample of PVA with an

average molecular weight of 45000 The polymer also contained 12

vinyl acetate groups As with PEO the amount of adsorption increases

with increase in M The isotherms are also of the high af 1047297nity type Γ at the plateau increases linearly with M 12

The hydrodynamic thickness was determined using PCS and theresults are given below

M 67000 43000 28000 17000 8000

δhnm 255 197 140 98 33

δh seems to increase linearly with increase in the molecular weight

The effect of solvency on adsorption was investigated by increasing

the temperature (the PVA molecules are less soluble at higher

temperature) or addition of electrolyte (KCl) [20] The results are

shown in Figs 12 and 13 for M =65100 As can be seen from Fig 12

increase of temperature results in reduction of solvency of the

mediumfor the chain (due to break down of hydrogen bonds) and this

results in an increase in the amount adsorbed Addition of KCl (which

reduces the solvency of the medium for the chain) results in anincrease in adsorption (as predicted by theory)

The adsorption of block and graft copolymers is more complex

since the intimate structure of the chain determines the extent of

adsorption [18] Randomcopolymers adsorbin an intermediate way to

that of the corresponding homopolymers Blockcopolymers retain the

adsorption preference of the individual blocks The hydrophilic block

(eg PEO) the buoy (previously referred to as the A chain) extends

away from the particle surface into the bulk solution whereas the

hydrophobic anchor block (previously referred to as the B chain) (eg

PS or PPO) provides 1047297rm attachment to the surface Fig 14 shows the

theoretical prediction of diblock copolymer adsorption according to

Fig 9 Adsorption isotherms for PEO on PS

Fig 10 Hydrodynamic thickness of PEO on PS as a function of the molecular weight

Fig 11 Adsorption isotherms of PVA with different molecular weights on polystyrene

latex at 25 degC

Fig 12 In1047298uence of temperature on adsorption

288 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 919

the Scheutjens and Fleer theory The surface density σ is plotted

versus the fraction of anchor segmentsν A The adsorption depends on

the anchorbuoy composition

The amount of adsorption is higher than for homopolymers and

the adsorbed layer thickness is more extended and dense Fig 15

shows the theoretical prediction for the adsorbed layer thickness δwhich is plotted as a function of ν A For a triblock copolymer AndashBndashA

with two buoy chains (A) and one anchor chain (B) the behaviour is

similar to that of diblock copolymers This is shown in Fig16 for PEOndash

PPOndashPEO block (Pluronic)

5 Interaction between particles containing adsorbed polymeric

surfactant layers steric stabilization

When two particles each with a radius R and containing an

adsorbed polymer layer with a hydrodynamic thickness δh approach

each other to a surfacendashsurface separation distance h that is smaller

than 2 δh the polymer layers interact with each other resulting in two

main situations [21] (i) The polymer chains may overlap with each

other (ii) The polymer layer may undergo some compression In both

cases there will be an increase in the local segment density of the

polymer chains in the interaction region This is schematicallyillustrated in Fig 17 The real situation is perhaps in between the

above two cases ie the polymer chains may undergo some

interpenetration and some compression

Provided the dangling chains (the A chains in AndashB AndashBndashA block or

BAn graft copolymers) are in a good solvent this local increase in

segment density in the interaction zone will result in strong repulsion

as a result of two main effects (i) Increase in the osmotic pressure in

the overlap region as a result of the unfavourable mixing of the

polymer chains when these are in good solvent conditions This is

referred to as osmotic repulsion or mixing interaction and it is

described by a free energy of interaction Gmix (ii) Reduction of the

con1047297gurational entropy of the chains in the interaction zone this

entropy reduction results from the decrease in the volume available

for thechains when these areeither overlapped or compressed This is

referred to as volume restriction interaction entropic or elastic

interaction and it is described by a free energy of interaction Gel

Combination of Gmix and Gel is usually referred to as the steric

interaction free energy Gs ie

Gs = Gmix + Gel eth17THORN

The sign of Gmix depends on the solvency of the medium for the

chains If in a good solvent ie the FloryndashHuggins interaction

parameter χ is less than 05 then Gmix is positive and the mixing

interaction leads to repulsion (see below) In contrast if χN05 (ie the

chains are in a poor solvent condition) Gmix is negative and the

mixing interaction becomes attractive Gel is always positive and

hence in some cases one can produce stable dispersions in a relatively

poor solvent (enhanced steric stabilisation)

51 Mixing interaction Gmix

This results from the unfavourable mixing of the polymer chainswhen these are in a good solvent conditions This is schematically

shown in Fig 18

Consider two spherical particles with the same radius and each

containing an adsorbed polymer layer with thickness δ Before

overlap one can de1047297ne in each polymer layer a chemical potential

for the solvent μ iα and a volume fraction for the polymer in the

layerϕ2 In the overlap region (volume element dV ) the chemical

potential of the solvent is reduced to μ i β This results from the increase

in polymer segment concentration in this overlap region

In the overlap region the chemical potential of the polymer chains

is now higher than in the rest of the layer (with no overlap) This

Fig 13 In1047298uence of addition of KCl on adsorption

Fig 14 Prediction of Adsorption of diblock copolymer

Fig15 Theoretical predictions of the adsorbed layer thickness for a diblock copolymer

Fig 16 Adsorbed amount (mg mminus2) versus fraction of anchor segment for an AndashBndashA

triblock copolymer (PEOndash

PPOndash

PEO)

289T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1019

amounts to an increase in the osmotic pressure in the overlap region

as a result solvent will diffuse from thebulk to the overlap region thus

separating the particles and hence a strong repulsive energy arises

from this effect The above repulsive energy can be calculated by

considering the free energy of mixing of two polymer solutions as for

example treated by Flory and Krigbaum [22] The free energy of

mixing is given by two terms (i) An entropy term that depends on the

volume fraction of polymer and solvent (ii)An energy term that is

determined by the FloryndashHuggins interaction parameter χ

Using the above theory one can derive an expression for the free

energy of mixing of two polymer layers (assuming a uniform segment

density distribution in each layer) surrounding two spherical particles

as a function of the separation distance h between the particles The

expression for Gmix is

Gmix

kT =

2V 22V 1

m2

2

1

2minusχ

δminus

h

2

2

3R + 2δ + h

2

eth18THORN

k is the Boltzmann constant T is the absolute temperature V 2 is the

molar volume of polymer V 1 is the molar volume of solvent and ν 2 is

the number of polymer chains per unit area

The sign of Gmix depends on the value of the FloryndashHuggins

interaction parameter χ if χb05 Gmix is positive and the interaction

is repulsive if χN05 Gmix is negative and the interaction is attractiveif χ=05 Gmix =0 and this de1047297nes the θ-condition

52 Elastic interaction Gel

Thisarises fromthe lossin con1047297gurational entropy of thechains on

the approach of a second particle As a result of this approach the

volume available for the chains becomes restricted resulting in loss of

the number of con1047297gurations This can be illustrated by considering a

simple molecule represented by a rod that rotates freely in a

hemisphere across a surface (Fig 19) When the two surfaces are

separated byan in1047297nitedistanceinfin thenumber of con1047297gurations of the

rod is Ω(infin) which is proportional to the volume of the hemisphere

When a secondparticleapproaches to a distance h suchthatit cuts the

hemisphere (loosing some volume) the volume available to the chains

is reduced and the number of con1047297gurations become Ω(h) which is

less than Ω(infin) For two 1047298at plates Gel is given by the following

expression

Gel

kT = 2m2 ln

X heth THORN

X infineth THORN

= 2m2Rel heth THORN eth19THORN

where Rel

(h) is a geometric function whose form depends on the

segment density distribution It should be stressed that Gel is always

positive and could play a major role in steric stabilisation It becomes

very strong when the separation distance between the particles

becomes comparable to the adsorbed layer thickness δ

Combination of Gmix and Gel with GA (the van der Waals attractive

energy)gives the total free energy of interaction GT (assuming there is

no contribution from any residual electrostatic interaction) ie

GT = Gmix + Gel + GA eth20THORN

A schematic representation of the variation of Gmix Gel GA and GT

with surfacendashsurface separation distance h is shown in Fig 20

Gmix increases very sharply with decrease of h when hb2δ Gel

increases very sharply with decrease of h when hbδ GT versus h

shows a minimum Gmin at separation distances comparable to 2δ

When h b2δ GT shows a rapid increase with decrease in h

The depth of the minimum depends on the Hamaker constant A

the particle radius R and adsorbed layer thickness δ Gmin increases

with increase of A and R At a given A and R Gmin increases with

decrease in δ (ie with decrease of the molecular weight M w of the

stabiliser This is illustrated in Fig 21 which shows the energyndash

distance curves as a function of δR The larger the value of δR the

smaller the value of Gmin In this case the system may approach

thermodynamic stability as is the case with nano-dispersions

6 Emulsions stabilized by polymeric surfactants

The most effective method for emulsion stabilization is to use

polymeric surfactants that stronglyadsorb at the OWor WO interface

Fig 17 Schematic representation of the interaction between particles containing adsorbed polymer layers

Fig 18 Schematic representation of polymer layer overlap

Fig 19 Schematic representation of con1047297gurational entropy loss on approach of a

second particle

290 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1119

and produce effective steric stabilization against strong 1047298occulation

coalescence and Ostwald ripening [23]

As mentioned above a graft copolymer of the ABn type was

synthesized by grafting several alkyl groups on an inulin (polyfruc-

tose) chain The polymeric surfactant (INUTECregSP1) consists of a

linear polyfructose chain (the stabilizing A chain) and several alkyl

groups (the B chains) that provide multi-anchor attachment to the oil

droplets This polymeric surfactant produces enhanced steric stabili-

zation both in water and high electrolyte concentrations as will be

discussed later

For water-in-oil (WO) emulsions an AndashBndashA block copolymer of

poly (12-hydroxystearic acid) (PHS) (the A chains) and poly (ethylene

oxide) (PEO) (the B chain) PHSndashPEOndashPHS is commercially available

(Arlacel P135 UNIQEMA) The PEO chain (that is soluble in the water

droplets) forms the anchor chain whereas the PHS chains form the

stabilizing chains PHS is highly soluble in most hydrocarbon solvents

and is strongly solvated by its molecules The structure of the PHSndash

PEOndashPHS block copolymer is schematically shown in Fig 22The conformation of the polymeric surfactant at the WO interface

is schematically shown in Fig 23

Emulsions of Isopar Mwater and cyclomethiconewater were

prepared using INUTECregSP1 5050 (vv) OW emulsions were

prepared and the emulsi1047297er concentration was varied from 025 to 2

(wv) based on the oil phase 05 (wv) emulsi1047297erwas suf 1047297cient for

stabilization of these 5050 (vv) emulsions [23]

The emulsions were stored at room temperature and 50 degC and

optical micrographs were taken at intervals of time (for a year) in

order to check the stability Emulsions prepared in water were very

stableshowingno change in droplet size distributionover more than a

year period and this indicated absence of coalescence Any weak

1047298occulation that occurred was reversible and the emulsion could be

redispersed by gentle shaking Fig 24 shows an optical micrograph fora dilute 5050 (vv) emulsion that was stored for 15 and 14 weeks at

50 degC

No change in droplet size was observed after storage for more than

1 year at 50 degC indicating absence of coalescence The same result was

obtained when using different oils Emulsions were also stable against

coalescence in the presence of high electrolyte concentrations (up to

4 mol dmminus3 or ~25 NaCl

The above stability in high electrolyte concentrations is not

observed with polymeric surfactants based on polethylene oxide

The high stability observed using INUTECregSP1 is related to its

strong hydration both in water and in electrolyte solutions The

hydration of inulin (the backbone of HMI) could be assessed using

cloud point measurements A comparison was also made with PEO

with two molecular weights namely 4000 and 20000

Solutionsof PEO 4000 and 20000 showeda systematic decrease of

cloud point with increase in NaCl or MgSO4 concentration In contrast

inulin showed no cloud point up to 4 mol dmminus3 NaCl and up to 1 mol

dmminus3 MgSO4

The above results explain the difference between PEO and inulin

With PEO the chains show dehydration when the NaCl concentration

is increased above 2 mol dmminus3 or 05 mol dmminus3 MgSO4 The inulin

chains remain hydrated at much higher electrolyte concentrations It

seems that the linear polyfructose chains remain strongly hydrated athigh temperature and high electrolyte concentrations

The high emulsion stability obtained whenusingINUTECregSP1 can be

accounted for by the following factors (i) The multi-point attachment

of the polymer by several alkyl chains that are grafted on the backbone

(ii) The strong hydration of the polyfructose ldquoloopsrdquo both in water and

high electrolyte concentrations (χ remains below 05 under these

conditions) (iii) Thehigh volumefraction (concentration) of the loops at

the interface (iv) Enhanced steric stabilization this is the case with

multi-point attachment which produces strong elastic interaction

Evidence for the high stability of the liquid 1047297lm between emulsion

droplets when using INUTECregSP1 was obtained by Exerowa et al [24]

using disjoining pressure measurements This is illustrated in Fig 25

which shows a plot of disjoining pressure versus separation distance

between two emulsion droplets at various electrolyte concentrations

The results show that by increasing the capillary pressure a stable

Newton Black Film (NBF) is obtained at a 1047297lm thickness of sim7 nm

The lack of rupture of the 1047297lm at the highest pressure applied of

45times104 Pa indicate the high stability of the 1047297lm in water and in high

electrolyte concentrations (up to 20 mol dmminus3 NaCl)

The lack of rupture of the NBF up to the highest pressure applied

namely 45times104 Pa clearly indicatesthe high stability of the liquid1047297lm

in the presence of high NaCl concentrations (up to 2 mol dm minus3) This

result is consistent with the high emulsion stability obtained at high

electrolyte concentrations and high temperature Emulsions of Isopar

M-in-water are very stable under such conditions and this could be

accounted for by the high stability of the NBF The droplet size of

5050 OW emulsions prepared using 2 INUTEC regSP1 is in the region

of 1ndash10 μ m This corresponds to a capillary pressure of ~3times104 Pa for

the 1 μ m drops and ~3times103 Pa for the 10 μ m drops These capillarypressures are lower than those to which the NBF have been sub-

jected to and this clearly indicates the high stability obtained against

coalescence in these emulsions

61 WO emulsions stabilized with PHS ndashPEOndashPHS block copolymer

WO emulsions (the oil being Isopar M) were prepared using PHSndash

PEOndashPHS block copolymer at high water volume fractions (N07) The

emulsions have a narrow droplet size distribution with a z -average

radius of 183 nm [25] They also remained 1047298uid up to high water

volume fractions (N06) This could be illustrated from viscosityndash

volume fraction curves as is shown in Fig 26

Fig 20 Energyndashdistance curves for sterically stabilized systems

Fig 21 Variation of Gmin with δR

291T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1219

The effective volume fraction ϕeff of the emulsions (the core

droplets plus the adsorbed layer) could be calculated from the relative

viscosity and using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth21THORN

Where η r is the relative viscosity ϕp is the maximum packing fraction(sim07) and [ η ] is the intrinsic viscosity that is equal to 25 for hard-

spheres

The calculations based on Eq (22) are shown in Fig 27 (square

symbols) From the effective volume fraction ϕeff and the core volume

fraction ϕ the adsorbed layer thickness could be calculated This was

found to be in the region of 10 nm at ϕ =04 and it decreased with

increase in ϕ

The WO emulsions prepared using the PHSndashPEOndashPHS block

copolymer remained stable both at room temperature and 50 degC

This is consistent with the structure of the block copolymer the B

chain (PEO) is soluble in water and it forms a very strong anchor at the

WO interface The PHS chains (the A chains) provide effective steric

stabilization since the chains are highly soluble in Isopar M and are

strongly solvated by its molecules

7 Suspensions stabilised using polymeric surfactants

There are generally two procedures for preparation of solidliquid

dispersions

(i) Condensation methods build-up of particles from mole-

cular units ie nucleation and growth A special procedure

is the preparation of latexes by emulsion or dispersion

polymerization

(ii) Dispersion methods in this case one starts with preformed

large particles or crystals which are dispersed in the liquid by

using a surfactant (wetting agent) with subsequent breaking up

of the large particles by milling (comminution) to achieve the

desirable particle size distribution A dispersing agent (usuallya

polymeric surfactant) is used for the dispersion process and

subsequent stabilization of the resulting suspension

There are generally two procedures for preparation of latexes

(i) Emulsion polymerization the monomers that are essentially

insoluble in the aqueous medium are emulsi1047297ed using a

surfactant and an initiator is added while heating the system

to produce the polymer particles that are stabilized electro-

statically (when using ionic surfactants) or sterically (when

using non-ionic surfactants)

(ii) Dispersion polymerization the monomers are dissolved in a

solvent in which the resulting polymer particlesare insoluble A

protective colloid (normally a block or graft copolymer) is

added to prevent 1047298occulation of the resulting polymers

particles that are produced on addition of an initiator This

method is usually applied for the production of non-aqueous

latex dispersions and is sometimes referred to as Non-Aqueous

Dispersion Polymerization (NAD)

Surfactants play a crucial role in the process of latex preparation

since they determine the stabilizing ef 1047297ciency and the effectiveness of

the surface active agent ultimately determines the number of particles

Fig 22 Schematic representation of the structure of PHSndashPEOndashPHS block copolymer

Fig 23 Conformation of PHSndash

PEOndash

PHS polymeric surfactant at the WO interface

292 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1319

and their size The effectiveness of any surface active agent in

stabilizing the particles is the dominant factor and the number of

micelles formed is relatively unimportant In the NAD process themonomer normally an acrylic is dissolved in a non-aqueous solvent

normally an aliphatic hydrocarbon and an oil soluble initiator and a

stabilizer (to protect the resulting particles from1047298occulation) is added

to the reaction mixture The most successful stabilizers used in NAD

are block and graft copolymers Preformed graft stabilizers based

on poly(12-hydroxy stearic acid) (PHS) are simple to prepare and

effective in NAD polymerization

Dispersion methods are used for the preparation of suspensions of

preformed particles The role of surfactants (or polymers) in the

dispersion process can be analyzed in terms of the three processesinvolved (i) Wetting of the powder by the liquid (ii) Breaking of the

aggregates and agglomerates (iii) Comminution of the resulting

particles and their subsequent stabilization All these processes are

affected by surfactants or polymers which adsorb on the powder

surface thus aiding the wetting of the powder break-up of the

aggregates and agglomerates and subsequent reduction of particle

size by wet milling

71 Polymeric surfactants in emulsion polymerization

Recently the graft copolymer of hydrophobically modi1047297ed inulin

(INUTECreg SP1) has been used in emulsion polymerization of styrene

methyl methacrylate butyl acrylate and several other monomers [26]All lattices were prepared by emulsion polymerisation using potas-

sium persulphate as initiator The z -average particle size was

determined by photon correlation spectroscopy (PCS) and electron

micrographs were also taken

Fig 24 Optical micrographs of OW emulsions stabilized with INUTEC regSP1 stored at

50 degC for 15 weeks (a) and 14 weeks (b)

Fig 25 Variation of disjoining pressure with equivalent 1047297lm thickness at various NaCl concentrations

Fig 26 Viscosityndashvolume fraction for WO emulsion stabilized with PHSndashPEOndashPHS

block copolymer experimental data calculated using Eq (22)

293T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 819

The microelectrophoresis technique is based on measurement of

the electrophoretic mobility u of the particles in the presence and

absence of the polymer layer From u one can calculate the zeta

potential ζ using the Huckel equation (which is applicable for small

particles and extended double layers ie κ Rbb1 where κ is the

DebyendashHuckel parameter that is related to the salt concentration

u = 23eeon η

eth15THORN

where ε is the relative permittivity of the medium and ε o is the

permittivity of free space

By measuring ζ of the particles with and without the adsorbed

polymer layer one can obtain the hydrodynamic thickness δh For

accurate measurements one should carry the measurements at various

electrolyte concentrations and extrapolate the results to the plateau

value Several automatic instruments are available for measurement of

the electrophoretic mobility Malvern zeta sizerndashCoulter Delsa sizer and

Broekahven instrument All these instruments are easy to use and the

measurement can be carried out within few minutes

4 Examples of the adsorption isotherms of nonionic polymericsurfactants

Fig 9 shows the adsorption isotherms for PEO with different

molecular weights on PS (at room temperature It can be seen that the

amount adsorbed in mgm-2 increases with increase in the polymer

molecular weight Fig 10 shows the variation of the hydrodynamic

thickness δh with molecular weight M δh shows a linear increase with

log M δh increases with n the number of segments in the chain

according to

δhasympn08 eth16THORN

Fig 11 shows the adsorption isotherms of PVA with various

molecular weights on PS latex (at 25 degC) [19] The polymers were

obtained by fractionation of a commercial sample of PVA with an

average molecular weight of 45000 The polymer also contained 12

vinyl acetate groups As with PEO the amount of adsorption increases

with increase in M The isotherms are also of the high af 1047297nity type Γ at the plateau increases linearly with M 12

The hydrodynamic thickness was determined using PCS and theresults are given below

M 67000 43000 28000 17000 8000

δhnm 255 197 140 98 33

δh seems to increase linearly with increase in the molecular weight

The effect of solvency on adsorption was investigated by increasing

the temperature (the PVA molecules are less soluble at higher

temperature) or addition of electrolyte (KCl) [20] The results are

shown in Figs 12 and 13 for M =65100 As can be seen from Fig 12

increase of temperature results in reduction of solvency of the

mediumfor the chain (due to break down of hydrogen bonds) and this

results in an increase in the amount adsorbed Addition of KCl (which

reduces the solvency of the medium for the chain) results in anincrease in adsorption (as predicted by theory)

The adsorption of block and graft copolymers is more complex

since the intimate structure of the chain determines the extent of

adsorption [18] Randomcopolymers adsorbin an intermediate way to

that of the corresponding homopolymers Blockcopolymers retain the

adsorption preference of the individual blocks The hydrophilic block

(eg PEO) the buoy (previously referred to as the A chain) extends

away from the particle surface into the bulk solution whereas the

hydrophobic anchor block (previously referred to as the B chain) (eg

PS or PPO) provides 1047297rm attachment to the surface Fig 14 shows the

theoretical prediction of diblock copolymer adsorption according to

Fig 9 Adsorption isotherms for PEO on PS

Fig 10 Hydrodynamic thickness of PEO on PS as a function of the molecular weight

Fig 11 Adsorption isotherms of PVA with different molecular weights on polystyrene

latex at 25 degC

Fig 12 In1047298uence of temperature on adsorption

288 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 919

the Scheutjens and Fleer theory The surface density σ is plotted

versus the fraction of anchor segmentsν A The adsorption depends on

the anchorbuoy composition

The amount of adsorption is higher than for homopolymers and

the adsorbed layer thickness is more extended and dense Fig 15

shows the theoretical prediction for the adsorbed layer thickness δwhich is plotted as a function of ν A For a triblock copolymer AndashBndashA

with two buoy chains (A) and one anchor chain (B) the behaviour is

similar to that of diblock copolymers This is shown in Fig16 for PEOndash

PPOndashPEO block (Pluronic)

5 Interaction between particles containing adsorbed polymeric

surfactant layers steric stabilization

When two particles each with a radius R and containing an

adsorbed polymer layer with a hydrodynamic thickness δh approach

each other to a surfacendashsurface separation distance h that is smaller

than 2 δh the polymer layers interact with each other resulting in two

main situations [21] (i) The polymer chains may overlap with each

other (ii) The polymer layer may undergo some compression In both

cases there will be an increase in the local segment density of the

polymer chains in the interaction region This is schematicallyillustrated in Fig 17 The real situation is perhaps in between the

above two cases ie the polymer chains may undergo some

interpenetration and some compression

Provided the dangling chains (the A chains in AndashB AndashBndashA block or

BAn graft copolymers) are in a good solvent this local increase in

segment density in the interaction zone will result in strong repulsion

as a result of two main effects (i) Increase in the osmotic pressure in

the overlap region as a result of the unfavourable mixing of the

polymer chains when these are in good solvent conditions This is

referred to as osmotic repulsion or mixing interaction and it is

described by a free energy of interaction Gmix (ii) Reduction of the

con1047297gurational entropy of the chains in the interaction zone this

entropy reduction results from the decrease in the volume available

for thechains when these areeither overlapped or compressed This is

referred to as volume restriction interaction entropic or elastic

interaction and it is described by a free energy of interaction Gel

Combination of Gmix and Gel is usually referred to as the steric

interaction free energy Gs ie

Gs = Gmix + Gel eth17THORN

The sign of Gmix depends on the solvency of the medium for the

chains If in a good solvent ie the FloryndashHuggins interaction

parameter χ is less than 05 then Gmix is positive and the mixing

interaction leads to repulsion (see below) In contrast if χN05 (ie the

chains are in a poor solvent condition) Gmix is negative and the

mixing interaction becomes attractive Gel is always positive and

hence in some cases one can produce stable dispersions in a relatively

poor solvent (enhanced steric stabilisation)

51 Mixing interaction Gmix

This results from the unfavourable mixing of the polymer chainswhen these are in a good solvent conditions This is schematically

shown in Fig 18

Consider two spherical particles with the same radius and each

containing an adsorbed polymer layer with thickness δ Before

overlap one can de1047297ne in each polymer layer a chemical potential

for the solvent μ iα and a volume fraction for the polymer in the

layerϕ2 In the overlap region (volume element dV ) the chemical

potential of the solvent is reduced to μ i β This results from the increase

in polymer segment concentration in this overlap region

In the overlap region the chemical potential of the polymer chains

is now higher than in the rest of the layer (with no overlap) This

Fig 13 In1047298uence of addition of KCl on adsorption

Fig 14 Prediction of Adsorption of diblock copolymer

Fig15 Theoretical predictions of the adsorbed layer thickness for a diblock copolymer

Fig 16 Adsorbed amount (mg mminus2) versus fraction of anchor segment for an AndashBndashA

triblock copolymer (PEOndash

PPOndash

PEO)

289T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1019

amounts to an increase in the osmotic pressure in the overlap region

as a result solvent will diffuse from thebulk to the overlap region thus

separating the particles and hence a strong repulsive energy arises

from this effect The above repulsive energy can be calculated by

considering the free energy of mixing of two polymer solutions as for

example treated by Flory and Krigbaum [22] The free energy of

mixing is given by two terms (i) An entropy term that depends on the

volume fraction of polymer and solvent (ii)An energy term that is

determined by the FloryndashHuggins interaction parameter χ

Using the above theory one can derive an expression for the free

energy of mixing of two polymer layers (assuming a uniform segment

density distribution in each layer) surrounding two spherical particles

as a function of the separation distance h between the particles The

expression for Gmix is

Gmix

kT =

2V 22V 1

m2

2

1

2minusχ

δminus

h

2

2

3R + 2δ + h

2

eth18THORN

k is the Boltzmann constant T is the absolute temperature V 2 is the

molar volume of polymer V 1 is the molar volume of solvent and ν 2 is

the number of polymer chains per unit area

The sign of Gmix depends on the value of the FloryndashHuggins

interaction parameter χ if χb05 Gmix is positive and the interaction

is repulsive if χN05 Gmix is negative and the interaction is attractiveif χ=05 Gmix =0 and this de1047297nes the θ-condition

52 Elastic interaction Gel

Thisarises fromthe lossin con1047297gurational entropy of thechains on

the approach of a second particle As a result of this approach the

volume available for the chains becomes restricted resulting in loss of

the number of con1047297gurations This can be illustrated by considering a

simple molecule represented by a rod that rotates freely in a

hemisphere across a surface (Fig 19) When the two surfaces are

separated byan in1047297nitedistanceinfin thenumber of con1047297gurations of the

rod is Ω(infin) which is proportional to the volume of the hemisphere

When a secondparticleapproaches to a distance h suchthatit cuts the

hemisphere (loosing some volume) the volume available to the chains

is reduced and the number of con1047297gurations become Ω(h) which is

less than Ω(infin) For two 1047298at plates Gel is given by the following

expression

Gel

kT = 2m2 ln

X heth THORN

X infineth THORN

= 2m2Rel heth THORN eth19THORN

where Rel

(h) is a geometric function whose form depends on the

segment density distribution It should be stressed that Gel is always

positive and could play a major role in steric stabilisation It becomes

very strong when the separation distance between the particles

becomes comparable to the adsorbed layer thickness δ

Combination of Gmix and Gel with GA (the van der Waals attractive

energy)gives the total free energy of interaction GT (assuming there is

no contribution from any residual electrostatic interaction) ie

GT = Gmix + Gel + GA eth20THORN

A schematic representation of the variation of Gmix Gel GA and GT

with surfacendashsurface separation distance h is shown in Fig 20

Gmix increases very sharply with decrease of h when hb2δ Gel

increases very sharply with decrease of h when hbδ GT versus h

shows a minimum Gmin at separation distances comparable to 2δ

When h b2δ GT shows a rapid increase with decrease in h

The depth of the minimum depends on the Hamaker constant A

the particle radius R and adsorbed layer thickness δ Gmin increases

with increase of A and R At a given A and R Gmin increases with

decrease in δ (ie with decrease of the molecular weight M w of the

stabiliser This is illustrated in Fig 21 which shows the energyndash

distance curves as a function of δR The larger the value of δR the

smaller the value of Gmin In this case the system may approach

thermodynamic stability as is the case with nano-dispersions

6 Emulsions stabilized by polymeric surfactants

The most effective method for emulsion stabilization is to use

polymeric surfactants that stronglyadsorb at the OWor WO interface

Fig 17 Schematic representation of the interaction between particles containing adsorbed polymer layers

Fig 18 Schematic representation of polymer layer overlap

Fig 19 Schematic representation of con1047297gurational entropy loss on approach of a

second particle

290 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1119

and produce effective steric stabilization against strong 1047298occulation

coalescence and Ostwald ripening [23]

As mentioned above a graft copolymer of the ABn type was

synthesized by grafting several alkyl groups on an inulin (polyfruc-

tose) chain The polymeric surfactant (INUTECregSP1) consists of a

linear polyfructose chain (the stabilizing A chain) and several alkyl

groups (the B chains) that provide multi-anchor attachment to the oil

droplets This polymeric surfactant produces enhanced steric stabili-

zation both in water and high electrolyte concentrations as will be

discussed later

For water-in-oil (WO) emulsions an AndashBndashA block copolymer of

poly (12-hydroxystearic acid) (PHS) (the A chains) and poly (ethylene

oxide) (PEO) (the B chain) PHSndashPEOndashPHS is commercially available

(Arlacel P135 UNIQEMA) The PEO chain (that is soluble in the water

droplets) forms the anchor chain whereas the PHS chains form the

stabilizing chains PHS is highly soluble in most hydrocarbon solvents

and is strongly solvated by its molecules The structure of the PHSndash

PEOndashPHS block copolymer is schematically shown in Fig 22The conformation of the polymeric surfactant at the WO interface

is schematically shown in Fig 23

Emulsions of Isopar Mwater and cyclomethiconewater were

prepared using INUTECregSP1 5050 (vv) OW emulsions were

prepared and the emulsi1047297er concentration was varied from 025 to 2

(wv) based on the oil phase 05 (wv) emulsi1047297erwas suf 1047297cient for

stabilization of these 5050 (vv) emulsions [23]

The emulsions were stored at room temperature and 50 degC and

optical micrographs were taken at intervals of time (for a year) in

order to check the stability Emulsions prepared in water were very

stableshowingno change in droplet size distributionover more than a

year period and this indicated absence of coalescence Any weak

1047298occulation that occurred was reversible and the emulsion could be

redispersed by gentle shaking Fig 24 shows an optical micrograph fora dilute 5050 (vv) emulsion that was stored for 15 and 14 weeks at

50 degC

No change in droplet size was observed after storage for more than

1 year at 50 degC indicating absence of coalescence The same result was

obtained when using different oils Emulsions were also stable against

coalescence in the presence of high electrolyte concentrations (up to

4 mol dmminus3 or ~25 NaCl

The above stability in high electrolyte concentrations is not

observed with polymeric surfactants based on polethylene oxide

The high stability observed using INUTECregSP1 is related to its

strong hydration both in water and in electrolyte solutions The

hydration of inulin (the backbone of HMI) could be assessed using

cloud point measurements A comparison was also made with PEO

with two molecular weights namely 4000 and 20000

Solutionsof PEO 4000 and 20000 showeda systematic decrease of

cloud point with increase in NaCl or MgSO4 concentration In contrast

inulin showed no cloud point up to 4 mol dmminus3 NaCl and up to 1 mol

dmminus3 MgSO4

The above results explain the difference between PEO and inulin

With PEO the chains show dehydration when the NaCl concentration

is increased above 2 mol dmminus3 or 05 mol dmminus3 MgSO4 The inulin

chains remain hydrated at much higher electrolyte concentrations It

seems that the linear polyfructose chains remain strongly hydrated athigh temperature and high electrolyte concentrations

The high emulsion stability obtained whenusingINUTECregSP1 can be

accounted for by the following factors (i) The multi-point attachment

of the polymer by several alkyl chains that are grafted on the backbone

(ii) The strong hydration of the polyfructose ldquoloopsrdquo both in water and

high electrolyte concentrations (χ remains below 05 under these

conditions) (iii) Thehigh volumefraction (concentration) of the loops at

the interface (iv) Enhanced steric stabilization this is the case with

multi-point attachment which produces strong elastic interaction

Evidence for the high stability of the liquid 1047297lm between emulsion

droplets when using INUTECregSP1 was obtained by Exerowa et al [24]

using disjoining pressure measurements This is illustrated in Fig 25

which shows a plot of disjoining pressure versus separation distance

between two emulsion droplets at various electrolyte concentrations

The results show that by increasing the capillary pressure a stable

Newton Black Film (NBF) is obtained at a 1047297lm thickness of sim7 nm

The lack of rupture of the 1047297lm at the highest pressure applied of

45times104 Pa indicate the high stability of the 1047297lm in water and in high

electrolyte concentrations (up to 20 mol dmminus3 NaCl)

The lack of rupture of the NBF up to the highest pressure applied

namely 45times104 Pa clearly indicatesthe high stability of the liquid1047297lm

in the presence of high NaCl concentrations (up to 2 mol dm minus3) This

result is consistent with the high emulsion stability obtained at high

electrolyte concentrations and high temperature Emulsions of Isopar

M-in-water are very stable under such conditions and this could be

accounted for by the high stability of the NBF The droplet size of

5050 OW emulsions prepared using 2 INUTEC regSP1 is in the region

of 1ndash10 μ m This corresponds to a capillary pressure of ~3times104 Pa for

the 1 μ m drops and ~3times103 Pa for the 10 μ m drops These capillarypressures are lower than those to which the NBF have been sub-

jected to and this clearly indicates the high stability obtained against

coalescence in these emulsions

61 WO emulsions stabilized with PHS ndashPEOndashPHS block copolymer

WO emulsions (the oil being Isopar M) were prepared using PHSndash

PEOndashPHS block copolymer at high water volume fractions (N07) The

emulsions have a narrow droplet size distribution with a z -average

radius of 183 nm [25] They also remained 1047298uid up to high water

volume fractions (N06) This could be illustrated from viscosityndash

volume fraction curves as is shown in Fig 26

Fig 20 Energyndashdistance curves for sterically stabilized systems

Fig 21 Variation of Gmin with δR

291T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1219

The effective volume fraction ϕeff of the emulsions (the core

droplets plus the adsorbed layer) could be calculated from the relative

viscosity and using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth21THORN

Where η r is the relative viscosity ϕp is the maximum packing fraction(sim07) and [ η ] is the intrinsic viscosity that is equal to 25 for hard-

spheres

The calculations based on Eq (22) are shown in Fig 27 (square

symbols) From the effective volume fraction ϕeff and the core volume

fraction ϕ the adsorbed layer thickness could be calculated This was

found to be in the region of 10 nm at ϕ =04 and it decreased with

increase in ϕ

The WO emulsions prepared using the PHSndashPEOndashPHS block

copolymer remained stable both at room temperature and 50 degC

This is consistent with the structure of the block copolymer the B

chain (PEO) is soluble in water and it forms a very strong anchor at the

WO interface The PHS chains (the A chains) provide effective steric

stabilization since the chains are highly soluble in Isopar M and are

strongly solvated by its molecules

7 Suspensions stabilised using polymeric surfactants

There are generally two procedures for preparation of solidliquid

dispersions

(i) Condensation methods build-up of particles from mole-

cular units ie nucleation and growth A special procedure

is the preparation of latexes by emulsion or dispersion

polymerization

(ii) Dispersion methods in this case one starts with preformed

large particles or crystals which are dispersed in the liquid by

using a surfactant (wetting agent) with subsequent breaking up

of the large particles by milling (comminution) to achieve the

desirable particle size distribution A dispersing agent (usuallya

polymeric surfactant) is used for the dispersion process and

subsequent stabilization of the resulting suspension

There are generally two procedures for preparation of latexes

(i) Emulsion polymerization the monomers that are essentially

insoluble in the aqueous medium are emulsi1047297ed using a

surfactant and an initiator is added while heating the system

to produce the polymer particles that are stabilized electro-

statically (when using ionic surfactants) or sterically (when

using non-ionic surfactants)

(ii) Dispersion polymerization the monomers are dissolved in a

solvent in which the resulting polymer particlesare insoluble A

protective colloid (normally a block or graft copolymer) is

added to prevent 1047298occulation of the resulting polymers

particles that are produced on addition of an initiator This

method is usually applied for the production of non-aqueous

latex dispersions and is sometimes referred to as Non-Aqueous

Dispersion Polymerization (NAD)

Surfactants play a crucial role in the process of latex preparation

since they determine the stabilizing ef 1047297ciency and the effectiveness of

the surface active agent ultimately determines the number of particles

Fig 22 Schematic representation of the structure of PHSndashPEOndashPHS block copolymer

Fig 23 Conformation of PHSndash

PEOndash

PHS polymeric surfactant at the WO interface

292 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1319

and their size The effectiveness of any surface active agent in

stabilizing the particles is the dominant factor and the number of

micelles formed is relatively unimportant In the NAD process themonomer normally an acrylic is dissolved in a non-aqueous solvent

normally an aliphatic hydrocarbon and an oil soluble initiator and a

stabilizer (to protect the resulting particles from1047298occulation) is added

to the reaction mixture The most successful stabilizers used in NAD

are block and graft copolymers Preformed graft stabilizers based

on poly(12-hydroxy stearic acid) (PHS) are simple to prepare and

effective in NAD polymerization

Dispersion methods are used for the preparation of suspensions of

preformed particles The role of surfactants (or polymers) in the

dispersion process can be analyzed in terms of the three processesinvolved (i) Wetting of the powder by the liquid (ii) Breaking of the

aggregates and agglomerates (iii) Comminution of the resulting

particles and their subsequent stabilization All these processes are

affected by surfactants or polymers which adsorb on the powder

surface thus aiding the wetting of the powder break-up of the

aggregates and agglomerates and subsequent reduction of particle

size by wet milling

71 Polymeric surfactants in emulsion polymerization

Recently the graft copolymer of hydrophobically modi1047297ed inulin

(INUTECreg SP1) has been used in emulsion polymerization of styrene

methyl methacrylate butyl acrylate and several other monomers [26]All lattices were prepared by emulsion polymerisation using potas-

sium persulphate as initiator The z -average particle size was

determined by photon correlation spectroscopy (PCS) and electron

micrographs were also taken

Fig 24 Optical micrographs of OW emulsions stabilized with INUTEC regSP1 stored at

50 degC for 15 weeks (a) and 14 weeks (b)

Fig 25 Variation of disjoining pressure with equivalent 1047297lm thickness at various NaCl concentrations

Fig 26 Viscosityndashvolume fraction for WO emulsion stabilized with PHSndashPEOndashPHS

block copolymer experimental data calculated using Eq (22)

293T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 919

the Scheutjens and Fleer theory The surface density σ is plotted

versus the fraction of anchor segmentsν A The adsorption depends on

the anchorbuoy composition

The amount of adsorption is higher than for homopolymers and

the adsorbed layer thickness is more extended and dense Fig 15

shows the theoretical prediction for the adsorbed layer thickness δwhich is plotted as a function of ν A For a triblock copolymer AndashBndashA

with two buoy chains (A) and one anchor chain (B) the behaviour is

similar to that of diblock copolymers This is shown in Fig16 for PEOndash

PPOndashPEO block (Pluronic)

5 Interaction between particles containing adsorbed polymeric

surfactant layers steric stabilization

When two particles each with a radius R and containing an

adsorbed polymer layer with a hydrodynamic thickness δh approach

each other to a surfacendashsurface separation distance h that is smaller

than 2 δh the polymer layers interact with each other resulting in two

main situations [21] (i) The polymer chains may overlap with each

other (ii) The polymer layer may undergo some compression In both

cases there will be an increase in the local segment density of the

polymer chains in the interaction region This is schematicallyillustrated in Fig 17 The real situation is perhaps in between the

above two cases ie the polymer chains may undergo some

interpenetration and some compression

Provided the dangling chains (the A chains in AndashB AndashBndashA block or

BAn graft copolymers) are in a good solvent this local increase in

segment density in the interaction zone will result in strong repulsion

as a result of two main effects (i) Increase in the osmotic pressure in

the overlap region as a result of the unfavourable mixing of the

polymer chains when these are in good solvent conditions This is

referred to as osmotic repulsion or mixing interaction and it is

described by a free energy of interaction Gmix (ii) Reduction of the

con1047297gurational entropy of the chains in the interaction zone this

entropy reduction results from the decrease in the volume available

for thechains when these areeither overlapped or compressed This is

referred to as volume restriction interaction entropic or elastic

interaction and it is described by a free energy of interaction Gel

Combination of Gmix and Gel is usually referred to as the steric

interaction free energy Gs ie

Gs = Gmix + Gel eth17THORN

The sign of Gmix depends on the solvency of the medium for the

chains If in a good solvent ie the FloryndashHuggins interaction

parameter χ is less than 05 then Gmix is positive and the mixing

interaction leads to repulsion (see below) In contrast if χN05 (ie the

chains are in a poor solvent condition) Gmix is negative and the

mixing interaction becomes attractive Gel is always positive and

hence in some cases one can produce stable dispersions in a relatively

poor solvent (enhanced steric stabilisation)

51 Mixing interaction Gmix

This results from the unfavourable mixing of the polymer chainswhen these are in a good solvent conditions This is schematically

shown in Fig 18

Consider two spherical particles with the same radius and each

containing an adsorbed polymer layer with thickness δ Before

overlap one can de1047297ne in each polymer layer a chemical potential

for the solvent μ iα and a volume fraction for the polymer in the

layerϕ2 In the overlap region (volume element dV ) the chemical

potential of the solvent is reduced to μ i β This results from the increase

in polymer segment concentration in this overlap region

In the overlap region the chemical potential of the polymer chains

is now higher than in the rest of the layer (with no overlap) This

Fig 13 In1047298uence of addition of KCl on adsorption

Fig 14 Prediction of Adsorption of diblock copolymer

Fig15 Theoretical predictions of the adsorbed layer thickness for a diblock copolymer

Fig 16 Adsorbed amount (mg mminus2) versus fraction of anchor segment for an AndashBndashA

triblock copolymer (PEOndash

PPOndash

PEO)

289T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1019

amounts to an increase in the osmotic pressure in the overlap region

as a result solvent will diffuse from thebulk to the overlap region thus

separating the particles and hence a strong repulsive energy arises

from this effect The above repulsive energy can be calculated by

considering the free energy of mixing of two polymer solutions as for

example treated by Flory and Krigbaum [22] The free energy of

mixing is given by two terms (i) An entropy term that depends on the

volume fraction of polymer and solvent (ii)An energy term that is

determined by the FloryndashHuggins interaction parameter χ

Using the above theory one can derive an expression for the free

energy of mixing of two polymer layers (assuming a uniform segment

density distribution in each layer) surrounding two spherical particles

as a function of the separation distance h between the particles The

expression for Gmix is

Gmix

kT =

2V 22V 1

m2

2

1

2minusχ

δminus

h

2

2

3R + 2δ + h

2

eth18THORN

k is the Boltzmann constant T is the absolute temperature V 2 is the

molar volume of polymer V 1 is the molar volume of solvent and ν 2 is

the number of polymer chains per unit area

The sign of Gmix depends on the value of the FloryndashHuggins

interaction parameter χ if χb05 Gmix is positive and the interaction

is repulsive if χN05 Gmix is negative and the interaction is attractiveif χ=05 Gmix =0 and this de1047297nes the θ-condition

52 Elastic interaction Gel

Thisarises fromthe lossin con1047297gurational entropy of thechains on

the approach of a second particle As a result of this approach the

volume available for the chains becomes restricted resulting in loss of

the number of con1047297gurations This can be illustrated by considering a

simple molecule represented by a rod that rotates freely in a

hemisphere across a surface (Fig 19) When the two surfaces are

separated byan in1047297nitedistanceinfin thenumber of con1047297gurations of the

rod is Ω(infin) which is proportional to the volume of the hemisphere

When a secondparticleapproaches to a distance h suchthatit cuts the

hemisphere (loosing some volume) the volume available to the chains

is reduced and the number of con1047297gurations become Ω(h) which is

less than Ω(infin) For two 1047298at plates Gel is given by the following

expression

Gel

kT = 2m2 ln

X heth THORN

X infineth THORN

= 2m2Rel heth THORN eth19THORN

where Rel

(h) is a geometric function whose form depends on the

segment density distribution It should be stressed that Gel is always

positive and could play a major role in steric stabilisation It becomes

very strong when the separation distance between the particles

becomes comparable to the adsorbed layer thickness δ

Combination of Gmix and Gel with GA (the van der Waals attractive

energy)gives the total free energy of interaction GT (assuming there is

no contribution from any residual electrostatic interaction) ie

GT = Gmix + Gel + GA eth20THORN

A schematic representation of the variation of Gmix Gel GA and GT

with surfacendashsurface separation distance h is shown in Fig 20

Gmix increases very sharply with decrease of h when hb2δ Gel

increases very sharply with decrease of h when hbδ GT versus h

shows a minimum Gmin at separation distances comparable to 2δ

When h b2δ GT shows a rapid increase with decrease in h

The depth of the minimum depends on the Hamaker constant A

the particle radius R and adsorbed layer thickness δ Gmin increases

with increase of A and R At a given A and R Gmin increases with

decrease in δ (ie with decrease of the molecular weight M w of the

stabiliser This is illustrated in Fig 21 which shows the energyndash

distance curves as a function of δR The larger the value of δR the

smaller the value of Gmin In this case the system may approach

thermodynamic stability as is the case with nano-dispersions

6 Emulsions stabilized by polymeric surfactants

The most effective method for emulsion stabilization is to use

polymeric surfactants that stronglyadsorb at the OWor WO interface

Fig 17 Schematic representation of the interaction between particles containing adsorbed polymer layers

Fig 18 Schematic representation of polymer layer overlap

Fig 19 Schematic representation of con1047297gurational entropy loss on approach of a

second particle

290 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1119

and produce effective steric stabilization against strong 1047298occulation

coalescence and Ostwald ripening [23]

As mentioned above a graft copolymer of the ABn type was

synthesized by grafting several alkyl groups on an inulin (polyfruc-

tose) chain The polymeric surfactant (INUTECregSP1) consists of a

linear polyfructose chain (the stabilizing A chain) and several alkyl

groups (the B chains) that provide multi-anchor attachment to the oil

droplets This polymeric surfactant produces enhanced steric stabili-

zation both in water and high electrolyte concentrations as will be

discussed later

For water-in-oil (WO) emulsions an AndashBndashA block copolymer of

poly (12-hydroxystearic acid) (PHS) (the A chains) and poly (ethylene

oxide) (PEO) (the B chain) PHSndashPEOndashPHS is commercially available

(Arlacel P135 UNIQEMA) The PEO chain (that is soluble in the water

droplets) forms the anchor chain whereas the PHS chains form the

stabilizing chains PHS is highly soluble in most hydrocarbon solvents

and is strongly solvated by its molecules The structure of the PHSndash

PEOndashPHS block copolymer is schematically shown in Fig 22The conformation of the polymeric surfactant at the WO interface

is schematically shown in Fig 23

Emulsions of Isopar Mwater and cyclomethiconewater were

prepared using INUTECregSP1 5050 (vv) OW emulsions were

prepared and the emulsi1047297er concentration was varied from 025 to 2

(wv) based on the oil phase 05 (wv) emulsi1047297erwas suf 1047297cient for

stabilization of these 5050 (vv) emulsions [23]

The emulsions were stored at room temperature and 50 degC and

optical micrographs were taken at intervals of time (for a year) in

order to check the stability Emulsions prepared in water were very

stableshowingno change in droplet size distributionover more than a

year period and this indicated absence of coalescence Any weak

1047298occulation that occurred was reversible and the emulsion could be

redispersed by gentle shaking Fig 24 shows an optical micrograph fora dilute 5050 (vv) emulsion that was stored for 15 and 14 weeks at

50 degC

No change in droplet size was observed after storage for more than

1 year at 50 degC indicating absence of coalescence The same result was

obtained when using different oils Emulsions were also stable against

coalescence in the presence of high electrolyte concentrations (up to

4 mol dmminus3 or ~25 NaCl

The above stability in high electrolyte concentrations is not

observed with polymeric surfactants based on polethylene oxide

The high stability observed using INUTECregSP1 is related to its

strong hydration both in water and in electrolyte solutions The

hydration of inulin (the backbone of HMI) could be assessed using

cloud point measurements A comparison was also made with PEO

with two molecular weights namely 4000 and 20000

Solutionsof PEO 4000 and 20000 showeda systematic decrease of

cloud point with increase in NaCl or MgSO4 concentration In contrast

inulin showed no cloud point up to 4 mol dmminus3 NaCl and up to 1 mol

dmminus3 MgSO4

The above results explain the difference between PEO and inulin

With PEO the chains show dehydration when the NaCl concentration

is increased above 2 mol dmminus3 or 05 mol dmminus3 MgSO4 The inulin

chains remain hydrated at much higher electrolyte concentrations It

seems that the linear polyfructose chains remain strongly hydrated athigh temperature and high electrolyte concentrations

The high emulsion stability obtained whenusingINUTECregSP1 can be

accounted for by the following factors (i) The multi-point attachment

of the polymer by several alkyl chains that are grafted on the backbone

(ii) The strong hydration of the polyfructose ldquoloopsrdquo both in water and

high electrolyte concentrations (χ remains below 05 under these

conditions) (iii) Thehigh volumefraction (concentration) of the loops at

the interface (iv) Enhanced steric stabilization this is the case with

multi-point attachment which produces strong elastic interaction

Evidence for the high stability of the liquid 1047297lm between emulsion

droplets when using INUTECregSP1 was obtained by Exerowa et al [24]

using disjoining pressure measurements This is illustrated in Fig 25

which shows a plot of disjoining pressure versus separation distance

between two emulsion droplets at various electrolyte concentrations

The results show that by increasing the capillary pressure a stable

Newton Black Film (NBF) is obtained at a 1047297lm thickness of sim7 nm

The lack of rupture of the 1047297lm at the highest pressure applied of

45times104 Pa indicate the high stability of the 1047297lm in water and in high

electrolyte concentrations (up to 20 mol dmminus3 NaCl)

The lack of rupture of the NBF up to the highest pressure applied

namely 45times104 Pa clearly indicatesthe high stability of the liquid1047297lm

in the presence of high NaCl concentrations (up to 2 mol dm minus3) This

result is consistent with the high emulsion stability obtained at high

electrolyte concentrations and high temperature Emulsions of Isopar

M-in-water are very stable under such conditions and this could be

accounted for by the high stability of the NBF The droplet size of

5050 OW emulsions prepared using 2 INUTEC regSP1 is in the region

of 1ndash10 μ m This corresponds to a capillary pressure of ~3times104 Pa for

the 1 μ m drops and ~3times103 Pa for the 10 μ m drops These capillarypressures are lower than those to which the NBF have been sub-

jected to and this clearly indicates the high stability obtained against

coalescence in these emulsions

61 WO emulsions stabilized with PHS ndashPEOndashPHS block copolymer

WO emulsions (the oil being Isopar M) were prepared using PHSndash

PEOndashPHS block copolymer at high water volume fractions (N07) The

emulsions have a narrow droplet size distribution with a z -average

radius of 183 nm [25] They also remained 1047298uid up to high water

volume fractions (N06) This could be illustrated from viscosityndash

volume fraction curves as is shown in Fig 26

Fig 20 Energyndashdistance curves for sterically stabilized systems

Fig 21 Variation of Gmin with δR

291T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1219

The effective volume fraction ϕeff of the emulsions (the core

droplets plus the adsorbed layer) could be calculated from the relative

viscosity and using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth21THORN

Where η r is the relative viscosity ϕp is the maximum packing fraction(sim07) and [ η ] is the intrinsic viscosity that is equal to 25 for hard-

spheres

The calculations based on Eq (22) are shown in Fig 27 (square

symbols) From the effective volume fraction ϕeff and the core volume

fraction ϕ the adsorbed layer thickness could be calculated This was

found to be in the region of 10 nm at ϕ =04 and it decreased with

increase in ϕ

The WO emulsions prepared using the PHSndashPEOndashPHS block

copolymer remained stable both at room temperature and 50 degC

This is consistent with the structure of the block copolymer the B

chain (PEO) is soluble in water and it forms a very strong anchor at the

WO interface The PHS chains (the A chains) provide effective steric

stabilization since the chains are highly soluble in Isopar M and are

strongly solvated by its molecules

7 Suspensions stabilised using polymeric surfactants

There are generally two procedures for preparation of solidliquid

dispersions

(i) Condensation methods build-up of particles from mole-

cular units ie nucleation and growth A special procedure

is the preparation of latexes by emulsion or dispersion

polymerization

(ii) Dispersion methods in this case one starts with preformed

large particles or crystals which are dispersed in the liquid by

using a surfactant (wetting agent) with subsequent breaking up

of the large particles by milling (comminution) to achieve the

desirable particle size distribution A dispersing agent (usuallya

polymeric surfactant) is used for the dispersion process and

subsequent stabilization of the resulting suspension

There are generally two procedures for preparation of latexes

(i) Emulsion polymerization the monomers that are essentially

insoluble in the aqueous medium are emulsi1047297ed using a

surfactant and an initiator is added while heating the system

to produce the polymer particles that are stabilized electro-

statically (when using ionic surfactants) or sterically (when

using non-ionic surfactants)

(ii) Dispersion polymerization the monomers are dissolved in a

solvent in which the resulting polymer particlesare insoluble A

protective colloid (normally a block or graft copolymer) is

added to prevent 1047298occulation of the resulting polymers

particles that are produced on addition of an initiator This

method is usually applied for the production of non-aqueous

latex dispersions and is sometimes referred to as Non-Aqueous

Dispersion Polymerization (NAD)

Surfactants play a crucial role in the process of latex preparation

since they determine the stabilizing ef 1047297ciency and the effectiveness of

the surface active agent ultimately determines the number of particles

Fig 22 Schematic representation of the structure of PHSndashPEOndashPHS block copolymer

Fig 23 Conformation of PHSndash

PEOndash

PHS polymeric surfactant at the WO interface

292 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1319

and their size The effectiveness of any surface active agent in

stabilizing the particles is the dominant factor and the number of

micelles formed is relatively unimportant In the NAD process themonomer normally an acrylic is dissolved in a non-aqueous solvent

normally an aliphatic hydrocarbon and an oil soluble initiator and a

stabilizer (to protect the resulting particles from1047298occulation) is added

to the reaction mixture The most successful stabilizers used in NAD

are block and graft copolymers Preformed graft stabilizers based

on poly(12-hydroxy stearic acid) (PHS) are simple to prepare and

effective in NAD polymerization

Dispersion methods are used for the preparation of suspensions of

preformed particles The role of surfactants (or polymers) in the

dispersion process can be analyzed in terms of the three processesinvolved (i) Wetting of the powder by the liquid (ii) Breaking of the

aggregates and agglomerates (iii) Comminution of the resulting

particles and their subsequent stabilization All these processes are

affected by surfactants or polymers which adsorb on the powder

surface thus aiding the wetting of the powder break-up of the

aggregates and agglomerates and subsequent reduction of particle

size by wet milling

71 Polymeric surfactants in emulsion polymerization

Recently the graft copolymer of hydrophobically modi1047297ed inulin

(INUTECreg SP1) has been used in emulsion polymerization of styrene

methyl methacrylate butyl acrylate and several other monomers [26]All lattices were prepared by emulsion polymerisation using potas-

sium persulphate as initiator The z -average particle size was

determined by photon correlation spectroscopy (PCS) and electron

micrographs were also taken

Fig 24 Optical micrographs of OW emulsions stabilized with INUTEC regSP1 stored at

50 degC for 15 weeks (a) and 14 weeks (b)

Fig 25 Variation of disjoining pressure with equivalent 1047297lm thickness at various NaCl concentrations

Fig 26 Viscosityndashvolume fraction for WO emulsion stabilized with PHSndashPEOndashPHS

block copolymer experimental data calculated using Eq (22)

293T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1019

amounts to an increase in the osmotic pressure in the overlap region

as a result solvent will diffuse from thebulk to the overlap region thus

separating the particles and hence a strong repulsive energy arises

from this effect The above repulsive energy can be calculated by

considering the free energy of mixing of two polymer solutions as for

example treated by Flory and Krigbaum [22] The free energy of

mixing is given by two terms (i) An entropy term that depends on the

volume fraction of polymer and solvent (ii)An energy term that is

determined by the FloryndashHuggins interaction parameter χ

Using the above theory one can derive an expression for the free

energy of mixing of two polymer layers (assuming a uniform segment

density distribution in each layer) surrounding two spherical particles

as a function of the separation distance h between the particles The

expression for Gmix is

Gmix

kT =

2V 22V 1

m2

2

1

2minusχ

δminus

h

2

2

3R + 2δ + h

2

eth18THORN

k is the Boltzmann constant T is the absolute temperature V 2 is the

molar volume of polymer V 1 is the molar volume of solvent and ν 2 is

the number of polymer chains per unit area

The sign of Gmix depends on the value of the FloryndashHuggins

interaction parameter χ if χb05 Gmix is positive and the interaction

is repulsive if χN05 Gmix is negative and the interaction is attractiveif χ=05 Gmix =0 and this de1047297nes the θ-condition

52 Elastic interaction Gel

Thisarises fromthe lossin con1047297gurational entropy of thechains on

the approach of a second particle As a result of this approach the

volume available for the chains becomes restricted resulting in loss of

the number of con1047297gurations This can be illustrated by considering a

simple molecule represented by a rod that rotates freely in a

hemisphere across a surface (Fig 19) When the two surfaces are

separated byan in1047297nitedistanceinfin thenumber of con1047297gurations of the

rod is Ω(infin) which is proportional to the volume of the hemisphere

When a secondparticleapproaches to a distance h suchthatit cuts the

hemisphere (loosing some volume) the volume available to the chains

is reduced and the number of con1047297gurations become Ω(h) which is

less than Ω(infin) For two 1047298at plates Gel is given by the following

expression

Gel

kT = 2m2 ln

X heth THORN

X infineth THORN

= 2m2Rel heth THORN eth19THORN

where Rel

(h) is a geometric function whose form depends on the

segment density distribution It should be stressed that Gel is always

positive and could play a major role in steric stabilisation It becomes

very strong when the separation distance between the particles

becomes comparable to the adsorbed layer thickness δ

Combination of Gmix and Gel with GA (the van der Waals attractive

energy)gives the total free energy of interaction GT (assuming there is

no contribution from any residual electrostatic interaction) ie

GT = Gmix + Gel + GA eth20THORN

A schematic representation of the variation of Gmix Gel GA and GT

with surfacendashsurface separation distance h is shown in Fig 20

Gmix increases very sharply with decrease of h when hb2δ Gel

increases very sharply with decrease of h when hbδ GT versus h

shows a minimum Gmin at separation distances comparable to 2δ

When h b2δ GT shows a rapid increase with decrease in h

The depth of the minimum depends on the Hamaker constant A

the particle radius R and adsorbed layer thickness δ Gmin increases

with increase of A and R At a given A and R Gmin increases with

decrease in δ (ie with decrease of the molecular weight M w of the

stabiliser This is illustrated in Fig 21 which shows the energyndash

distance curves as a function of δR The larger the value of δR the

smaller the value of Gmin In this case the system may approach

thermodynamic stability as is the case with nano-dispersions

6 Emulsions stabilized by polymeric surfactants

The most effective method for emulsion stabilization is to use

polymeric surfactants that stronglyadsorb at the OWor WO interface

Fig 17 Schematic representation of the interaction between particles containing adsorbed polymer layers

Fig 18 Schematic representation of polymer layer overlap

Fig 19 Schematic representation of con1047297gurational entropy loss on approach of a

second particle

290 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1119

and produce effective steric stabilization against strong 1047298occulation

coalescence and Ostwald ripening [23]

As mentioned above a graft copolymer of the ABn type was

synthesized by grafting several alkyl groups on an inulin (polyfruc-

tose) chain The polymeric surfactant (INUTECregSP1) consists of a

linear polyfructose chain (the stabilizing A chain) and several alkyl

groups (the B chains) that provide multi-anchor attachment to the oil

droplets This polymeric surfactant produces enhanced steric stabili-

zation both in water and high electrolyte concentrations as will be

discussed later

For water-in-oil (WO) emulsions an AndashBndashA block copolymer of

poly (12-hydroxystearic acid) (PHS) (the A chains) and poly (ethylene

oxide) (PEO) (the B chain) PHSndashPEOndashPHS is commercially available

(Arlacel P135 UNIQEMA) The PEO chain (that is soluble in the water

droplets) forms the anchor chain whereas the PHS chains form the

stabilizing chains PHS is highly soluble in most hydrocarbon solvents

and is strongly solvated by its molecules The structure of the PHSndash

PEOndashPHS block copolymer is schematically shown in Fig 22The conformation of the polymeric surfactant at the WO interface

is schematically shown in Fig 23

Emulsions of Isopar Mwater and cyclomethiconewater were

prepared using INUTECregSP1 5050 (vv) OW emulsions were

prepared and the emulsi1047297er concentration was varied from 025 to 2

(wv) based on the oil phase 05 (wv) emulsi1047297erwas suf 1047297cient for

stabilization of these 5050 (vv) emulsions [23]

The emulsions were stored at room temperature and 50 degC and

optical micrographs were taken at intervals of time (for a year) in

order to check the stability Emulsions prepared in water were very

stableshowingno change in droplet size distributionover more than a

year period and this indicated absence of coalescence Any weak

1047298occulation that occurred was reversible and the emulsion could be

redispersed by gentle shaking Fig 24 shows an optical micrograph fora dilute 5050 (vv) emulsion that was stored for 15 and 14 weeks at

50 degC

No change in droplet size was observed after storage for more than

1 year at 50 degC indicating absence of coalescence The same result was

obtained when using different oils Emulsions were also stable against

coalescence in the presence of high electrolyte concentrations (up to

4 mol dmminus3 or ~25 NaCl

The above stability in high electrolyte concentrations is not

observed with polymeric surfactants based on polethylene oxide

The high stability observed using INUTECregSP1 is related to its

strong hydration both in water and in electrolyte solutions The

hydration of inulin (the backbone of HMI) could be assessed using

cloud point measurements A comparison was also made with PEO

with two molecular weights namely 4000 and 20000

Solutionsof PEO 4000 and 20000 showeda systematic decrease of

cloud point with increase in NaCl or MgSO4 concentration In contrast

inulin showed no cloud point up to 4 mol dmminus3 NaCl and up to 1 mol

dmminus3 MgSO4

The above results explain the difference between PEO and inulin

With PEO the chains show dehydration when the NaCl concentration

is increased above 2 mol dmminus3 or 05 mol dmminus3 MgSO4 The inulin

chains remain hydrated at much higher electrolyte concentrations It

seems that the linear polyfructose chains remain strongly hydrated athigh temperature and high electrolyte concentrations

The high emulsion stability obtained whenusingINUTECregSP1 can be

accounted for by the following factors (i) The multi-point attachment

of the polymer by several alkyl chains that are grafted on the backbone

(ii) The strong hydration of the polyfructose ldquoloopsrdquo both in water and

high electrolyte concentrations (χ remains below 05 under these

conditions) (iii) Thehigh volumefraction (concentration) of the loops at

the interface (iv) Enhanced steric stabilization this is the case with

multi-point attachment which produces strong elastic interaction

Evidence for the high stability of the liquid 1047297lm between emulsion

droplets when using INUTECregSP1 was obtained by Exerowa et al [24]

using disjoining pressure measurements This is illustrated in Fig 25

which shows a plot of disjoining pressure versus separation distance

between two emulsion droplets at various electrolyte concentrations

The results show that by increasing the capillary pressure a stable

Newton Black Film (NBF) is obtained at a 1047297lm thickness of sim7 nm

The lack of rupture of the 1047297lm at the highest pressure applied of

45times104 Pa indicate the high stability of the 1047297lm in water and in high

electrolyte concentrations (up to 20 mol dmminus3 NaCl)

The lack of rupture of the NBF up to the highest pressure applied

namely 45times104 Pa clearly indicatesthe high stability of the liquid1047297lm

in the presence of high NaCl concentrations (up to 2 mol dm minus3) This

result is consistent with the high emulsion stability obtained at high

electrolyte concentrations and high temperature Emulsions of Isopar

M-in-water are very stable under such conditions and this could be

accounted for by the high stability of the NBF The droplet size of

5050 OW emulsions prepared using 2 INUTEC regSP1 is in the region

of 1ndash10 μ m This corresponds to a capillary pressure of ~3times104 Pa for

the 1 μ m drops and ~3times103 Pa for the 10 μ m drops These capillarypressures are lower than those to which the NBF have been sub-

jected to and this clearly indicates the high stability obtained against

coalescence in these emulsions

61 WO emulsions stabilized with PHS ndashPEOndashPHS block copolymer

WO emulsions (the oil being Isopar M) were prepared using PHSndash

PEOndashPHS block copolymer at high water volume fractions (N07) The

emulsions have a narrow droplet size distribution with a z -average

radius of 183 nm [25] They also remained 1047298uid up to high water

volume fractions (N06) This could be illustrated from viscosityndash

volume fraction curves as is shown in Fig 26

Fig 20 Energyndashdistance curves for sterically stabilized systems

Fig 21 Variation of Gmin with δR

291T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1219

The effective volume fraction ϕeff of the emulsions (the core

droplets plus the adsorbed layer) could be calculated from the relative

viscosity and using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth21THORN

Where η r is the relative viscosity ϕp is the maximum packing fraction(sim07) and [ η ] is the intrinsic viscosity that is equal to 25 for hard-

spheres

The calculations based on Eq (22) are shown in Fig 27 (square

symbols) From the effective volume fraction ϕeff and the core volume

fraction ϕ the adsorbed layer thickness could be calculated This was

found to be in the region of 10 nm at ϕ =04 and it decreased with

increase in ϕ

The WO emulsions prepared using the PHSndashPEOndashPHS block

copolymer remained stable both at room temperature and 50 degC

This is consistent with the structure of the block copolymer the B

chain (PEO) is soluble in water and it forms a very strong anchor at the

WO interface The PHS chains (the A chains) provide effective steric

stabilization since the chains are highly soluble in Isopar M and are

strongly solvated by its molecules

7 Suspensions stabilised using polymeric surfactants

There are generally two procedures for preparation of solidliquid

dispersions

(i) Condensation methods build-up of particles from mole-

cular units ie nucleation and growth A special procedure

is the preparation of latexes by emulsion or dispersion

polymerization

(ii) Dispersion methods in this case one starts with preformed

large particles or crystals which are dispersed in the liquid by

using a surfactant (wetting agent) with subsequent breaking up

of the large particles by milling (comminution) to achieve the

desirable particle size distribution A dispersing agent (usuallya

polymeric surfactant) is used for the dispersion process and

subsequent stabilization of the resulting suspension

There are generally two procedures for preparation of latexes

(i) Emulsion polymerization the monomers that are essentially

insoluble in the aqueous medium are emulsi1047297ed using a

surfactant and an initiator is added while heating the system

to produce the polymer particles that are stabilized electro-

statically (when using ionic surfactants) or sterically (when

using non-ionic surfactants)

(ii) Dispersion polymerization the monomers are dissolved in a

solvent in which the resulting polymer particlesare insoluble A

protective colloid (normally a block or graft copolymer) is

added to prevent 1047298occulation of the resulting polymers

particles that are produced on addition of an initiator This

method is usually applied for the production of non-aqueous

latex dispersions and is sometimes referred to as Non-Aqueous

Dispersion Polymerization (NAD)

Surfactants play a crucial role in the process of latex preparation

since they determine the stabilizing ef 1047297ciency and the effectiveness of

the surface active agent ultimately determines the number of particles

Fig 22 Schematic representation of the structure of PHSndashPEOndashPHS block copolymer

Fig 23 Conformation of PHSndash

PEOndash

PHS polymeric surfactant at the WO interface

292 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1319

and their size The effectiveness of any surface active agent in

stabilizing the particles is the dominant factor and the number of

micelles formed is relatively unimportant In the NAD process themonomer normally an acrylic is dissolved in a non-aqueous solvent

normally an aliphatic hydrocarbon and an oil soluble initiator and a

stabilizer (to protect the resulting particles from1047298occulation) is added

to the reaction mixture The most successful stabilizers used in NAD

are block and graft copolymers Preformed graft stabilizers based

on poly(12-hydroxy stearic acid) (PHS) are simple to prepare and

effective in NAD polymerization

Dispersion methods are used for the preparation of suspensions of

preformed particles The role of surfactants (or polymers) in the

dispersion process can be analyzed in terms of the three processesinvolved (i) Wetting of the powder by the liquid (ii) Breaking of the

aggregates and agglomerates (iii) Comminution of the resulting

particles and their subsequent stabilization All these processes are

affected by surfactants or polymers which adsorb on the powder

surface thus aiding the wetting of the powder break-up of the

aggregates and agglomerates and subsequent reduction of particle

size by wet milling

71 Polymeric surfactants in emulsion polymerization

Recently the graft copolymer of hydrophobically modi1047297ed inulin

(INUTECreg SP1) has been used in emulsion polymerization of styrene

methyl methacrylate butyl acrylate and several other monomers [26]All lattices were prepared by emulsion polymerisation using potas-

sium persulphate as initiator The z -average particle size was

determined by photon correlation spectroscopy (PCS) and electron

micrographs were also taken

Fig 24 Optical micrographs of OW emulsions stabilized with INUTEC regSP1 stored at

50 degC for 15 weeks (a) and 14 weeks (b)

Fig 25 Variation of disjoining pressure with equivalent 1047297lm thickness at various NaCl concentrations

Fig 26 Viscosityndashvolume fraction for WO emulsion stabilized with PHSndashPEOndashPHS

block copolymer experimental data calculated using Eq (22)

293T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1119

and produce effective steric stabilization against strong 1047298occulation

coalescence and Ostwald ripening [23]

As mentioned above a graft copolymer of the ABn type was

synthesized by grafting several alkyl groups on an inulin (polyfruc-

tose) chain The polymeric surfactant (INUTECregSP1) consists of a

linear polyfructose chain (the stabilizing A chain) and several alkyl

groups (the B chains) that provide multi-anchor attachment to the oil

droplets This polymeric surfactant produces enhanced steric stabili-

zation both in water and high electrolyte concentrations as will be

discussed later

For water-in-oil (WO) emulsions an AndashBndashA block copolymer of

poly (12-hydroxystearic acid) (PHS) (the A chains) and poly (ethylene

oxide) (PEO) (the B chain) PHSndashPEOndashPHS is commercially available

(Arlacel P135 UNIQEMA) The PEO chain (that is soluble in the water

droplets) forms the anchor chain whereas the PHS chains form the

stabilizing chains PHS is highly soluble in most hydrocarbon solvents

and is strongly solvated by its molecules The structure of the PHSndash

PEOndashPHS block copolymer is schematically shown in Fig 22The conformation of the polymeric surfactant at the WO interface

is schematically shown in Fig 23

Emulsions of Isopar Mwater and cyclomethiconewater were

prepared using INUTECregSP1 5050 (vv) OW emulsions were

prepared and the emulsi1047297er concentration was varied from 025 to 2

(wv) based on the oil phase 05 (wv) emulsi1047297erwas suf 1047297cient for

stabilization of these 5050 (vv) emulsions [23]

The emulsions were stored at room temperature and 50 degC and

optical micrographs were taken at intervals of time (for a year) in

order to check the stability Emulsions prepared in water were very

stableshowingno change in droplet size distributionover more than a

year period and this indicated absence of coalescence Any weak

1047298occulation that occurred was reversible and the emulsion could be

redispersed by gentle shaking Fig 24 shows an optical micrograph fora dilute 5050 (vv) emulsion that was stored for 15 and 14 weeks at

50 degC

No change in droplet size was observed after storage for more than

1 year at 50 degC indicating absence of coalescence The same result was

obtained when using different oils Emulsions were also stable against

coalescence in the presence of high electrolyte concentrations (up to

4 mol dmminus3 or ~25 NaCl

The above stability in high electrolyte concentrations is not

observed with polymeric surfactants based on polethylene oxide

The high stability observed using INUTECregSP1 is related to its

strong hydration both in water and in electrolyte solutions The

hydration of inulin (the backbone of HMI) could be assessed using

cloud point measurements A comparison was also made with PEO

with two molecular weights namely 4000 and 20000

Solutionsof PEO 4000 and 20000 showeda systematic decrease of

cloud point with increase in NaCl or MgSO4 concentration In contrast

inulin showed no cloud point up to 4 mol dmminus3 NaCl and up to 1 mol

dmminus3 MgSO4

The above results explain the difference between PEO and inulin

With PEO the chains show dehydration when the NaCl concentration

is increased above 2 mol dmminus3 or 05 mol dmminus3 MgSO4 The inulin

chains remain hydrated at much higher electrolyte concentrations It

seems that the linear polyfructose chains remain strongly hydrated athigh temperature and high electrolyte concentrations

The high emulsion stability obtained whenusingINUTECregSP1 can be

accounted for by the following factors (i) The multi-point attachment

of the polymer by several alkyl chains that are grafted on the backbone

(ii) The strong hydration of the polyfructose ldquoloopsrdquo both in water and

high electrolyte concentrations (χ remains below 05 under these

conditions) (iii) Thehigh volumefraction (concentration) of the loops at

the interface (iv) Enhanced steric stabilization this is the case with

multi-point attachment which produces strong elastic interaction

Evidence for the high stability of the liquid 1047297lm between emulsion

droplets when using INUTECregSP1 was obtained by Exerowa et al [24]

using disjoining pressure measurements This is illustrated in Fig 25

which shows a plot of disjoining pressure versus separation distance

between two emulsion droplets at various electrolyte concentrations

The results show that by increasing the capillary pressure a stable

Newton Black Film (NBF) is obtained at a 1047297lm thickness of sim7 nm

The lack of rupture of the 1047297lm at the highest pressure applied of

45times104 Pa indicate the high stability of the 1047297lm in water and in high

electrolyte concentrations (up to 20 mol dmminus3 NaCl)

The lack of rupture of the NBF up to the highest pressure applied

namely 45times104 Pa clearly indicatesthe high stability of the liquid1047297lm

in the presence of high NaCl concentrations (up to 2 mol dm minus3) This

result is consistent with the high emulsion stability obtained at high

electrolyte concentrations and high temperature Emulsions of Isopar

M-in-water are very stable under such conditions and this could be

accounted for by the high stability of the NBF The droplet size of

5050 OW emulsions prepared using 2 INUTEC regSP1 is in the region

of 1ndash10 μ m This corresponds to a capillary pressure of ~3times104 Pa for

the 1 μ m drops and ~3times103 Pa for the 10 μ m drops These capillarypressures are lower than those to which the NBF have been sub-

jected to and this clearly indicates the high stability obtained against

coalescence in these emulsions

61 WO emulsions stabilized with PHS ndashPEOndashPHS block copolymer

WO emulsions (the oil being Isopar M) were prepared using PHSndash

PEOndashPHS block copolymer at high water volume fractions (N07) The

emulsions have a narrow droplet size distribution with a z -average

radius of 183 nm [25] They also remained 1047298uid up to high water

volume fractions (N06) This could be illustrated from viscosityndash

volume fraction curves as is shown in Fig 26

Fig 20 Energyndashdistance curves for sterically stabilized systems

Fig 21 Variation of Gmin with δR

291T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1219

The effective volume fraction ϕeff of the emulsions (the core

droplets plus the adsorbed layer) could be calculated from the relative

viscosity and using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth21THORN

Where η r is the relative viscosity ϕp is the maximum packing fraction(sim07) and [ η ] is the intrinsic viscosity that is equal to 25 for hard-

spheres

The calculations based on Eq (22) are shown in Fig 27 (square

symbols) From the effective volume fraction ϕeff and the core volume

fraction ϕ the adsorbed layer thickness could be calculated This was

found to be in the region of 10 nm at ϕ =04 and it decreased with

increase in ϕ

The WO emulsions prepared using the PHSndashPEOndashPHS block

copolymer remained stable both at room temperature and 50 degC

This is consistent with the structure of the block copolymer the B

chain (PEO) is soluble in water and it forms a very strong anchor at the

WO interface The PHS chains (the A chains) provide effective steric

stabilization since the chains are highly soluble in Isopar M and are

strongly solvated by its molecules

7 Suspensions stabilised using polymeric surfactants

There are generally two procedures for preparation of solidliquid

dispersions

(i) Condensation methods build-up of particles from mole-

cular units ie nucleation and growth A special procedure

is the preparation of latexes by emulsion or dispersion

polymerization

(ii) Dispersion methods in this case one starts with preformed

large particles or crystals which are dispersed in the liquid by

using a surfactant (wetting agent) with subsequent breaking up

of the large particles by milling (comminution) to achieve the

desirable particle size distribution A dispersing agent (usuallya

polymeric surfactant) is used for the dispersion process and

subsequent stabilization of the resulting suspension

There are generally two procedures for preparation of latexes

(i) Emulsion polymerization the monomers that are essentially

insoluble in the aqueous medium are emulsi1047297ed using a

surfactant and an initiator is added while heating the system

to produce the polymer particles that are stabilized electro-

statically (when using ionic surfactants) or sterically (when

using non-ionic surfactants)

(ii) Dispersion polymerization the monomers are dissolved in a

solvent in which the resulting polymer particlesare insoluble A

protective colloid (normally a block or graft copolymer) is

added to prevent 1047298occulation of the resulting polymers

particles that are produced on addition of an initiator This

method is usually applied for the production of non-aqueous

latex dispersions and is sometimes referred to as Non-Aqueous

Dispersion Polymerization (NAD)

Surfactants play a crucial role in the process of latex preparation

since they determine the stabilizing ef 1047297ciency and the effectiveness of

the surface active agent ultimately determines the number of particles

Fig 22 Schematic representation of the structure of PHSndashPEOndashPHS block copolymer

Fig 23 Conformation of PHSndash

PEOndash

PHS polymeric surfactant at the WO interface

292 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1319

and their size The effectiveness of any surface active agent in

stabilizing the particles is the dominant factor and the number of

micelles formed is relatively unimportant In the NAD process themonomer normally an acrylic is dissolved in a non-aqueous solvent

normally an aliphatic hydrocarbon and an oil soluble initiator and a

stabilizer (to protect the resulting particles from1047298occulation) is added

to the reaction mixture The most successful stabilizers used in NAD

are block and graft copolymers Preformed graft stabilizers based

on poly(12-hydroxy stearic acid) (PHS) are simple to prepare and

effective in NAD polymerization

Dispersion methods are used for the preparation of suspensions of

preformed particles The role of surfactants (or polymers) in the

dispersion process can be analyzed in terms of the three processesinvolved (i) Wetting of the powder by the liquid (ii) Breaking of the

aggregates and agglomerates (iii) Comminution of the resulting

particles and their subsequent stabilization All these processes are

affected by surfactants or polymers which adsorb on the powder

surface thus aiding the wetting of the powder break-up of the

aggregates and agglomerates and subsequent reduction of particle

size by wet milling

71 Polymeric surfactants in emulsion polymerization

Recently the graft copolymer of hydrophobically modi1047297ed inulin

(INUTECreg SP1) has been used in emulsion polymerization of styrene

methyl methacrylate butyl acrylate and several other monomers [26]All lattices were prepared by emulsion polymerisation using potas-

sium persulphate as initiator The z -average particle size was

determined by photon correlation spectroscopy (PCS) and electron

micrographs were also taken

Fig 24 Optical micrographs of OW emulsions stabilized with INUTEC regSP1 stored at

50 degC for 15 weeks (a) and 14 weeks (b)

Fig 25 Variation of disjoining pressure with equivalent 1047297lm thickness at various NaCl concentrations

Fig 26 Viscosityndashvolume fraction for WO emulsion stabilized with PHSndashPEOndashPHS

block copolymer experimental data calculated using Eq (22)

293T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1219

The effective volume fraction ϕeff of the emulsions (the core

droplets plus the adsorbed layer) could be calculated from the relative

viscosity and using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth21THORN

Where η r is the relative viscosity ϕp is the maximum packing fraction(sim07) and [ η ] is the intrinsic viscosity that is equal to 25 for hard-

spheres

The calculations based on Eq (22) are shown in Fig 27 (square

symbols) From the effective volume fraction ϕeff and the core volume

fraction ϕ the adsorbed layer thickness could be calculated This was

found to be in the region of 10 nm at ϕ =04 and it decreased with

increase in ϕ

The WO emulsions prepared using the PHSndashPEOndashPHS block

copolymer remained stable both at room temperature and 50 degC

This is consistent with the structure of the block copolymer the B

chain (PEO) is soluble in water and it forms a very strong anchor at the

WO interface The PHS chains (the A chains) provide effective steric

stabilization since the chains are highly soluble in Isopar M and are

strongly solvated by its molecules

7 Suspensions stabilised using polymeric surfactants

There are generally two procedures for preparation of solidliquid

dispersions

(i) Condensation methods build-up of particles from mole-

cular units ie nucleation and growth A special procedure

is the preparation of latexes by emulsion or dispersion

polymerization

(ii) Dispersion methods in this case one starts with preformed

large particles or crystals which are dispersed in the liquid by

using a surfactant (wetting agent) with subsequent breaking up

of the large particles by milling (comminution) to achieve the

desirable particle size distribution A dispersing agent (usuallya

polymeric surfactant) is used for the dispersion process and

subsequent stabilization of the resulting suspension

There are generally two procedures for preparation of latexes

(i) Emulsion polymerization the monomers that are essentially

insoluble in the aqueous medium are emulsi1047297ed using a

surfactant and an initiator is added while heating the system

to produce the polymer particles that are stabilized electro-

statically (when using ionic surfactants) or sterically (when

using non-ionic surfactants)

(ii) Dispersion polymerization the monomers are dissolved in a

solvent in which the resulting polymer particlesare insoluble A

protective colloid (normally a block or graft copolymer) is

added to prevent 1047298occulation of the resulting polymers

particles that are produced on addition of an initiator This

method is usually applied for the production of non-aqueous

latex dispersions and is sometimes referred to as Non-Aqueous

Dispersion Polymerization (NAD)

Surfactants play a crucial role in the process of latex preparation

since they determine the stabilizing ef 1047297ciency and the effectiveness of

the surface active agent ultimately determines the number of particles

Fig 22 Schematic representation of the structure of PHSndashPEOndashPHS block copolymer

Fig 23 Conformation of PHSndash

PEOndash

PHS polymeric surfactant at the WO interface

292 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1319

and their size The effectiveness of any surface active agent in

stabilizing the particles is the dominant factor and the number of

micelles formed is relatively unimportant In the NAD process themonomer normally an acrylic is dissolved in a non-aqueous solvent

normally an aliphatic hydrocarbon and an oil soluble initiator and a

stabilizer (to protect the resulting particles from1047298occulation) is added

to the reaction mixture The most successful stabilizers used in NAD

are block and graft copolymers Preformed graft stabilizers based

on poly(12-hydroxy stearic acid) (PHS) are simple to prepare and

effective in NAD polymerization

Dispersion methods are used for the preparation of suspensions of

preformed particles The role of surfactants (or polymers) in the

dispersion process can be analyzed in terms of the three processesinvolved (i) Wetting of the powder by the liquid (ii) Breaking of the

aggregates and agglomerates (iii) Comminution of the resulting

particles and their subsequent stabilization All these processes are

affected by surfactants or polymers which adsorb on the powder

surface thus aiding the wetting of the powder break-up of the

aggregates and agglomerates and subsequent reduction of particle

size by wet milling

71 Polymeric surfactants in emulsion polymerization

Recently the graft copolymer of hydrophobically modi1047297ed inulin

(INUTECreg SP1) has been used in emulsion polymerization of styrene

methyl methacrylate butyl acrylate and several other monomers [26]All lattices were prepared by emulsion polymerisation using potas-

sium persulphate as initiator The z -average particle size was

determined by photon correlation spectroscopy (PCS) and electron

micrographs were also taken

Fig 24 Optical micrographs of OW emulsions stabilized with INUTEC regSP1 stored at

50 degC for 15 weeks (a) and 14 weeks (b)

Fig 25 Variation of disjoining pressure with equivalent 1047297lm thickness at various NaCl concentrations

Fig 26 Viscosityndashvolume fraction for WO emulsion stabilized with PHSndashPEOndashPHS

block copolymer experimental data calculated using Eq (22)

293T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1319

and their size The effectiveness of any surface active agent in

stabilizing the particles is the dominant factor and the number of

micelles formed is relatively unimportant In the NAD process themonomer normally an acrylic is dissolved in a non-aqueous solvent

normally an aliphatic hydrocarbon and an oil soluble initiator and a

stabilizer (to protect the resulting particles from1047298occulation) is added

to the reaction mixture The most successful stabilizers used in NAD

are block and graft copolymers Preformed graft stabilizers based

on poly(12-hydroxy stearic acid) (PHS) are simple to prepare and

effective in NAD polymerization

Dispersion methods are used for the preparation of suspensions of

preformed particles The role of surfactants (or polymers) in the

dispersion process can be analyzed in terms of the three processesinvolved (i) Wetting of the powder by the liquid (ii) Breaking of the

aggregates and agglomerates (iii) Comminution of the resulting

particles and their subsequent stabilization All these processes are

affected by surfactants or polymers which adsorb on the powder

surface thus aiding the wetting of the powder break-up of the

aggregates and agglomerates and subsequent reduction of particle

size by wet milling

71 Polymeric surfactants in emulsion polymerization

Recently the graft copolymer of hydrophobically modi1047297ed inulin

(INUTECreg SP1) has been used in emulsion polymerization of styrene

methyl methacrylate butyl acrylate and several other monomers [26]All lattices were prepared by emulsion polymerisation using potas-

sium persulphate as initiator The z -average particle size was

determined by photon correlation spectroscopy (PCS) and electron

micrographs were also taken

Fig 24 Optical micrographs of OW emulsions stabilized with INUTEC regSP1 stored at

50 degC for 15 weeks (a) and 14 weeks (b)

Fig 25 Variation of disjoining pressure with equivalent 1047297lm thickness at various NaCl concentrations

Fig 26 Viscosityndashvolume fraction for WO emulsion stabilized with PHSndashPEOndashPHS

block copolymer experimental data calculated using Eq (22)

293T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1419

Emulsion polymerization of styrene or methylmethacrylate showed

an optimum weight ratio of (INUTEC)Monomer of 00033 for PS and

0001 for PMMA particles The (initiator)(Monomer) ratio was

kept constant at 000125 The monomer conversion was higher than

85 in all cases Latex dispersions of PS reaching 50 and of PMMA

reaching 40 could be obtained using such low concentration of

INUTECregSP1 Fig 27 shows the variation of particle diameter with

monomer concentration

The stability of the latexes was determined by determining the

critical coagulation concentration (CCC) using CaCl2 The CCC was low

(00175ndash005 mol dmminus3) but this was higher than that for the latex

prepared without surfactant Post addition of INUTEC regSP1 resulted in

a large increase in the CCC as is illustrated in Fig 28 which shows log

W minus log C curves (where W is the ratio between the fast 1047298occulation

rate constant to the slow 1047298occulation rate constant referred to as the

stability ratio) at various additions of INUTECregSP1

As with the emulsions the high stability of the latex when using

INUTECregSP1 is due to the strong adsorption of the polymeric

surfactant on the latex particles and formation of strongly hydrated

loops and tails of polyfructose that provide effective steric stabiliza-

tion Evidence for the strong repulsion produced when using

INUTECregSP1 was obtained from Atomic Force Microscopy investiga-

tions [27] whereby the force between hydrophobic glass spheres and

hydrophobic glass plate both containing an adsorbed layer of

INUTECregSP1 was measured as a function of distance of separation

both in water and in the presence of various Na2SO4 concentrations

The results are shown in Figs 29 and 30

Fig 27 Electron micrographs of the latexes

Fig 28 In1047298uence of post addition of INUTECreg

SP1 on the latex stability

Fig 29 Forcendashdistance curves between hydrophobised glass surfaces containing

adsorbed INUTECreg

SP1 in water

294 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1519

711 Dispersion polymerization

This method is usually applied for the preparation of non-aqueous

latex dispersions and hence it is referred to as NAD Themethod has also

been adapted to prepare aqueouslatex dispersions by using an alcoholndash

water mixture In the NAD process the monomer normally an acrylic

is dissolved in a non-aqueous solvent normally an aliphatic hydro-

carbon and an oil soluble initiator and a stabilizer (to protect the

resulting particles from 1047298occulation) is added to the reaction mixture

The most successful stabilizers used in NAD are block and graft copo-

lymers Preformed graft stabilizers based on poly(12-hydroxy stearic

acid) (PHS) are simple to prepare and effective in NAD polymerization

Commercial 12-hydroxystearic acid contains 8ndash15 palmitic and

stearic acids which limits the molecular weight during polymerization

to an average of 1500ndash2000 This oligomer may be converted to a

lsquorsquomacromonomer rsquorsquo by reacting the carboxylic group with glycidyl

methacrylate The macromonomer is then copolymerized with an

equal weight of methyl methacrylate (MMA) or similar monomer to

give a lsquorsquocombrsquorsquo graft copolymer with an average molecular weight of 10000ndash20000 The graft copolymer contains on average 5ndash10 PHS

chains pendent from a polymeric anchor backbone of PMMA This a

graft copolymer can stabilize latex particles of various monomers The

major limitation of the monomer composition is that the polymer

produced should be insoluble in the medium used

NAD polymerization is carried in two steps (i) Seed stage the

diluent portion of the monomer portion of dispersant and initiator

(azo or peroxy type) are heated to form an initial low-concentration

1047297ne dispersion (ii) Growth stage the remaining monomer together

with more dispersant and initiator are then fed over the course of

several hours to complete the growth of the particles A small amount

of transfer agent is usually added to control the molecular weight

Excellent control of particle size is achieved by proper choice of the

designed dispersant and correct distribution of dispersant between

the seed and growth stages NAD acrylic polymers are applied in

automotive thermosetting polymers and hydroxy monomers may be

included in the monomer blend used

72 Polymeric surfactants for stabilisation of preformed latex dispersions

For this purpose polystyrene (PS) latexes were prepared using the

surfactant-free emulsion polymerisation Two latexes with z -average

diameter of 427 and 867 (as measured using Photon Correlation

Spectroscopy PCS) that are reasonably monodisperse were prepared

[28] Two polymeric surfactants namely Hypermer CG-6 and Atlox

4913 (UNIQEMA UK) were used Both are graft (lsquorsquocombrsquorsquo) type

consisting of polymethylmethacrylatepolymethacrylic acid (PMMA

PMA) backbone with methoxy-capped polyethylene oxide (PEO) side

chains (M =750 Da) Hypermer CG-6 is the same graft copolymer asAtlox 4913 but it contains higher proportion of methacrylic acid in the

backbone The average molecular weight of the polymer is ~5000 Da

Fig 31 shows a typical adsorption isotherm of Atlox 4913 on the two

latexes

Similar results were obtained for Hypermer CG-6 but the plateau

adsorption was lower (12 mg mminus2 compared with 15 mg mminus2 for

Atlox 4913) It is likely that the backbone of Hypermer CG-6 that

contains more PMA is more polar and hence less strongly adsorbed

The amount of adsorption was independent of particle size

The in1047298uence of temperature on adsorption is shown in Fig 32The

amount of adsorption increases with increase of temperature This is

due to the poorer solvency of the medium for the PEO chains The PEO

chains become less hydrated at higher temperature and the reduction

of solubility of the polymer enhances adsorptionThe adsorbed layer thickness of the graft copolymer on the latexes

was determined using rheological measurements Steady state (shear

stress σ minusshear rate γ ) measurements were carried out and the results

were 1047297tted to the Bingham equation to obtain the yield value σ β and

the high shear viscosity η of the suspension

σ = σ β + ηγ eth22THORN

As an illustration Fig 33 shows a plot of σ β versus volume fraction

ϕ of the latex for Atlox 4913 Similar results were obtained for latexes

stabilised using Hypermer CG-6

At any given volume fraction the smaller latex has higherσ β when

compared to the larger latex This is dueto thehigher ratio of adsorbed

Fig 30 Forcendashdistance curves for hydrophobised glass surfaces containing adsorbed

INUTECregSP1 at various Na2SO4 concentrations

Fig 31 Adsorption isotherms of Atlox 4913 on the two latexes at 25 degC

Fig 32 Effect of temperature on adsorption of Atlox 4913 on PS

295T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1619

layer thickness to particle radius ΔR for the smaller latex The

effective volume fraction of the latex ϕeff is related to the core volume

fraction ϕ by the equation

eff = 1 + Δ

R

3

eth23THORN

As discussed before ϕeff can be calculated from the relative

viscosity η r using the DoughertyndashKrieger equation

η r = 1minuseff

p

minus η frac12 p

eth24THORN

where ϕp is the maximum packing fraction

The maximum packing fraction ϕp can be calculated using the

following empirical Eq (28)

η 1=2r minus1

= 1

p

η 1=2

minus1

+ 125 eth25THORN

The results showed a gradual decrease of adsorbed layer thickness

Δ with increase of the volume fraction ϕ For the latex with diameter

D of 867 nm and Atlox 4913 Δ decreased from 175 nm at ϕ=036 to

65 at ϕ =057 For Hypermer CG-6 with the same latex Δ decreased

from 118 nm at ϕ =049 to 65 at ϕ =057 The reduction of Δ with

increase in ϕ may be due to overlap andor compression of the

adsorbed layers as the particles come close to each other at higher

volume fraction of the latex

The stability of the latexes was determined using viscoelastic

measurements For this purpose dynamic (oscillatory) measurements

were used to obtain the storage modulus G

the elastic modulus Gprime

and the viscous modulus GPrime as a function of strain amplitude γ o and

frequency ω (rad sminus1) The method relies on application of a sinusoidal

strain or stress and the resulting stress or strain is measured

simultaneously For a viscoelastic system the strain and stress sine

waves oscillate with the same frequency but out of phase From the

time shift Δt and ω one can obtain the phase angle shift δ

The ratio of the maximum stress σ o to the maximum strain γ ogives the complex modulus |G|

jGj = σ oγ o

eth26THORN

|G| can be resolved into two components Storage (elastic) modulus Gprime

the real component of the complex modulus loss (viscous) modulusGPrime

the imaginary component of the complex modulus The complex

modulus can be resolved into Gprime and GPrime using vector analysis and the

phase angle shift δ

G V = jGj cosδ eth27THORN

GW = jGj sinδ eth28THORN

Gprime is measured as a function of electrolyte concentration andor

temperature to assess the latex stability As an illustration Fig 34shows the variation of Gprime with temperature for latex stabilised with

Atlox 4913 in the absence of any added electrolyte and in the presence

of 01 02 and 03 mol dmminus3 Na2SO4 In the absence of electrolyte Gprime

showed no change with temperature up to 65 degC

In the presence of 01 mol dmminus3 Na2SO4 Gprime remained constant up

to 40 degC above which Gprime increased with further increase of

temperature This temperature is denoted as the critical 1047298occulation

temperature (CFT) The CFT decreases with increase in electrolyte

concentration reaching ~30 degC in 02 and 03 mol dmminus3 Na2SO4 This

reduction in CFT with increase in electrolyte concentration is due to

the reduction in solvency of the PEO chains with increase in

electrolyte concentrations The latex stabilised with Hypermer CG-6

gave relatively higher CFT values when compared with that stabilised

using Atlox 4913

73 Use of polymeric surfactants for preparation and stabilisation of

nano-emulsions

Nano-emulsions are systems that cover the size range 20ndash200 nm

[29ndash31] They can be transparent translucent or turbid depending on

the droplet radius and refractive index difference between the

droplets and the continuous phase This can be understood from

consideration of the dependence of light scattering (turbidity) on the

above two factors For droplets with a radius that is less than (120) of

the wave length of the light the turbidity τ is given by the following

equation

τ = KN oV 2 eth29THORN

Where K is an optical constant that is related to the difference in

refractive index between the droplets np and the medium no and N o is

the number of droplets each with a volume V

It is clear from Eq (30) that τ decreases with decrease of K ie

smaller (npminusno) decrease of N o and decrease of V Thus to produce a

transparent nano-emulsion one has to decrease the difference

between the refractive index of the droplets and the medium (ie

try to match the two refractive indices) If such matching is not

possible then one has to reduce the droplet size (by high pressure

homogenization) to values below 50 nm It is also necessary to use a

nano-emulsion with low oil volume fraction (in the region of 02)

Fig 34 Variationof Gprime

withtemperature in water andat various Na2SO4 concentrations

Fig 33 Variation of yield stress with latex volume fraction for Atlox 4913

296 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1719

Nano-emulsions are only kinetically stable They have to distin-

guished from microemulsions (that cover the size range 5ndash50 nm)

which are mostly transparent and thermodynamically stable The longterm physical stability of nano-emulsions (with no apparent 1047298occula-

tion or coalescence) make them unique and they are sometimes

referred to as ldquoApproaching Thermodynamic Stabilityrdquo The inherently

high colloid stabilityof nano-emulsions canbe well understood from a

consideration of their steric stabilisation (when using nonionic

surfactants andor polymers) and how this is affected by the ratio of

theadsorbedlayer thickness to droplet radius as was discussedbefore

Unless adequately prepared (to control the droplet size distribu-

tion) and stabilised against Ostwald ripening (that occurswhen the oil

hassome 1047297nite solubility in the continuous medium) nano-emulsions

may show an increase in the droplet size and an initially transparent

system may become turbid on storage

The attraction of nano-emulsions for application in Personal Careand Cosmetics as well as in Health care is due to the following

advantages (i) The very small droplet size causes a large reduction in

the gravity force and the Brownian motion may be suf 1047297cient for

overcoming gravity mdash This means that no creaming or sedimentation

occurs on storage (ii) The small droplet size also prevents any

1047298occulation of the droplets Weak 1047298occulation is prevented and this

enables the system to remain dispersed with no separation (iii) The

small droplets also prevent their coalescence since these droplets are

non-deformable and hence surface 1047298uctuations are prevented In

addition the signi1047297cant surfactant 1047297lm thickness (relative to droplet

Fig 35 Silicone oil in water nano-emulsions stabilized with INUTECregSP1

Fig 36 r 3

versus t for nano-emulsions based on hydrocarbon oils

297T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1819

radius) prevents any thinning or disruption of the liquid 1047297lm between

the droplets

The production of small droplets (submicron) requires application

of high energy The process of emulsi1047297cation is generally inef 1047297cient

Simple calculations show that the mechanical energy required for

emulsi1047297cation exceeds the interfacial energy by several orders of

magnitude For example to produce a nano-emulsion atϕ=01 with an

average radius R of 200 nm using a surfactant that gives an interfacial

tension γ =10 mNm

minus1

the net increase in surface free energy is Aγ = 3ϕγ R =15times104 Jmminus3 The mechanical energy required in a

homogenizer is 15times107 Jmminus3 ie an ef 1047297ciency of 01 The rest of

the energy (999) is dissipated as heat

The intensity of the process or the effectiveness in making small

droplets is often governed by the net power density (ε (t ))

p = e t eth THORNdt eth30THORN

where t is the time during which emulsi1047297cation occurs

Break up of droplets will only occur at high ε values which means

that the energy dissipated at low ε levels is wasted Batch processes

are generally less ef 1047297cient than continuous processes This shows why

with a stirrer in a large vessel most of the energy applies at low

intensity is dissipated as heat In a homogenizer p is simply equal to

the homogenizer pressureSeveral procedures may be applied to enhance the ef 1047297ciency of

emulsi1047297cation when producing nano-emulsions (i) One should

optimise the ef 1047297ciency of agitation by increasing ε and decreasing

dissipation time (ii)The nano-emulsion is preferably prepared at high

volume faction of the disperse phase and diluted afterwards However

very high ϕ values may result in coalescence during emulsi1047297cation

(iii) Add more surfactant whereby creating a smaller γ eff and possibly

diminishing recoalescence (iv) Use surfactant mixture that show

more reduction in γ the individual components (v) If possible

dissolve the surfactant in the disperse phase rather than the

continuous phase mdash this often leads to smaller droplets (vi) It may

be useful to emulsify in steps of increasing intensity particularly with

nano-emulsions having highly viscous disperse phase

The high kinetic stability of nano-emulsions can be explained fromconsideration of the energyndashdistance curves for sterically stabilized

dispersions shown in Fig 21 It can be seen from Fig 21 that the depth

of the minimum decrease with increasing δR With nano-emulsions

having a radius in the region of 50 nm andan adsorbed layer thickness

of say 10 nm the value of δR is 02 This high value (when compared

with the situation with macroemulsions where δR is at least an order

of magnitude lower) results in a very shallow minimum (which could

be less than kT ) This situation results in very high stability with no

1047298occulation (weak or strong) In addition the very small size of the

droplets and the dense adsorbed layers ensures lack of deformation of

the interface lack of thinning and disruption of the liquid 1047297lm

between the droplets and hence coalescence is also prevented

One of the main problems with Nanoemulsions is Ostwald

ripening which results from the difference in solubility between

small and large droplets The difference in chemical potential of

dispersed phase droplets between different sized droplets as given by

Lord Kelvin [32]

s r eth THORN = s infineth THORNexp2γ V m

rRT

eth31THORN

where s(r ) is the solubility surrounding a particle of radius r c (infin) is

the bulk phase solubility and V m is the molar volume of the dispersed

phase

The quantity (2γ V mRT ) is termed the characteristic length It has

an order of ~ 1 nm or less indicating that the difference in solubility of

a 1 μ m droplet is of the order of 01 or less

Theoretically Ostwald ripening should lead to condensation of all

droplets into a single drop (ie phase separation) This does not occur

in practice since the rate of growth decreases with increase of droplet

size

For two droplets of radii r 1 and r 2 (where r 1br 2)

RT

V mln

s r 1eth THORN

s r 2eth THORN

= 2γ

1

r 1minus

1

r 2

eth32THORN

Eq (33) shows that the larger the difference between r 1 and r 2 the

higher the rate of Ostwald ripening

Ostwald ripening can be quantitatively assessed from plots of thecube of the radius versus time t [33ndash35]

r 3 = 8

9

s infineth THORNγ V mD

ρRT

eth33THORN

where D is the diffusion coef 1047297cient of the disperse phase in the

continuous phase and ρ is its density

Ostwald ripening can be reduced by incorporation of a second

component which is insoluble in the continuous phase (eg squalane)

In this case signi1047297cant partitioning between different droplets occurs

with the component having low solubility in the continuous phase

expected to be concentrated in the smaller droplets During Ostwald

ripening in two component disperse phase system equilibrium is

established when the difference in chemical potential between

different size droplets (which results from curvature effects) is

balanced by the difference in chemical potential resulting from

partitioning of the two components If the secondary component has

zero solubility in the continuous phase the size distribution will not

deviate from the initial one (the growth rate is equal to zero) In the

case of limited solubility of the secondarycomponent the distribution

is the same as governed by Eq (33) ie a mixture growth rate is

obtained which is still lower than that of the more soluble component

Another method for reducing Ostwald ripening depends on

modi1047297cation of the interfacial 1047297lm at the OW interface According

to Eq (33) reduction in γ results in reduction of Ostwald ripening

However this alone is not suf 1047297cient since one has to reduce γ by

several orders of magnitude Walstra [3637] suggested that by using

surfactants which are strongly adsorbed at the OW interface (ie

polymeric surfactants) and which do not desorb during ripening therate could be signi1047297cantly reduced An increase in the surface

dilational modulus ε and decrease in γ would be observed for the

shrinking drops The difference in ε between the droplets would

balance the difference in capillary pressure (ie curvature effects)

To achieve the above effect it is useful to use AndashBndashA block

copolymers that are soluble in the oil phase and insoluble in the

continuous phase A strongly adsorbed polymeric surfactant that has

limited solubility in the aqueous phase can also be used eg

hydrophobically modi1047297ed inulin INUTECregSP1 This is illustrated in

Fig 35 which shows plots of r 3 versus time for 20 vv silicone oil-in-

water emulsions at two concentrations of INUTECregSP1 (16 top

curve and 24 bottom curve) [38] The concentration of INUTECregSP1

is much lower than that required when using non-ionic surfactants

The rate of Ostwald ripening is 11times10minus

29 and 24times10minus

30 m3sminus

1 at16 and 24 INUTECregSP1 respectively These rates are ~3 orders of

magnitude lower than those obtained using a nonionic surfactant

Addition of 5 glycerol was found to decrease the rate of Ostwald

ripening in some nano-emulsions which may be due to the lower oil

solubility in the waterndashglycerol mixture

Various nano-emulsions with hydrocarbon oils of different

solubility were prepared using INUTECregSP1 Fig 36 shows plots of r 3

versus t for nano-emulsions of the hydrocarbon oils that were stored

at 50 degC It can be seen that both parraf 1047297num liquidum with low and

high viscosity give almost a zero-slope indicating absence of Ostwald-

ripening in this case This is not surprising since both oils have very

low solubility and the hydrophobically modi1047297ed inulin INUTECregSP1

strongly adsorbs at the interface giving high elasticity that reduces

both Ostwald ripening and coalescence

298 T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299

8182019 Polymeric Surfactants in Disperse System

httpslidepdfcomreaderfullpolymeric-surfactants-in-disperse-system 1919

With the more soluble hydrocarbon oils namely isohexadecane

there is an increase in r 3 with time giving a rate of Ostwald ripening of

41times10minus27 m3 sminus1 The rate for this oil is almost three orders of a

magnitude lower than that obtained with a non-ionic surfactant

namely laureth-4 (C12-alkylchain with 4 mol ethylene-oxide) This

clearly shows the effectiveness of INUTECregSP1 in reducing Ostwald-

ripening This reduction can be attributed to the enhancement of the

Gibbs-dilational elasticity which results from the multi-point attach-

ment of the polymeric surfactant with several alkyl-groups to the oil-droplets This results in a reduction of the molecular diffusion of the

oil from the smaller to the larger droplets

References

[1] Tadros Tharwat F Applied Surfactants Germany Wiley-VCH 2005[2] Tadros ThF In Goddard ED Gruber JV editors Principles of Polymer Science and

Technology in Cosmetics and Personal Care NY Marcel Dekker 1999[3] Tadros Tharwat In Holmberg K editor Novel Surfactants NY Marcel Dekker

2003[4] JW Gibbs lsquorsquoScienti1047297c Papersrsquorsquo Longman Green London (1906) M Volmer Kinetik

der Phase Buildung Steinkopf Dresden (1939)[5] Piirma I Polymeric Surfactants Surfactant Science Series No42NY Marcel

Dekker 1992[6] Stevens CV Meriggi A Peristerpoulou M Christov PP Booten K Levecke B et al

Biomacromolecules 200121256[7] Hirst EL McGilvary DI Percival EG J Chem Soc 19501297

[8] Suzuki M In Suzuki M Chatterton NJ editors Science and Technologyof FructansBoca Raton Fl CRC Press 1993 p 21

[9] Flory PJ Principles of Polymer Chemistry NY Cornell University Press 1953[10] Tadros ThF In Buscall R Corner T Stageman editors Polymer Colloids Applied

SciencesLondon Elsevier 1985 p 105[11] Silberberg A J Chem Phys 1968482835[12] CA Hoeve J Polym Sci 197030 361 1971 34 1[13] Roe RJ J Chem Phys 1974604192[14] Scheutjens JMHM Fleer GJ J Phys Chem 1979831919

[15] Scheutjens JMHM Fleer GJ J Phys Chem 198084178[16] Scheutjens JMHM Fleer GJ Adv Colloid Interface Sci 198216341[17] GJ Fleer MA Cohen-Stuart JMHM Scheutjens T Cosgrove and B Vincent

London Chapman and Hall 1993[18] Obey TM Grif 1047297ths PC In Goddard ED Gruber JV editors Principles of Polymer

Science and Technology in Cosmetics and Personal Care NY Marcel Dekker1999

[19] Garvey MJ Tadros ThF Vincent B J Colloid Interface Sci 19744957[20] van den Boomgaard Th King TA Tadros ThF Tang H Vincent B J Colloid Interface

Sci 19786168[21] Napper DH Polymeric Stabilization of Dispersions London Academic Press1983

[22] Flory PJ Krigbaum WR J Chem Phys 1950181086[23] Tadros ThF Vandamme A Levecke B Booten K Stevens CV Adv Colloid InterfaceSci 2004108ndash109207

[24] Exerowa D Gotchev G Kolarev T Khristov Khr Levecke B Tadros T Langmuir2007231711

[25] Tadros ThF Dederen C Taelman MC Cosmet Toilet 199711275[26] Nestor J Esquena J Solans C Levecke B Booten K Tadros ThF Langmuir

2005214837[27] Nestor J Esquena J Solans C Luckham PF Levecke B Tadros ThF J Colloid Interface

Sci 2007311430[28] Liang W Bognolo G Tadros ThF Langmuir 1995112899[29] Nakajima H Tomomossa S Okabe M First Emulsion Conference Paris 1993[30] Nakajima H In Solans C Konieda H editors Industrial Applications of

Microemulsions Marcel Dekker 1997[31] Tadros Tharwat Izquierdo P Esquena J Solans C Adv Colloid Interface Sci

2004108ndash109303[32] Thompson (Lord Kelvin) W Phil Mag 187142448[33] Lifshitz IM Slesov VV Sov Phys 195935331[34] Wagner C Z Electroche 193535581

[35] Kabalnov AS Schukin ED Adv Colloid Interface Sci 19923869[36] Walstra P In Becher P editor Encyclopedia of Emulsion Technology NY Marcel

Dekker 1983[37] Walstra P Smoulders PEA In Binks BP editor Modern Aspects of Emulsion

Science Cambridge The Royal Society of Chemistry 1998[38] Tadros T Vandekerckhove E Vandamme A Levecke B Booten K Cosmet Toilet

2005120(No2)45

299T Tadros Advances in Colloid and Interface Science 147 ndash148 (2009) 281ndash 299