Polymer Brushes: On the Way to Tailor-Made...

1

Abstract

In recent years, the synthesis of polymer brushes through surface-initiated polymer-ization reactions has received significant attention. In this overview, several differentsynthetic strategies for the generation of polymer brushes are reviewed. The uniquephysical properties of polymer brushes that arise from the covalent anchoring of thepolymer chains to the solid substrate are discussed and compared to the propertiesof polymer layers deposited by other techniques of thin film generation. Finally,examples are provided that highlight some recent developments aimed at strategiesfor the functionalization of surfaces with polymer brushes, at ways of realizingsmart surfaces with switchable properties, and at the generation of micro- and nano-structured polymer monolayers.

1Growth of Polymer Molecules at Surfaces: Introductory Remarks

Thin coatings applied to the surface of materials can improve the properties ofobjects dramatically as they allow control of the interaction of a material with itsenvironment. This has been known more or less empirically to man for severalthousand years. Lacquer generated from tree sap was used in China some 7000years ago as a protective coating for wooden objects. Cold process coatings were alsoused around 3000 bc, where Egyptian ship builders used beeswax, gelatin and clayto produce varnishes and enamels and (later) coatings from pitch and balsam towaterproof their ships. The early Greeks and Romans, as well as the ancient Asiancultures in China, Japan and Korea, used lacquers and varnishes applied to homesand ships for decoration and as protective measures against adverse environmentalconditions. In modern times, the coatings industry is a multi-billion dollar businessand – especially if the value of the protected objects is considered – a very importantcontribution to the world economy. Today, however, the application range of coatingsextends much beyond the simple decoration and protection aspects, and functionalcoatings have become an enabling technology in a vast variety of different high-tech

Polymer Brushes: On the Way to Tailor-Made Surfaces

J�rgen R�he

Polymer Brushes. Rigoberto C. Advincula (Ed.)Copyright / 2004 WILEY-VCH Verlag GmbH & Co. KGaA, WeinheimISBN: 3-527-31033-9


areas. Fields in which such high-tech coatings are applied range from computerchips [1] and hard disk manufacturing [2] to the use of special coatings in biomedi-cal and aviation applications [3,4]. Accordingly, many different techniques have beendeveloped for the generation of protective coatings, and these will be discussedfurther below.

Surface-initiated polymerization reactions as a new pathway for the preparationof functional, high-tech coatings have recently received much attention [5,6]. Thistechnique is based on the growth of polymer molecules at the surface of a substratein situ from surface-bound initiators, which results in the attachment of polymermolecules through covalent bonds to this substrate (Figure 1). Polymer layers inwhich the polymer chains are irreversibly attached to the substrate are especiallyattractive for a variety of applications, as such layers can have a good long-term sta-bility, even in rather adverse environments. For example, it poses no problem toexpose surfaces with such surface-attached coatings to good solvents for the poly-mers without being concerned that the polymer will be either dissolved or displaced,and that the coating is more or less rapidly removed from the surface. In addition tothe issue of stability, the number of functional groups present at a surface can alsobe greatly enhanced by connecting large polymer molecules with functional groupsto the surface instead of binding the functional groups directly to that surface. Sucha “skyscraper” approach allows high densities of functional groups to be obtained atthe surface of the substrate through moving from the strictly two-dimensionalarrangement of these groups present in typical surfaces to a more three-dimensionalsituation. An example, which illustrates such a behavior is the attachment of DNAprobe molecules to surface-attached polymer chains, which can significantlyenhance the sensitivity of a DNA-chip (Figure 2).

Systems in which the polymer chains are attached with one end to a solid sub-strate are very interesting, not only from a chemical but also from a physical pointof view. If the grafting density of the polymer molecules is very high, the polymerchains adopt a rather unusual conformation wherein the individual coils overlap[7–9]. Under these conditions, the polymer molecules are strongly stretched away

2

Figure 1 Schematic depiction of the growth of polymer mole-cules at a surface of a solid substrate through surface-initiatedpolymerization.

2 Coatings: From First Principles to High-Tech Applications

from the surface and achieve a molecular shape which is far from the typical ran-dom coil conformation that polymer molecules assume in solution. Such surface-attached films with strongly stretched chains are usually referred to as “polymerbrushes” [10]. Polymer brushes are very interesting systems, as the strong stretchingof the polymer chains leads to concurrent drastic changes in the physical propertiesof the systems. For unstretched polymer chains, a slight molecular deformationleads to a moderate increase of the energy stored in the system (entropy elasticity).However, when the molecules are already strongly stretched – as is in the case of apolymer brush – the energy penalty for the same small deformation is large.Accordingly, in all situations where the stretching of the polymer chains is of con-cern – for example, during the shearing of such surfaces or when the film is pene-trated by other polymer chains from solution – very strong differences can be ob-served to the behavior of free coils [11–13].

Whilst systems in which polymer chains have one end tethered to a substrateappeared some years ago to be quite exotic, and significant doubts persisted thatsuch brushes with high grafting densities could be obtained in practice, the develop-ment of methods where polymers are grown directly on the surface of a substrate byusing surface-initiated polymerization has led to a large number of such systemsbecoming available.

However, before describing more detailed aspects of surface-initiated polymeriza-tion, more general aspects of coatings will be briefly discussed.

2Coatings: From First Principles to High-Tech Applications

For a large number of chemical and physical processes – both in daily life and intechnical applications – the bulk properties of a material as well as the structureand composition of its surfaces determine the performance of the entire system. Inorder to control the interaction of a material with its environment, coatings consist-ing of thin organic films are frequently applied to the surfaces of these solids (Fig-ure 3). In many cases, the coating serves simply as a barrier against a hostile envi-

3

Figure 2 Fluorescence image obtained from a DNA chip basedon a oligonucleotide functionalized polymer brush. The patternand the intensity of the spots allows for the determination of thesequence of the unknown analyte-DNA.


ronment and allows for protection against corrosion or other chemical or photo-chemical degradation. Although corrosion protection is certainly the most promi-nent aspect of surface coatings as far as market and materials volumes are con-cerned, thin organic coatings are also applied in a large number of high-tech appli-cations, ranging form microelectronics [1] to biomedical devices [3,4].

When considering such applications, thin organic coatings are applied to controlthe interactions between the material and its environment. Examples of interfaceproperties which can be controlled by deposition of a thin organic film onto a sur-face include friction [11,13–17], adhesion, adsorption of molecules from the sur-rounding environment, or wetting with water or other liquids. In medical applica-tions, coatings allow control of the interaction of biological cells and biomoleculeswith artificial materials in order to enhance the biocompatibility of an implant, or toavoid the nonspecific adsorption of proteins onto the active surfaces of an analyticaldevice [18].

It is well known that coatings, even when only a few Angstroms thick, can influ-ence the surface properties of a material so strongly that the chemical nature of theunderlying material becomes completely hidden and the interaction of the wholesystem with the surrounding environment is governed by these extremely thin coat-ings (“stealth effect”). This is an advantageous situation for materials engineering asit allows optimization of the bulk and surface properties of a material separatelyfrom each other. In addition, the application of functional coatings allows the cover-age of a surface with groups which interact with other molecules in their environ-ment through specific molecular recognition processes. Such a strategy is, for exam-ple, very important for the control of the adhesion of biological cells to artificial sub-strates. In such a case, thin layers containing cell recognition peptide sequences caninduce strong adhesion of the cells to the substrate surfaces, to which they otherwisewould show only a very unfavorable adhesion behavior [19].

One example of a system where the covering of a surface with an ultrathin coat-ing is a prerequisite for that system to function is a computer hard disk [2] (Figure4). If the uncoated surface of a thin film magnetic disk is subjected to strong shear,such as the sliding of a read/write head on the disk surface, then almost instanta-neous damage can be observed. The disk shows, even upon the first contact with thehead, a strong stick-slip behavior and a high friction coefficient, while the debris

4

Figure 3 Schematic depiction of use of thin polymer coatingsto control the interaction of a material with the surroundingenvironment.

2 Coatings: From First Principles to High-Tech Applications

generated by this damage leads to rapid failure of the disk. However, if a film of aperfluoropolyether of typically only 2–4 nm thickness is attached to the disk, the tri-bological properties are greatly improved, the wear is reduced, and the mean time tofailure of the disk is greatly prolonged (Figure 4).

A second example where ultrathin organic coatings control the performance ofthe whole system is the control of interface properties of materials in contact withblood. If artificial materials are brought into contact with blood, then blood proteinssuch as fibrinogen adsorb very rapidly to the surfaces of the implant or sensor sur-face, followed by the adhesion of blood cells to these protein layers. This reactioncascade leads almost immediately to strong changes of the surface composition ofthe active surfaces of the sensor or implant. After a short period of time, blood clots

5

a) b)

Figure 4 (a) Computer hard disks are protectedagainst mechanical wear by ultrathin layers of per-fluorinated polymers. (b) Hard disk in an acceler-ated wear test: (i) unlubricated and (ii) after appli-cation of 1.5 nm chemisorbed and 1 nm physi-sorbed lubricant; the high friction coefficient and

the strong noise indicate a strong stick-slipbehavior, which is the beginning of a catastrophicfailure of the system. (Reprinted with kind permis-sion from Ref. [2]; .American Society of Mechani-cal Engineering, 1996.)

Figure 5 SEM image of a fibrinnetwork and thrombocyteson the surface of an artificalheart value (picture courtesy ofDr. A. Schlitt, University Hospi-tal Mainz, Germany).


become attached to the surfaces of the blood-exposed materials (Figure 5). Even-tually, the blood clots can break off from the surface into the blood stream, wherethey pose a life-threatening situation for the patient [18]. It has been shown, that theapplication of just a polymer layer which is just a couple of nanometers thick candramatically reduce the adhesion of the blood proteins and thereby greatly improvethe blood compatibility of the material [20–22].

3Surface-Coating Techniques

Depending on the type of interaction between the molecules which are constituentsof the coating and the substrate which is to be modified, two classes of strategies forthe deposition of thin organic coatings can be distinguished. In one of these, themolecules interact with the substrate by physical forces [7–9], whilst the other classconsists of molecules which are attached to the surfaces through chemical bonds. Inthe latter case, a monomolecular layer or a surface-attached network is very strongly(“irreversibly”) attached to the surface. This classification is not simply a formality,but the type – and accordingly the strength of interaction – also has a very stronginfluence on the physical properties of the coating, the film thicknesses which canbe obtained through such a method, and the long-term stability of the coating inproblematic environments.

A number of technologically important coating techniques rely on physical inter-actions between the deposited molecules and the substrate, including:

. painting/droplet evaporation

. spray coating

. spin coating

. dip coating

. doctor blading

Although being quite different in detail, a common feature of all of these pro-cesses is that the molecules are deposited from solution and the solvent evaporatesduring the coating process (Figure 6). The techniques described above are somewhatempirical in nature, as certain parameters such as the rate of evaporation of the sol-vent depend on specific details of the individual process and are accordingly difficultto predict a priori, but in many cases are simple to reproduce. Accordingly if thedeposition conditions are properly controlled, layers with well-defined thickness andgood homogeneity can be generated without major effort. Several of these processes,such as dip- and spin-coating, allow the deposition of extremely thin film coatings(starting from just a few nanometers thickness), but essentially no upper limit tofilm thickness exists, if appropriate conditions are applied.

In contrast to these rather empirical processes, more sophisticated coating tech-niques have been developed, including the Langmuir-Blodgett technique [23], theadsorption of monomolecular layers of homo- and block copolymers [7] from solu-tion, and the Layer-by-Layer (LbL) [24] technique in which multilayer stacks of oppo-

6

3 Surface-Coating Techniques

sitely charged polyelectrolytes are deposited onto a (charged) substrate. These tech-niques allow for much better control of the internal structure of the deposited layers,and also for extremely high precision with regard to the thickness of the coatings.All of the coating techniques, except perhaps for the Langmuir-Blodgett technique,are – from a technological viewpoint – rather simple, and the generation of layerstypically requires no complicated set-ups to generate the coatings. The moleculesare attached to their substrates by physical interactions, and consequently the forcesholding them at the surface are rather weak. In some cases this situation is desir-able, but in others it becomes problematic as it is more likely to lead to adhesive fail-ure of the system. Under unfavorable conditions, the films can be subject to destruc-tion by the “Big Four Ds”:

7

a)

b)

c)

Figure 6 Schematic illustration of different processes used forthe deposition of organic molecules and/or polymers on sur-faces: (a) spin-coating. (b) Langmuir-Blodgett-Kuhn (LBK) tech-nique; (c) adsorption from solution.


. desorption during solvent exposure;

. displacement by molecules which have stronger interaction with the surface;

. dewetting (for films above the glass transition temperature, Tg); and

. delamination (for films below Tg).

Desorption and displacement are especially important, as coatings are usually notprepared and kept under ideal (i.e., ultrahigh vacuum) conditions, but rather areexposed to environments containing all sorts of contaminants. In these “real-life”environments, contaminants are present on every surface, and/or competing adsor-bates will fight for surface sites during or after the coating process. Examples of mol-ecules which are present in many different environments, and which typically com-pete quite efficiently with coating materials for surface sites, include water, ions,polyelectrolyte molecules, or oils. The contaminants or displacing agents mighthave such a strong interaction with the surface that the molecules of the coating canno longer remain in contact with the substrate, but eventually will instead be locatedon top of a thin layer consisting of contaminant/displacer molecules. In this respect,polar surfaces which absorb ambient water are especially problematic, as water hasa strong affinity for such surfaces and exhibits a very high adsorption enthalpy.Accordingly, water functions as a very efficient competitor for surface sites and easi-ly displaces adsorbed molecules from such high-energy surfaces. What renders thesituation even worse is that under these conditions the surface properties of thematerial become strongly dependent on the history of the sample – that is, whichenvironment the sample has been exposed before use – and this is, potentially, avery problematic situation.Dewetting occurs in all systems, where the surface tension of the substrate is

lower than that of the coating material if the molecules are allowed to reach an equi-librium. This may be achieved either by heating the film above Tg, or by exposure tomolecules which can act as a plasticizer for the polymers of the coating.

In contrast, delamination occurs if the films are in the glassy state and subjectedto wide temperature swings, or if the coating swells in the environment to which ithad been exposed, while the substrate does not swell. In such cases, strong mechan-ical stress develops at the interface, and this may cause the entire film to peel off,leading to large-scale adhesive failure.

An alternative to the above-mentioned procedures which allows improvement inthe long-term stability of coatings even in very adverse environments, is to attachthe molecules of the coating to the surface of the substrate through chemical bonds.The price which must be paid for an enhanced stability of the system is a more com-plicated coating procedure and/or the requirement to choose the coating conditionsmore carefully, so that the surface reaction proceeds in high yield and with limitedside reactions. A current, very frequently employed strategy for the preparation ofwell-controlled surface layers is the use of small molecules with a reactive headgroup that is amenable to form a covalent bond with a corresponding chemical moi-ety on the surface of the substrate, which is to be modified. As this process is self-limiting – that is, the surface-attachment reaction stops when all the reactive surfacegroups have been consumed or are no longer accessible – such layers are commonly

8

3 Surface-Coating Techniques

called self-assembled monolayers (SAM) [25]. Examples are silanes on oxide sur-faces, phosphates or phosphonate on metal(oxide)s, and thiols or disulfides on noblemetal surfaces (Figure 7).

In this way, surface coatings can be obtained which are very stable and may evenhave a strong degree of positional and orientational order. In some cases, even crys-talline packing of the surface-attached molecules has been observed. If moleculesare assembled that carry at their tail end a specific chemical moiety or a biochemi-cally active group, it is possible to obtain a more or less strict 2D arrangement ofthese functionalities (Figure 8) [26]. Examples are molecules which contain fluoro-carbon segments in the assembling units [27–29], and can convert a hydrophilic sur-face into a highly water-repellent hydrophobic one, or the introduction of “ligands”as recognition sites in bio-affinity assays. In this way, surfaces can be generated –for example, on top of the transducer of a biosensor – that very specifically bind pro-teins from solution [30,31].

9

Figure 7 Schematic of the self-assembly process and examplesof anchor groups used for the modification of surfaces with self-assembled monolayers (SAMs) of organic molecules.

Figure 8 Example of a structure prepared viasoft lithography from acid and methyl-termi-nated thiols. First, the Me-terminated thiol wasstamped onto a gold surface; second, the

unmodified areas were backfilled from a solu-tion containing the acid-terminated thiol.(Reprinted with kind permission from Ref. [26];.Wiley-VCH, 1998.)


In some of these applications the intrinsic limitations of this strictly 2D arrange-ment of the functional groups are evident: the maximal surface density of the func-tional moieties is limited by the surface area cross-section of the assembled unit. Insome cases it is even lower than the arrangement of the individual functional unitsat such high packing densities in some cases leads to a mutual blocking or, at least,to a limited accessibility. One obvious solution to the above problem is the extensioninto the third dimension – that is, the use of polymers carrying the functionalgroups along the chain, thus generating higher cross-sectional densities of thesegroups and simultaneously guaranteeing good accessibility.

4Surface-Attached Polymers

Most approaches which aim at attaching polymers to a surface use a system wherethe polymer carries an “anchor” group either as an end group or in a side chain.This anchor group can be reacted with appropriate sites at the substrate surface,thus yielding surface-attached monolayers of polymer molecules (termed “graftingto”) (Figure 9) [32–37]. While the attachment of terminally functionalized polymersto the surface leads to layers, where one group is connected to the surface, side chainattachment usually leads to multiple attachment points and, accordingly, a ratherflat conformation of the polymer molecules. In the latter case, the functionalgroups of different molecules compete for reactive sites on the surface, andaccordingly the amount of polymer which can be immobilized depends strongly onthe reaction conditions, and especially on the concentration of the polymer insolution.This chemical linking of polymers to a substrate surface is, in principle, closely

related to the formation of self-assembled monolayers of low molecular-weight com-pounds described above. Accordingly, if such (end)functionalized polymers are avail-

10

a) b) c)

Figure 9 Schematic illustration of different processes usedfor the attachment of polymers to surfaces: (a) “grafting to”;(b) grafting via incorporation of surface-bound monomericunits; (c) “grafting from/surface-initiated polymerization”.

4 Surface-Attached Polymers

able (which is a nontrivial condition, as the synthesis of polymers with reactive endgroups is far from being trivial), the attachment of the polymers is, from a chemicalpoint of view, rather simple.

Another straightforward technique for the attachment of polymers to surfaceswhich allows the generation of a great variety of functional surfaces is to carry out apolymerization reaction in the presence of a substrate onto which monomers hadbeen attached [32,38–41]. In such a polymerization reaction, the surface-attachedmonomers are incorporated into growing polymer chains in the very same way astheir peers in solution (Figure 9). However, once one or more surface-attachedmonomers are incorporated, the polymer is “glued” firmly to the surface. Duringthe process, a macroradical initially attacks the monomers on the surface, while in asecond step further monomers units are added, so that the chain grows again, awayfrom the surface. However, careful studies of the polymerization mechanism haveshown that the “grafting to” step represents the bottle-neck of the reaction and thuslimits the polymer immobilization [42,43]. Accordingly, very similar layers areobtained by using immobilized monomers as by the chemisorption of preformedchains. A general limitation of the technique is that the substrate must be immersedin a polymerization solution, but if this poses no problem it is one of the simplesttechniques to generate surface-attached layers, especially as there is no need tosynthesize a polymer functionalized with an anchor group.

Although “grafting to” reactions are easy to perform, it should be noted that cer-tain rather strict limitations apply to the structures which can be realized by use of a“grafting to” strategy. First, the use of reactive anchor groups for the surface-attach-ment of polymers imposes some rather strict limitations on the choice of functionalgroups available for incorporation into the polymer. One of the reasons for this isthat the functional groups on the polymer can compete with the anchor moieties forsurface sites. Especially if the aim is to immobilize functional polymers containinghighly polar or charged groups onto polar surfaces, the adsorption of functionalgroups to the surface can be very strong and compete very effectively with the chem-isorption process. Such competition between anchor and functional groups hasbeen observed, for example, in the case of the attachment of a low molecular-weightalkoxysilane containing amine groups to a silicon oxide surface [44–46]. In such asystem, interactions between the basic amine groups of the SAM-forming silaneand the rather acidic silanol groups of the silicon (oxide) substrate can strongly com-pete with the condensation reaction of the alkoxysilyl moiety with the substrate sila-nol groups. As result, layers are obtained which contain both physisorbed moleculesdue to acid–base interactions and chemically attached molecules.

Second, in order to obtain a fast and complete surface attachment reaction with ahigh surface density of chains covalently bound to the substrate, rather reactiveanchor groups are required. These groups, however, tend not to tolerate the simulta-neous presence of a large variety of functional groups in the polymer. For example,if a chlorosilyl group is chosen as an anchor group for the attachment of the poly-mer to a silicon oxide surface, this choice excludes the incorporation of many func-tional groups into the polymer, including amine-, hydroxyl- or carboxylic acid moi-eties, as these would react with the chlorosilyl groups.

11


At first view, there is a tendency to consider that resorting to less reactive anchorgroups – for example, using a less reactive alkoxysilane instead of a highly reactivechlorosilane – would solve the problem. This, however, is incorrect as a more in-depth analysis shows. If the nucleophilicity of the anchor group is reduced, thisaffects both the undesired side reaction, which is the reaction of the functionalgroup with the anchor group, and which leads to loss of anchor moieties, and thedesired reaction of the anchor group with a group at the surface of the substrate,which results in a successful chemisorption reaction. Accordingly, both reactionsare slowed down at the same time, and the ratio between the rates of the two reac-tions remain the same in all cases.

Another complication inherent to “grafting to” processes is an intrinsic limitationof the film thickness, and accordingly the number of functional groups per surfacearea which can be obtained by using such an approach. Films generated by chemi-sorption from solution are limited to (dry) thicknesses of typically 1 to 5 nm. Thislimitation has both kinetic and thermodynamic origins. With increasing coverage ofthe surface with attached chains, the polymer concentration at the interface quicklybecomes larger than the concentration of polymers in solution. Additional chains,which are to become attached to the surface, must diffuse against this concentrationgradient that ever increases with increasing grafting density of the attached polymer(Figure 10). This diffusion slows down the immobilization reaction at the surfacefurther and further as the reaction proceeds. Thus, the rate of the attachment reac-tion levels off rather quickly and further polymer is linked to the substrate only at anextremely slow rate due to this kinetic hindrance. Indeed, it has been shown [42,43]both theoretically and experimentally that once the surface-attached coils overlap,the attachment of further polymer molecules takes place on a logarithmic timescale, and already at rather low graft density time frames of thousands or even mil-lions of years would be required to add a few more nanometers of polymer to thelayer. Accordingly, as far as practical reaction times are concerned, films generatedby this technique are intrinsically limited with regard to the film thickness. Further-

12

a) b)

Figure 10 Schematic illustration of the “graft-ing to” process. (a) Chains that are to beattached to the surface can easily reach the sur-face at low graft densities. (b) The attachmentprocess comes to a virtual halt as soon as the

surface is covered with polymers, as thealready attached chains form a kinetic barrieragainst which incoming chains have to diffuseto reach the surface.

5 Polymer Brushes: General Features

more it should be noted that, even if this kinetic limitation is somehow circum-vented, the attachment of chains to a strongly covered surface becomes unfavorablealso for thermodynamic reasons. At high grafting densities the surface-attachedpolymer chains are in a rather stretched conformation due to the presence of strongsegment–segment interactions, as will be discussed in more detail below. A chain,which is now becoming attached to the surface, must change from a coil conforma-tion in solution to a stretched (“brush-like”) conformation at the surface. Theentropy loss during this process, however, is only compensated by the establishmentof one chemical bond, namely the one connecting the polymer to the surface.Hence, the higher the graft density of the chains at the surface, the stronger will bethe entropy penalty, and this rapidly precludes the attachment of further chains.

5Polymer Brushes: General Features

As mentioned briefly above, the term “polymer brush” refers to a system in whichchains of polymer molecules are attached with one or with a few anchor points to asurface in such a way that the graft density of the polymers is high enough that the

13

a) b)

c)

Figure 11 Artist’s perception of the terms (a) “mushroom”,(b) “pancake” and (c) “brush” used for the different possibleconformations of surface-attached polymers.


surface-attached chains become crowded and are stretched away from the surface(Figure 11). From the stretching of the polymer chains perpendicular to the surface,several new physical phenomena arise. Examples are ultralow friction surfaces[11,12] obtained through coating of two surfaces that slide against each other withpolymer brushes, or the so-called autophobic behavior [47–50], in which materialscoated with surface-attached polymer chains do not become wetted by free polymer,even if the surface-attached and the free chains are chemically identical (Figure 12).

In the following discussion, the focus will be placed on polymer brushes at solidsurfaces, although brush-like chain conformations can also be obtained at theboundary between phases in block copolymers [8] or in so-called molecular “bottle-brushes” [51–53]. In the latter system, polymers are attached as side chains to thebackbone of a polymer molecule, so that every segment of the backbone carries sucha polymeric side chain. Although the overall physical picture for the different sys-tems is very similar, here only chains attached to solid surfaces at one end will bedescribed and discussed.

When polymer molecules are tethered to a surface, two basic cases must be distin-guished depending on the graft density of the attached chains [8–10]:

1. If the distance between two anchoring sites is larger than the size of the sur-face-attached polymers, the segments of the individual chains do not “feel”each other and behave more or less like single chains “nailed” down onto thesurface by one end. Depending on the strength of interaction of the polymersegments with the surface, again two cases must be distinguished [10]. If theinteraction between the polymer and the surface is weak (or even repulsive),the chains form a typical random coil that is linked to the surface through a“stem” of varying size. For such a situation, the term “mushroom” conforma-tion has been coined (Figure 11). However, if the segments of the surface-attached chains adsorb strongly to the underlying surface, the polymer mole-cules obtain a flat, “pancake”-like conformation (Figure 11).

2. A completely different picture is obtained if the chains are attached to thesurface at such short distances between the anchor points that the polymermolecules overlap. In this case, the segments of the chains try to avoid eachother as much as possible and minimize segment–segment interactions by

14

Figure 12 Optical micrograph of adewetted polystyrene layer (initial thickness60 nm) on top of a polystyrene brush(6 nm; prepared via grafting from). Thispicture was taken after annealing the samplefor 40 h at 180 CC (scale bar = 200 lm).(Reprinted from Ref. [50], with kind permis-sion; .American Chemical Society, 1996.)

6 Theory of Polymer Brushes

stretching away from the surface (Figure 11). This chain stretching, however,reduces the number of possible polymer conformations, which is equivalentto a reduction in the entropy of the chains. This loss of entropy gives rise to aretracting force trying to keep the chains coiled, as occurs in a stretched pieceof rubber. Thus, a new equilibrium at a higher energy level is obtained inwhich the chains are stretched perpendicular to the surface.

6Theory of Polymer Brushes

The theoretical description of polymer brushes attached to surfaces of differenttopologies – that is, planar and curved surfaces – is well developed [7–9]. However,as in this book the main focus is set on new developments concerning the chemicalmethodology, only a very brief outline of the theory of brushes is provided here. Fora more detailed discussion, the reader is referred to reviews recently published onthis subject [7–9].The key idea behind the theoretical description of polymer brushes is that the free

energy F of the chains is obtained from a balance between the interaction energybetween the statistical segments Fint and energy difference between stretched andunstretched polymer chains Fel (elastic free energy) caused by the entropy loss of thechains:

F = Fint + Fel (1)

The most important parameters, which are of interest for a description of brush sys-tems, are the segment density profile (u(z)) of the surface-attached chains and/orthe brush height h as a function of the graft density r, the molecular weight (/degreeof polymerization) of the surface-attached chains, and the solvent quality of the con-tacting medium (Figure 13).The first description of such a brush system has been attempted by Alexander [54]

for monodisperse chains consisting of N segments, which are attached to a flat,non-adsorbing surface with an average distance of the anchor points dmuch smaller

15

Figure 13 Two hundred chains of a polymer brush (chainlength N = 100) under good solvent conditions. (Reproducedwith kind permission from Ref. [11]; .Springer, 1998.)


than the radius of gyration of the same unperturbed chains not in contact with thesurface (Figure 14). If both the interaction energy resulting from binary monomer–monomer interactions and the elastic energy of a Gaussian chain are calculated andminimized in respect to the brush height h, the following equation is obtained forbrushes in a good solvent:

h ~ N N r1/3 (2)

In a poor solvent – that is, close to H conditions – the exponent describing theinfluence of the grafting density is slightly different and

h ~ N N r1/2 (3)

is obtained. It should be noted, that in both cases the brush height scales linearlywith the degree of polymerization/molecular weight of the polymer molecules,which is a much stronger dependency than that of the size of a polymer coil in solu-

16

Figure 14 Schematic illustration of the Alexandermodel for the theoretic description of polymerbrushes. The chain segments with the “blobs”(indicated by the circles) behave as random(“Gaussian”) coils. (d represents the average dis-tance between anchor points.)

Figure 15 Schematic illustration of segment density profiles forsurface-attached polymers in different regimes. For details, seethe text (adopted from [9]).

6 Theory of Polymer Brushes

tion on the molecular weight, where the radius of gyration Rg, scales with Rg ~ N0.59

for a polymer in a good solvent and Rg ~ N0.50 for solutions close to H conditions.

Although the Alexander model is very simple, it predicts the experimentally ob-served scaling behavior more or less correctly and allows an understanding of someof the most striking properties of polymer brushes, such as lubrication and the wet-ting behavior. More sophisticated models have been developed to describe the seg-ment density profile of the brushes (Figure 15). To this numerical and analyticalself-consistent field (SCF), theories [55–57] for such systems have been proposedbased on the assumptions that, for strong stretching and high molecular weights ofthe brushes, fluctuations around the most favorable configuration of the polymerchain diminish. A general result of the SCF calculations is, that the segment densityprofile is more or less parabolic as long as the grafting density is moderate and themolecular weight of the brush chains is high. At very high grafting densities theSCF assumptions are no longer valid, as three body interactions between the poly-mer segments become significant. The results of the SCF calculations have beenverified both experimentally and in simulations. For the latter, molecular dynamicsand Monte Carlo methods have been employed [58]. If smaller differences in thenumerical coefficient are neglected, then the SCF results are in good agreementwith the results from simple scaling arguments.

In addition to these somewhat straightforward calculations, more complicated sit-uations have also been tackled where the polymer chains have a distinct polydisper-sity [59], are in specific topologies such as attached to small particles [60], which ex-hibit a significant curvature also on the molecular scale, and to brushes which carrycharges along the polymer chain [61] (Figure 16). In particular, the latter case canbecome very complicated if the polymer chains interact specifically with ions in thesurrounding medium, as under these circumstances the situation can no longer bedescribed by simple mean field approaches, but specific complex formation and(local) changes in the solubility of the polymer play a key role in describing theswelling behavior of such brushes.

17

Figure 16 Schematic illustration of a polyelectrolyte brush (PEL brush).


7Synthesis of Polymer Brushes

An obvious requirement for forcing polymer molecules into brush-like conforma-tions is that the strength of anchoring of the molecules to the interface is sufficientlyhigh that the molecules are connected irreversibly to the surface of the substrate. Asecond requirement is that the synthetic strategy allows for the generation of graft-ing densities high enough to cause sufficient repulsive segment–segment interac-tions within the surface-attached chains to induce significant chain stretching. Inparticular, the latter condition imposes some strict limitations onto the appropriatesynthetic strategy for brush formation as the chains lose a considerable amount ofentropy when stretched into an elongated form.

In the following section, four different approaches to reach these goals will bebriefly discussed. A complete review of the published literature on this subjectwould clearly be beyond the limits of this introductory chapter.

Approach 1

In the first approach, amphiphilic block copolymers consisting of a water-solubleblock and a water-insoluble block are spread at the air-water interface [62,63]. Thewater-soluble block attempts to dissolve into the aqueous subphase, but is anchoredto the air-water interface by the hydrophobic block. Upon compression of the thusobtained Langmuir monolayer, the distance between the anchor points of the poly-mer chains decreases and the hydrophilic block is stretched away from the surfaceinto the aqueous subphase. A prerequisite for this is that the hydrophobic block is inthe molten state, because only is it possible for a rearrangement of the chains withinthe film to occur upon compression. Furthermore, it is important that the hydro-philic balance is chosen in such a way that the loss of chains to the subphase andthe formation of micelles can be avoided. The thus obtained films can be crosslinkedthrough photochemical reactions and transferred to a solid substrate.

Approach 2

In the second case, block copolymers or end-functionalized polymers are physi-sorbed to a solid surface [7]. End-functionalized polymers can be discussed togetherwith block copolymers as they are structurally very similar to such systems in termsof their essential physics of adsorption to a solid surface. In some ways they can beviewed as block copolymers with a very short block, consisting only of one unit. Inthe block copolymer concept, one block adsorbs strongly at the surface and acts asan anchor for the polymer chains. The other block adsorbs only weakly at the sur-face – that is, the interactions of the polymer with the solvent are stronger thanthose with the surface – and so the block floats in the solvent like a buoy. Althoughduring the past, many different polymer layers have been prepared by this route, the

18

7 Synthesis of Polymer Brushes

chemical variability of these systems is somewhat limited as a solvent must be avail-able in which the block copolymer adsorbs to the surface without formation ofmicelles either in solution or at the surface. Furthermore, as the layer formationrequires the diffusion of polymer molecules through the layer of already attachedchains, this limits the range of graft densities that can be obtained using this tech-nique. In addition, as the interaction with the surface is based simply on physicalinteractions, anchoring of the molecules to the substrate surface is relatively weak,and this further limits the graft densities available and decreases the stability of thefilms.

Approach 3

As has been discussed above, the chemisorption of polymer molecules leads tochains which are covalently attached to surfaces [32–37]. Although situations can beenvisioned in which the polymer chains are slightly stretched, such processes arestrongly limited in terms of the obtainable graft density, especially for high molecu-lar-weight polymers, and this results in only relatively weak stretched polymerchains.

Approach 4

Much higher graft densities can be obtained when the polymer chains are grown atthe surface of the substrate in situ (Figure 17) [5,6]. To this initiator, species areeither generated or self-assembled at the surface of the substrate, followed by initia-tion of chain growth from these surface-attached initiators, for example by con-trolled or free radical chain polymerization. The surface-polymerization can bestarted thermally either through a chemical process or photochemically. In this way,polymer monolayers with film thicknesses of more than 2000 nm in the dry statehave been obtained (Figure 18). In this case, polymer molecules with number aver-

19

Figure 17 Common synthetic strategy for the gen-eration of polymer brushes via surface-initiated poly-merization. An initiator molecule is deposited on asurface by means of a self-assembly process via thereaction of an anchor group to suitable surface sitesand, subsequently, chains are grown on the surfacefrom the initiating sites.


age molecular weights of several 106 g mol–1 are attached at distances of anchorpoints of less than 3 nm.

Surface-initiated polymerization reactions work for any polymer which can beobtained by a chain growth reaction such as free and controlled radical polymerization,carbocationic polymerization, anionic polymerization, and ring-openingmetathesis po-lymerization (Table 1) [63–97]. The different polymerization reactions can be carried outon surfaces of very different topologies (planar, curved, and irregular surfaces), andallow for the generation of polymers from a wide spectrum of different monomers.

It would be far beyond the scope of this overview to try to review all recent devel-opments on the synthesis of such systems, and a large variety of different syntheticroutes for the generation of polymer brushes through surface-initiated polymeriza-tions will be detailed in the following chapters. However, at this point some com-ments should be made on controlled or living polymerization reactions for thegrowth of polymer molecules through surface-attached initiators. In this respect, liv-

20

a)

b)

Figure 18 (a) Optical waveguide spectrum(symbols) obtained from a PMMA brushdeposited on an evaporated SiO2 layer. Thesolid line was obtained from model calcula-tions based on a Fresnel formalism assuminga 2200 nm-thick polymer layer. The sample was

prepared in neat MMA at 50 CC, polymerizationtime: 96 h. (b) Thickness of PMMA brushes asa function of monomer concentration; poly-merizations were carried out at 60 CC for 18 hin toluene as a solvent (if required).

7 Synthesis of Polymer Brushes 21

Table 1 Selected systems for the generation of polymer brushes via surface-initiatedpolymerization. The list is by no means exhaustive, and is only meant to demonstratethe wide variety of synthetic strategies that have been developed over the past decade.

Mechanism Initiator/initiating species Maximum thickness(nm)

Reference(s)

Free radicalSi

Me

Me

ON

H

O

NN

Me

Me

CN

CN

CO2H ~ 100 nm 63,64

O Si

Me

Me

O

O

CN

MeNNMeMe

CN

S O

O

CN

MeNNMeMe

CN

(CH2)11

Si

Me

Me

NN

CN

CNMe

O

up to 2200 nm 65–72

O

O

O

O

OH

n.a. 73,74

TEMPO

Si

Me

Me

OO

ON

120 nm 75,76

ATRP Si (CH2)n O

O

Br150 nm;700 nm(water accelerated)

77–83

SiO (CH2)2 SO2Cl 100 nm 84–86

Si

Me

Me

O (CH2)2 CH2Cl n.a. 87–89

Si

Me

Me

OO

O

Me

Br< 60 nm 82,90

Others Various systems for cationic andanionic polymerizations, RAFTand reverse ATRP

< 40 nm 91–97

NA= not applicable.


ing systems with rapid initiation are of major interest as they allow, in principle, sur-face-attached polymer chains with relatively narrow molecular weight distributionsto be obtained. This facilitates comparison with theoretical models developed forsurface-attached polymer brushes, provided that the initiation process is sufficientlyefficient to allow high graft densities and that the molecular weight of the surface-attached chains is high enough to allow such a discussion. Indeed, controlled poly-merization approaches are expected to become even more interesting for the synthe-sis of surface-attached polymer brushes, as a large variety of functional brushes canalso be obtained by using these methods. At present, major efforts are made – espe-cially in the area of controlled radical polymerization – to polymerize functionalizedmonomers to create high molecular-weight compounds with low polydispersity.

8Polymer Brushes as Functional Materials

For many applications of polymer brushes, it is not simply protection against me-chanical or chemical damage that is important. Rather, where the polymer layer acts

22

Figure 19 Examples of functional groups incorporated into polymer brushes.

8 Polymer Brushes as Functional Materials

as a barrier against contact with the environment, a more specific chemical responseto the surrounding medium is desirable. Examples of this situation include layersinto which DNA, protein molecules or complexing agents – each of which shows aspecific reaction towards certain metals – are chemically incorporated [99]. To thisend, polymers with desired functional groups can be formed directly from the corre-sponding monomers (Figure 19). For example, brushes carrying either charges(“polyelectrolyte brushes”) [71,74,75,100–102] or pendant mesogenic units (“LC-brushes”) [103,104] have been prepared using this direct route. An alternative wouldbe first to generate a brush from a simple and inexpensive precursor monomer con-taining a reactive group, and this can then be transformed into the final moietythrough a polymer analogous reaction. Examples of such compounds are monomerscarrying an active ester, epoxide, azalactone or amine groups [99].

It is quite evident that, in principle, the direct approach is much simpler as thedesired brush can be prepared in a one-step reaction. However, this places somerather stringent requirements on the availability of the monomer, because if anincorporation of repeat units with especially valuable groups into the polymer isdesired, then the amount of the valuable monomer needed for the brush generationis rather large. The reason for this is that the molecular weight of the brushes is, formost polymerization mechanisms, directly connected to the monomer concentra-tion; consequently, if high molecular-weight polymers are desired, then relativelylarge amounts of monomer are required. A second requirement is that the func-tional group is compatible with the polymerization process used for brush forma-tion. Monomers containing moieties that show excessive transfer properties such assulfur groups cannot be used in direct polymerization processes as they would leadto side reaction and/or only low molecular-weight brushes. This is especially impor-tant, as for a surface-initiated polymerization reaction any chain transfer is equiva-lent to a termination reaction, because after the transfer further polymer is only gen-erated in solution and removed in a subsequent extraction of the film.

In addition to this, a two-step pathway for the generation of functional brusheshas the advantage that it is not necessary to study the polymerization behavior ofeach new monomer with a new functional group “from scratch” because a numberof different functionalities can be incorporated using the same precursor monomer.Examples are brushes of homo- or copolymers with N-hydroxysuccinimide ester orepoxide groups through which a large variety of different functionalities can beintroduced by aminolysis. For example, the preparation of brushes that carry thiol,pyrene, oligoethyleneoxide or bioactive groups such as peptides or oligonucleotideunits have been reported using the same precursor monomer. If the direct polymer-ization procedure is applied, then each and every one of these monomers must bestudied with regard to the polymerization kinetics in order to obtain an in-depthunderstanding of the brush-forming properties.The use of “living” polymerization reactions – that is, reactions where the num-

ber of active or dormant and thus potentially active species remains more or lessconstant on the time scale of the polymerization reaction – allows the generation ofbrushes which carry at the end pointing away from the surface a functional group,or brushes which consist of a copolymer [96,98,105,106]. The latter constitute a very

23


interesting system, as all polymer molecules are surface-attached and accordinglylarge-scale, irreversible reorganizations of the chains are prohibited, and the mor-phology of the polymer film is directly coupled to the composition of the copolymerbrush. Thus, upon exposure to an environment – which is selective for one of thetwo components – the morphology of the layers of such copolymer brushes can beeasily switched from one morphology to the other, and monolayers with very unusu-al topographies can be obtained.

Another interesting system is generated if not all of the initiator is used up duringthe polymerization reaction, or if two different initiators are co-immobilized on thesubstrate surface. In such a case, after completion of the growth of one polymer spe-cies, some initiator is still present which can be used to kick-off the polymerizationof another monomer [105,107,108]. This then results in the growth of a second typeof polymer in direct neighborhood to the chains already attached to the surface.Such systems – which commonly are called “mixed brushes” – seem especiallyattractive as the two polymers can have very different interaction strengths with thesurrounding of the film. This situation is very similar to that of block copolymerbrushes described above. If one environment strongly prefers one polymer over theother, whilst a second environment favors the reverse situation, surfaces withswitchable surface chemistries are generated. When the polymer layer is alternatelyexposed to one or the other environment, the internal structure of the polymerchanges accordingly and a system that can adapt to the substrate environment(“smart surfaces”) is obtained.

9Microstructured Polymer Brushes

The (micro-)patterning of polymer brushes is especially interesting as all the poly-mer molecules are permanently attached to the surface [73,109]. This is an impor-tant aspect, both for the generation of the patterns as well as for applications of themicrostructured surfaces, as it allows exposure of the microstructures to good sol-vents for the polymers. The latter aspect is especially important for biological appli-cations, as it allows strong swelling of the brush and provides a soft “cushion” forthe biological system at the surface of the substrate. This is of special significance asproteins tend to denature in contact with hard, solid surfaces. Also, from the view-point of preparing microstructured systems, the generation of thick, surface-attached monolayers is rather attractive as it allows the washing away of reagentsafter completion of a chemical reaction in the patterned structures, and hence thegeneration of multifunctional chemical patterns with high resolution (Figures 20and 21). Indeed, in addition to simple chemical structures being “written” into thefilm, the use of step-and-repeat procedures allows the generation of very compli-cated chemical surfaces and structures. This contrasts strongly with the conven-tional lithographic procedures used in the semiconductor industry where, upon irra-diation and solvent exposure, a relief is generated and hence topological rather thanchemical structures are generated on the surface of the substrate.

24

9 Microstructured Polymer Brushes 25

a)

b)

Figure 20 (a) Process used by Hawker et al.for the generation of polymer brushes with spa-tially resolved properties. A poly(t-butyl meth-acrylate) brush is covered with a photoresistcontaining a photoacid generator. Upon illumi-nation of the sample through a mask, protonsare generated in the illuminated areas. The pro-tons diffuse into the underlying brush andhydrolyze the ester groups. (b) Illustration ofthe different wetting properties of a sampleprepared as described in (a). The water on thesample only wets the illuminated areas – thatis, the areas in which the chains were trans-formed to a poly-(methacrylic acid).(Reprinted with kind permission from Ref. [77];.American Chemical Society, 2000.)


In principle, three different strategies can be followed for the generation of chem-ically micropatterned brushes, besides the trivial photoablation of the polymers bydeep UV-irradiation:

1. Deposition of the initiator in a patterned fashion and/or spatially addresseddeactivation of a complete initiator monolayer.

2. Spatially controlled growth of the polymer molecules through local address-ing of the initiator and/or confinement of the monomer access.

3. Spatially addressed chemical transformations of precursor brushes.

26

a) b)

c)

Figure 21 System used by Carter, Hawker etal. for the tuning of the feature size on nanos-tructures via a combined process consisting of(a) nanoimprinting and (b) surface-initiatedpolymerization from initiator sites (“inimers”)embedded into the mold; the AFM and SEM

images shown in (c) demonstrate the featuresize of lines after nonoimprinting (A,B) andafter the subsequent surface-initiated polymeri-zation (C,D). (Reprinted with kind permissionfrom Ref. [111]; .American Chemical Society,2003.)

9 Microstructured Polymer Brushes

In the first case, the initiator is deposited (by inkjet printing or stamping) in cer-tain areas of the substrate [110,111]. In a subsequent reaction step, the polymer isgenerated through growth of the polymer chains. In further reaction steps, initiatorcan be deposited in other, still uncovered areas of the substrate. Alternatively, a com-plete initiator monolayer can be formed and in selected areas the initiator deacti-vated or photochemically destroyed and evaporated, followed by growth of thebrushes. In the latter case (photoablation of the initiator), new initiator can beattached to the substrate in the thus obtained blank areas, either directly or after ashort etching process.

In the second case, the surface-attached polymer chains can be generated throughphotopolymerization reactions or other means, spatially to kick-off the polymeriza-tion reaction. An alternative, in which many different polymers can be formed in

27

a)

b)

Figure 22 (a) Schematic description of the l-stamping processused for the spatially resolved deposition of laminin to a brushcontaining active ester groups. (b) Neuronal cells aligning alongthe laminin grid deposited via this process.


one polymerization step, would be to supply the monomer only locally, followed bysimultaneous induction of the polymerization process for all monomers in the dif-ferent locations– for example, through flood exposure with UV irradiation orthrough thermal initiation (depending on the monomer).

In the third case, a precursor polymer brushwould first be formed in prettymuch thesame way as has been described above for the generation of functional brushes. Theonly difference here is that the transformation reactions into the final functionalbrush are carried out locally through administering the reagents [77] (Figure 22).

10Surface-Initiated Polymerization: The Overall Picture

Without attempting to gather all the information available for the synthesis of poly-mer monolayers by growing chains away from the surface, it can be safely statedthat surface-initiated polymerization to generate tailor-made surfaces is becomingincreasingly accepted. The number of systems for which the surface-attached layersof the initiator and the polymers have been well characterized, and the mechanismof the growth of surface-attached layers is well understood, has grown considerablyduring the past few years. Although only a short time ago it was doubtful that sys-tems with a high graft density and strong stretching of the polymer chains could besynthesized at all, a variety of approaches is now available that allows to study thephysics of densely grafted polymer brushes at ease. As most systems described inthe literature have been extensively extracted with good solvents for the polymer, itis also clear that these systems show a high stability, even under rather adverse con-ditions. This is a clear distinction from other techniques of deposition of thin poly-mer films, where only weaker interactions to the substrate are employed.

By using surface-initiated polymerization reactions, a wide range of monolayerscontaining functional groups has been synthesized over a wide spectrum of sub-strates. Indeed, one of the most attractive features of surface-initiated polymeriza-tion reactions is that they lead to highly swellable polymer layers attached to spheri-cal, tubular, planar, and even very irregular surfaces such as those of components ofcomplex microsystems.

Another key feature of surface-initiated polymerization is that the application of alocal stimulus allows local initiation of the polymerization reaction which, in turn,yields a spatially addressed growth of polymer chains. The performance of suchlocally addressable attachment of polymer molecules to generate chemically l-struc-tured surfaces is particularly of interest, as this cannot be achieved by other tech-niques of thin layer deposition.

Although at present it seems premature to outline the practical implicationsof surface-initiated polymerization reactions in depth, the future will undoubtedlyreveal a wide variety of applications, most notably in the areas of microsystems tech-nology and biomedical devices. On this basis, aspects of the practical applicationsof systems generated by surface-initiated polymerization will be discussed inChapter 17.

28

References 29

References

1 Introduction to Microlithography, 2nd Ed.(Eds.: L. F. Thompson, C. G. Willson,M. J. Bowden) American Chemical Society,Washington DC, 1994.

2 J. RPhe, V. Novotny, T. Clarke, G. B. Street,J. Tribol. Trans. ASME 1996, 118, 663.

3 M. Tirrell, E. Kokkoli, M. Biesalski, Surf. Sci.2002, 500, 61.

4 Surface and Interfacial Aspects of BiomedicalPolymers (Ed.: J. P. Andrade), Plenum Press,New York, 1985.

5 B. Zhao, W. J. Brittain, Prog. Polym. Sci.2000, 25, 677.

6 J. Pyun, T. Kowalewski, K. Matyjaszewski,Macromol. Rapid Commun. 2003, 24, 1043.

7 A. Halperin, M. Tirrell, T. P. Lodge, Adv.Polym. Sci. 1992, 100, 31.

8 R. A. L. Jones, R. W. Richards, Polymers atSurfaces and Interfaces, Cambridge UniversityPress, Cambridge, 1999.

9 G. J. Fleer, M. A. Cohen Stuart, J. M. H. M.Scheutjens, T. Cosgrove, B. Vincent, Poly-mers at Interfaces, Chapman & Hall, London,1993.

10 P. G. deGennes, J. Physique 1976, 37, 1445.11 G. S. Grest, in: Polymers in Confined Environ-ments, Vol. 138, 1999, pp. 149.

12 J. Klein, Y. Kamiyama, H. Yoshizawa,J. N. Israelachvili, G. H. Fredrickson, P. Pin-cus, L. J. Fetters,Macromolecules 1993, 26,5552.

13 J. Klein, E. Kumacheva, D. Mahalu, D. Pera-hia, L. J. Fetters, Nature 1994, 370, 634.

14 G. S. Grest, Curr. Opin. Colloid Interface Sci.1997, 2, 271.

15 G. S. Grest, Phys. Rev. Lett. 1996, 76, 4979.16 J. Klein, Colloid Surfaces – Physicochem. Eng.Aspects 1994, 86, 63.

17 J. Klein, E. Kumacheva, D. Perahia,D. Mahalu, S. Warburg, Faraday Disc. 1994,173.

18 B. D. Ratner, A. S. Hoffmann, F. J. Schoen,J. E. Lemons, Biomaterials Science, An Intro-duction to Materials in Medicine, AcademicPress, San Diego, 1996.

19 E. Ruoslahti, Annu. Rev. Cell Dev. Biol. 1996,12, 697.

20 K. L. Prime, G. M. Whitesides, J. Am. Chem.Soc. 1993, 115, 10714.

21 J. M. Harris, S. Zalipsky, Poly(ethylene glycol):Chemistry and Biological Applications, Ameri-can Chemical Society, Washington DC, 1997.

22 T. Lehmann, J. RPhe,Macromol. Symp. 1999,142, 1.

23 K. Mathauer, F. Embs, G. Wegner, in: Com-prehensive Polymer Science, Vol. 1. Suppl. (Ed.:G. Allen), Pergamon Press, Oxford, 1992,p. 449.

24 Multilayer Thin Films. Sequential Assembly ofNanocomposite Materials (Ed.: G. Decher),Wiley-VCH, Weinheim, 2002.

25 A. Ulman, An Introduction to UltrathinOrganic Films, Academic Press, New York,1991.

26 Y. N. Xia, G. M. Whitesides, Angew. Chem.Int. Ed. 1998, 37, 551.

27 M. J. Pellerite, E. J. Wood, V. W. Jones,J. Phys. Chem. B 2002, 106, 4746.

28 A. Hozumi, K. Ushiyama, H. Sugimura,O. Takai, Langmuir 1999, 15, 7600.

29 P. W. Hoffmann, M. Stelzle, J. F. Rabolt,Langmuir 1997, 13, 1877.

30 D. Kambhampati, P. E. Nielsen, W. Knoll,Biosensors Bioelectronics 2001, 16, 1109.

31 W. Knoll, M. Zizlsperger, T. Liebermann,S. Arnold, A. Badia, M. Liley, D. Piscevic,F. J. Schmitt, J. Spinke, Colloids Surfaces –Physicochem. Eng. Aspects 2000, 161, 115.

32 K. P. Krenkler, R. Laible, K. Hamann, Angew.Makromol. Chem. 1953, 53, 101.

33 N. Tsubokawa, M. Hosoya, K. Yanadori,Y. Sone, J. Macromol. Sci. – Chem 1990, A27,445.

34 N. Tsubokawa, A. Kuroda, Y. Sone, J. Polym.Sci. 1989, A27, 1701.

35 H. Ben Ouada, H. Hommel, A. P. Legrand,H. Balard, E. Papirer, J. Colloid Interface Sci.1988, 122, 441.

36 K. Bridger, B. Vincent, Eur. Polym. J. 1980,16, 1017.

37 R. Laible, K. Hamann, Adv. Colloid InterfaceSci 1980, 13, 65.

38 A. V. Dmitrenko, N. E. Shadrina, S. S. Ivan-chev, N. N. Ulinskaya, A. M. Volkov, J. Chro-matogr. 1990, 520, 21.

39 K. Hashimoto, T. Fujisawa, M. Kobayashi,R. Yosomiya, J. Macromol. Sci. – Chem 1982,A18, 173.

Polymer Brushes: On the Way to Tailor-Made Surfaces30

40 K. Hashimoto, T. Fujisawa, M. Kobayashi,R. Yosomiya, J. Appl. Polym. Sci. 1982, 27,4529.

41 V. I. Trachenko, Y. N. Zil’berman, T. F. Shats-kaya, E. G. Pomerantseva, Polym. Sci. USSR1986, 28, 646.

42 A. Kopf, J. Baschnagel, J. Wittmer, K. Binder,Macromolecules 1996, 29, 1433.

43 R. Zajac, A. Chakrabarti, Physical Rev. E1995, 52, 6536.

44 H. Ishida, C. H. Chiang, J. L. Koenig, Poly-mer 1982, 23, 251.

45 S. Naviroj, J. L. Koenig, H. Ishida, J. Macro-mol. Sci. Physics 1983, B22, 291.

46 A. M. Zaper, J. L. Koenig, Polymer Composites1985, 6, 156.

47 G. Reiter, Actualite Chimique 2003, 12.48 G. Reiter, R. Khanna, Phys. Rev. Lett. 2000,85, 2753.

49 G. Reiter, R. Khanna, Langmuir 2000, 16,6351.

50 G. Reiter, P. Auroy, L. Auvray,Macromolecules1996, 29, 2150.

51 K. L. Beers, S. G. Gaynor, K. Matyjaszewski,S. S. Sheiko, M. Moller,Macromolecules 1998,31, 9413.

52 M. Wintermantel, M. Gerle, K. Fischer,M. Schmidt, I. Wataoka, H. Urakawa, K.Kajiwara, Y. Tsukahara,Macromolecules 1996,29, 978.

53 Y. Tsukahara, K. Mizuno, A. Segawa, Y.Yamashita,Macromolecules 1989, 22, 1546.

54 S. Alexander, J. Physique 1977, 38, 977.55 E. B. Zhulina, O. V. Borisov, V. A. Pryamit-

syn, T. M. Birshtein,Macromolecules 1991,24, 140.

56 C. M. Wijmans, J. Scheutjens, E. B. Zhulina,Macromolecules 1992, 25, 2657.

57 S. T. Milner, T. A. Witten, M. E. Cates,Macro-molecules 1988, 21, 2610.

58 P. Y. Lai, K. Binder, J. Chem. Physics 1992,97, 586.

59 S. T. Milner, T. A. Witten, M. E. Cates,Macro-molecules 1989, 22, 853.

60 R. Hariharan, C. Biver, J. Mays, W. B. Russel,Macromolecules 1998, 31, 7506.

61 P. Pincus,Macromolecules 1991, 24, 2912.62 H. Ahrens, S. Forster, C. A. Helm, Phys. Rev.Lett. 1998, 81, 4172.

63 H. Ahrens, S. Forster, C. A. Helm,Macromol-ecules 1997, 30, 8447.

64 G. Boven, M. Oosterling, G. Challa,A. J. Schouten, Polymer 1990, 31, 2377.

65 G. Boven, R. Folkersma, G. Challa,A. J. Schouten, Polym. Commun. 1991, 32,50.

66 T. Stohr, J. Heinz, J. RPhe,Mater. Sci. Eng.C-Biomimetic and Supramolecular Systems1999, 8-9, 225.

67 T. Stohr, J. RPhe,Macromolecules 2000, 33,4501.

68 O. Prucker, J. RPhe,Macromolecules 1998, 31,592.

69 O. Prucker, J. RPhe,Macromolecules 1998, 31,602.

70 J. RPhe,Macromol. Symp. 1998, 126, 215.71 M. Biesalski, D. Johannsmann, J. RPhe,Macromol. Symp. 1999, 145, 113.

72 O. Prucker, J. RPhe, Langmuir 1998, 14,6893.

73 O. Prucker, J. Habicht, I. J. Park, J. RPhe,Mater. Sci. Eng. C-Biomimetic and Supramo-lecular Systems 1999, 8-9, 291.

74 X. Guo, M. Ballauff, Langmuir 2000, 16,8719.

75 X. Guo, A. Weiss, M. Ballauff,Macromole-cules 1999, 32, 6043.

76 M. Husseman, E. E. Malmstrom, M. McNa-mara, M. Mate, D. Mecerreyes, D. G. Benoit,J. L. Hedrick, P. Mansky, E. Huang, T. P.Russell, C. J. Hawker,Macromolecules 1999,32, 1424.

77 M. Husemann, M. Morrison, D. Benoit,K. J. Frommer, C. M. Mate, W. D. Hinsberg,J. L. Hedrick, C. J. Hawker, J. Am. Chem.Soc. 2000, 122, 1844.

78 X. Y. Chen, D. P. Randall, C. Perruchot,J. F. Watts, T. E. Patten, T. von Werne,S. P. Armes, J. Colloid Interface Sci. 2003,257, 56.

79 C. Perruchot, M. A. Khan, A. Kamitsi,S. P. Armes, T. von Werne, T. E. Patten,Langmuir 2001, 17, 4479.

80 J. B. Kim, W. X. Huang, M. L. Bruening,G. L. Baker,Macromolecules 2002, 35, 5410.

81 W. X. Huang, J. B. Kim, M. L. Bruening,G. L. Baker,Macromolecules 2002, 35, 1175.

82 A. Ramakrishnan, R. Dhamodharan, J.RPhe,Macromol. Rapid Commun. 2002, 23,612.

83 T. von Werne, T. E. Patten, J. Am. Chem. Soc.2001, 123, 7497.

84 K. Matyjaszewski, P. J. Miller, N. Shukla,B. Immaraporn, A. Gelman, B. B. Luokala,T. M. Siclovan, G. Kickelbick, T. Vallant,

References 31

H. Hoffmann, T. Pakula,Macromolecules1999, 32, 8716.

85 M. Ejaz, K. Ohno, Y. Tsujii, T. Fukuda,Mac-romolecules 2000, 33, 2870.

86 M. Ejaz, Y. Tsujii, T. Fukuda, Polymer 2001,42, 6811.

87 Y. Tsujii, M. Ejaz, S. Yamamoto, T. Fukuda,K. Shigeto, K. Mibu, T. Shinjo, Polymer 2002,43, 3837.

88 T. von Werne, T. E. Patten, J. Am. Chem. Soc.1999, 121, 7409.

89 X. Huang, M. J. Wirth,Macromolecules 1999,32, 1694.

90 X. Y. Huang, M. J. Wirth, Anal. Chem. 1997,69, 4577.

91 J. D. Jeyaprakash, S. Samuel, R. Dhamod-haran, J. RPhe,Macromol. Rapid Commun.2002, 23, 277.

92 R. A. Sedjo, B. K. Mirous, W. J. Brittain,Mac-romolecules 2000, 33, 1492.

93 M. Baum, W. J. Brittain,Macromolecules2002, 35, 610.

94 M. D. K. Ingall, C. H. Honeyman, J. V. Mer-cure, P. A. Bianconi, R. R. Kunz, J. Am.Chem. Soc. 1999, 121, 3607.

95 R. Jordan, A. Ulman, J. F. Kang, M. H. Rafai-lovich, J. Sokolov, J. Am. Chem. Soc. 1999,121, 1016.

96 B. Zhao, W. J. Brittain, J. Am. Chem. Soc.1999, 121, 3557.

97 B. Zhao, W. J. Brittain,Macromolecules 2000,33, 342.

98 B. Zhao, W. J. Brittain,Macromolecules 2000,33, 8813.

99 J. RPhe, W. Knoll, J. Macromol. Sci. -Polym.Rev. 2002, C42, 91.

100 M. Biesalski, J. RPhe,Macromolecules 2003,36, 1222.

101 M. Biesalski, D. Johannsmann, J. RPhe,J. Chem. Phys. 2002, 117, 4988.

102 M. Biesalski, J. RPhe,Macromolecules 1999,32, 2309.

103 B. Peng, J. RPhe, D. Johannsmann, Adv.Mater. 2000, 12, 821.

104 B. Peng, D. Johannsmann, J. RPhe,Macro-molecules 1999, 32, 6759.

105 S. A. Prokhorova, A. Kopyshev, A. Ramak-rishnan, H. Zhang, J. RPhe, Nanotechnology2003, 14, 1098.

106 B. Zhao, W. J. Brittain, W. S. Zhou, S. Z. D.Cheng, J. Am. Chem. Soc. 2000, 122, 2407.

107 S. Minko, D. Usov, E. Goreshnik, M.Stamm,Macromol. Rapid Commun. 2001, 22,206.

108 A. Sidorenko, S. Minko, K. Schenk-Meuser,H. Duschner, M. Stamm, Langmuir 1999, 15,8349.

109 O. Prucker, M. Schimmel, G. Tovar, W.Knoll, J. RPhe, Adv. Mater. 1998, 10, 1073.

110 T. A. von Werne, D. S. Germack, E. C. Hag-berg, V. V. Sheares, C. J. Hawker, K. R. Car-ter, J. Am. Chem. Soc. 2003, 125, 3831.

111 R. R. Shah, D. Merreceyes, M. Husemann,I. Rees, N. L. Abbott, C. J. Hawker, J. L.Hedrick,Macromolecules 2000, 33, 597.

Polymer Brushes: On the Way to Tailor-Made...

Documents

Transcript of Polymer Brushes: On the Way to Tailor-Made...