Supercritical Fluid Extraction and Separation of ...kbkf.kkft.bme.hu/EBCP_SF.pdfSupercritical Fluid...

Supercritical Fluid Extraction and Separation of Biologically Active

Components

Béla SimándiBudapest University of Technology and

Economics

Department of Chemical and Environmental Process Engineering

Historical notes

• 1822 – Baron Cagniard de la Tour

Critical point of a substance

• 1879 – 1880 – Hannay and Hogarth

KI, KBr, CoCl2 in alcohol (TC = 243 °C, PC = 63 bar)

• 1880 - 1920 - van der Waals, Amagat, Joule,

Thomson, Clausius

Pioneer work in research of thermodynamics of one and two component supercritical system

Historical notes

• 1896 – VillardSolubilities of camphor, stearic acid, paraffin wax in methane, ethylene, CO2

• 1943 – Messmore, 1955 - ZhuzeDeasphalting of petroleum oils with propane/propylene (T = 100 °C, P = 100 – 150 bar)

• 1964 – ZoselThe patent includes 68 examples of extractions and separarations

• 1970 – 1980 commercial applications

Supercritical Fluid Extraction (SFE) Units (V > 0.1m3)

SCOPE of presentation

• Physico-chemical properties

• Supercritical Fluid Extraction (SFE)

• SFE of plant constituents

SFE of plant constituents

essential oilstriterpenes and steroidscarotenoidsalkaloidsplant lipids

SCOPE of presentation

• Physico-chemical properties

• Supercritical Fluid Extraction (SFE)

• SFE of plant constituents

• Down stream processing

• Modelling

• Economic issues

• Conclusion

Life Long Learning 2012, Graz

What is a supercritical fluid?

Supercritical fluid

Density of carbon dioxide vs. pressure

Density of carbon dioxide vs. temperature

Density of water vs. temperature

Range over which near-critical extraction have been carried out

Viscosity of carbon dioxide at the near critical region

Viscosity of water

Specific heat capacity of carbon dioxide

•Bruno and Ely, 1991

Thermal conductivity of CO2

Bruno and Ely, 1991

Transport properties of supercritical solvents

Gas Fluid Liquid

ρ (kg/m3) 1 200 – 700 1000

η (Pas) 10-5 10-4 10-3

D (cm2/s) 10-1 10-3 - 10-4 10-5

Schneider et al., 2000

Fluid phase behaviour

B

A

yA

Phase diagram of ethanol – CO2

•Ploishuk et al, 2001.

The six basic types of fluid phase behaviour

Van Koynenburg and Scott, 1980

High-pressure phase equilibria

),,(),,( i

v

ii

l

i yPTfxPTf =

i

v

iv

i Py

f=φi

l

il

i Px

f=φi

l

ii

v

i xy φφ =

∫∞

−

∂∂−==

≠

V

nVTi

v

ii

i

v

i

RT

PVdV

n

P

V

RT

RTPy

yPTf

ij

ln1

ln),,(

ln,,

φ

22 2

)(

bVbV

Ta

bV

RTP

−+−

−=

P = P(T,V,ni)

The Peng-Robinson equation of state

Kikic, de Loos, 2001

Solubility of solids in supercritical fluids

SFSff 22 =

== ∫

P

P

SVsubSS

sub RT

dPvPff

2

22222 expφ

PyfSFSF

222 φ=

=∫SF

P

P

SV

sub S RT

dPv

P

Py

2

22

22

2

exp

φ

φ

Kikic, de Loos, 2001

Solubility of naphthalene in CO2

Solvating power of supercritical fluids

Critical properties of used solvents

Solvent TC (°C) P C (bar)

Ethylene (C2H4) 9 50,3

Ethane (C2H6) 32 48,8

Propylene (C3H6) 92 46,2

Propane (C3H8) 97 42,4

n-pentane (C5H12) 197 33,7

Benzene (C6H6) 289 48,9

Toluene (C7H8) 319 41,1

Supercritical fluids in food and pharmaindustries

• Carbon dioxide (PC=73.8 bar, TC=31.1°C)

Carbon dioxide

• GRAS (generally regarded as safe)• gentle conditions (PC=73.8 bar,

TC=31.1°C)• inert, odourless, tasteless• easily removed from products• non-explosive• readily available, inexpensive• benign solvent

Supercritical fluids

• Carbon dioxide (PC=73.8 bar, TC=31.1°C)• Nitrous oxide (PC=71.0 bar, TC=36.5°C)• Propane (PC=42.5 bar, TC=96.7°C)

Propane

• near-critical liquid (TC=96.7°C)• flammable, explosive mixtures with air• maximum permitted residue: 50 ppm• significant solvent loss• waste solvent


• Carbon dioxide (PC=73.8 bar, TC=31.1°C)• Propane (PC=42.5 bar, TC=96.7°C)• Water (PC=220.5 bar, TC=374.2°C)


• Carbon dioxide (PC=73.8 bar, TC=31.1°C)• Propane (PC=42.5 bar, TC=96.7°C)• Water (PC=220.5 bar, TC=374.2°C) • Entrainers (ethanol, ethyl acetate,

acetone)

Modifiers (co-solvent, entrainer)

Solvent T C (°C) P C (bar)

n-pentane 196,6 33,7

n-hexane 234,5 30,3

methanol 239,5 80,8

ethanol 241 61,4

n-buthanol 288,9 45

acetone 235 47

Dimethyl ether 126,9 54

Phase diagram of ethanol – CO2

•Ploishuk et al, 2001.

Change of critical parameters by addition of entrainer

Concentration acetone methanol ethanol n-buthanol

mol% Tc(°C) P c(bar) Tc(°C) P c(bar) Tc(°C) P c(bar) Tc(°C) P c(bar)

1 34,7 77,9 32,7 76,5 32,7 76,6 36,5 80,3

2 36,8 79,7 34,7 78,2 35,7 78,3 42,5 87,5

4 43,7 85,7 37,7 81,7 40,5 84,3 56,1 108

•Pure CO2: Tc=31,3°C, Pc=73,8 bar

Developing a Commercial-scale SFE Process

• Analytical/laboratory unit: 1 mL to 1 Lscreening unitsamples for analytical evaluation

• Pilot plant unit: 2 to 100 Lprocessing variablessamples for customers

• Semi-commercial plant: 100 to 200 Lproduction for market tests

• Commercial plant: 200 to 6500 L

Supercritical fluid extraction unit

I. separator

ExtractorCO2 storage

CoolerPumpHeat exchanger

II. separator

Heat exchanger Cooler/

condenser

ValveValve ValveValve

Heat exchanger

Pump process in the T – s diagram

1 liquid

2 undercooler

3 extraction pressure

4 extraction temperature

5 two phase region

5-6 gaseous part

5-7 liquid part

8 gas

Multipurpose extraction unit(NATEX, Austria)

Preparation of raw materials

• Crushing/grinding

Effect of particle size: SFE of paprika

0

2

4

6

8

10

12

0 5 10 15 20 25 30 35

kg CO2/kg dry paprika

Yie

ld (

%)

Run 32

Run 33

Run 34

Run 35

d0 = 0.96 mm

d0 = 0.45 mm

Preparation of raw materials

• Crushing/grinding• Rolling/pelletizing• Wetting/drying

Effect of moisture content: SFE of olive fruit

0

2

4

6

8

10

12

0 20 40 60 80 100 120 140 160

kg CO2/kg dry olive fruit

Yie

ld (

%)

Moisture content: 5.12 %

Moisture content: 47.67 %

Preparation of raw material

• Crushing/grinding• Rolling/pelletizing• Wetting/drying• Chemical/biochemical pretreatment

Extraction conditions

• Pressure• Temperature• Solvent composition (co-solvents)• Extraction time• Solvent flow rate• Upflow/downflow• Packed bed structure (H/D, axial mixing)• Mixing

Separation conditions

• Pressure• Temperature• Multiple separators (fractionation)• Absorbers• Adsorbers • Membranes

Formation of secondary plant products

Process of primary metabolism

CO2 H2O

Pentose cycle

Saccharides

Glycolysis

Phosphoenol-pyruvic acid

Pyruvate

Acetyl CoA

Krebs cycle

CO2


CO2 H2O

Pentose cycle

Saccharides

Glycolysis


Pyruvate

Acetyl CoA

Krebs cycle

CO2

Products of primary metabolism

Oligo and polisaccharides

Eritrose-4-PO4

Shikimic acid

Aromatic amino acids

Aliphatic amino acids

Malonyl CoAFatty acids

IPP biological isopren

Squalene

Proteines


CO2 H2O

Pentose cycle

Saccharides

Glycolysis


Pyruvate

Acetyl CoA

Krebs cycle

CO2

Products of primary metabolism

Oligo and polisaccharides

Eritrose-4-PO4

Shikimic acid

Aromatic amino acids

Aliphatic amino acids

Malonyl CoAFatty acids

IPP biological isopren

Squalene

Secondary plant metabolism

Glycosides

Mucilages and gums

Phenolic compounds

Lignans

Alkaloids

Fats and waxes

Tetracyclynes

Antraquinones

Alkaloids

Peptides

Terpenes

Terpenoids

Isoprene (C5) units

Isoprene does not participate in the biogenesis

Biosynthesis of Limonene and Carvone

Bouwmeester, H.J. et al.: Plant Physiol., 117, 901-912 (1998).

Terpenoids

• C10 monoterpenes (essential oil)• C15 sesquiterpenes (essential oil)• C20 diterpenes (antioxidants)• C30 triterpenes (phytosterols)• C40 tetraterpenes (carotenes)• (C5)n polyterpenes (caoutchouc)

Solubility

• typical essential oil components

Stahl, E., Quirin, K.-W., Gerard, D.: Dense Gases for Extraction and Refining, Springer-Verlag, Berlin, 1988.

Comparative chemical composition of distilled oils and SFE products

Plant Compound Steamdistillation

SFE

Dill(Anethum graveolens L.)

limonened-carvone

55.736.5

42.648.6

Coriander(Coriandrum sativum L.)

pinenelinalool

15.368.5

7.775.5

Celery(Apium graveolens L.)

limonene3-butylphthalide

50.523.6

33.440.6

Parslay(Petroselinum crispum L.)

α-pineneβ-pinenemyristicin

apiole

24.021.97.438.5

1.53.04.084.9

Alteration of essential oil components during distillation

linalyl acetate → linaloollavandin (Lavandula intermedia Emeric)clary sage (Salvia sclarea L.)

glycosides → thymol

thyme (Thymus vulgaris L.)unknown precursor → α-terpineol

terpinene-4-ol

lavandin (Lavandula intermedia Emeric)pulegone → dihydrocarvone

spearmint (Mentha spicata L.)

Fractionation of fennel oil(Foeniculum vulgare Mill.)

Life Long Learning 2012, Graz

Fractionation of fennel oil (PE=302 bar, TE=38°C)

1st separator:75 bar, 23.5 °C

1st separator:86 bar, 23.5 °C

Effects of the separation parameters on fractionation of fennel oil

0.75 1 1.5 2 2.5 3 4 5 6 7

P S, bar

TS, °

C

18

22

26

30

34

38

42

68 72 76 80 84 88 92

Composition of orange oil

Component Composition (%)

α-Pinene 0.45

Myrcene 1.77

d-Limonene 90.60

Octanal 0.59

Decanal 0.52

Linalool 0.37

Geranial 0.12

Phase equilibria of orange peel oilBrunner, G.: Industrial Process Development: Counter current Multistage Gas Extraction Processes, Proceedings of 4th International Symposium on Supercritical Fluids, 745-756, Sendai, Japan, 1997.

Separation factors of orange peel oil fractions

x

yK =

Distribution coefficient:

Separation factor:

.frAroma

Terpenes

K

K

−

=α

Brunner, G.: Industrial Process Development: Counter current Multistage Gas Extraction Processes, Proceedings of 4th International Symposium on Supercritical Fluids, 745-756, Sendai, Japan, 1997.

α

Fractionation of citrus oil

CO2

Aroma

CO2

Feed

Terpenes

Results of fractionation

Pilot plant data on deterpenation of orange oil

Aromatic content in feed oil 4.1 %

Aromatic content inconcentrated oil

18.9 %

Recovery of aromatic fraction 90 %

Solvent-to-oil ratio 100

SFE of sesquiterpene lactonesChamomile (Matricaria chamomilla L.)

chamazulene

OH

O

(-)-α-bisabololoxide B

O

OH

(-)-α-bisabololoxide A

OH

(-)-α-bisabolol

O

O

OH

-D-glucose-O

OH

β

apigenin-7-glucoside

O

O

OH

OH

OH

apigenin

OOCOCH3

O

OH

matricine

O

O

trans-en-in-dicycloether

OO

cis-en-in-dicycloether

Fractionated extraction of chamomile

P

(bar)

T

(°°°°C)

Matricine

(%)

αααα-bisabolol

(%)

Spiroether

(%)

Yield

(%)

80 40 3.5 27 14 0.12

90 40 17.0 21.0 16.7 0.4

300 40 7.5 9.2 11.7 1.65


SFE of feverfew (Chrysanthemum partheniumBernh.): effects of pressure and temperature on the

yield

Recovery of parthenolide

O

O

O

parthénolide

Diterpenes investigated by SFE

OOH

OH

O

OH

rosmanol

OH

OH

sclareol

O

OO

OH

O

O

O

OHO ONH OH

O O

O

taxol

baccatin-III

O

H

O O O

OOH

OH

carnosol

OH

O

O

O

H

OHOH

O

O

Isolation of diterpenesfrom plants

Plant Component

Clary sage(Salvia sclarea L.)

sclareol

Yews(Taxus brevifolia Nutt.,Taxus cuspidata Capita)

paclitaxel baccatin III

Rosemary(Rosmarinus officinalis L.),

Sage(Salvia officinalis L.)

rosmanol7-methyl-epirosmanol

carnosol carnosolic acid

Basic sceleton of triterpenes

OH

H

α-amirin

OH

H

β-amirin

OH

OH

faradiol

H

H

OH

β-sitosterol

Isolation of triterpenesPlant Component

Mulberry(Morus alba L.)

α-amyrin acetate

Neem(Azadirachta indica A. Juss)

nimbinsalannin

azadirachtinMarigold

(Calendula officinalis L.)faradiol and its esters

Chaste tree(Vitex agnus castus L.)

β-amyrinβ-sitosterol

Dandelion(Taraxacum officinale Web.)

β-amyrinβ-sitosterol

SFE of dandelion leaves:effects of pressure and temperature on the yields

Recovery of sterols from dandelion leaves

Recovery of carotenoids

OH

OH

lutein (alfalfa)

β-carotene (carrot)

OH

OH

O

capsanthin (paprika)

OH

O

O Obixin (annetto)

lycopene (tomato)

10,589 11,214 11,839 12,464 13,089 13,714 14,339 14,964 15,589 16,214 16,839 17,464 18,089 18,714 19,339 above

SFE of pigments from paprika

Extraction of hop (Humulus lupulus L.)

Water Organic solvent Steam distillation

Water extract Total resin

Petrol soluble Petrol insoluble

‘Hard resin’

Essential oil

‘Soft resin’

α acid β acid Uncharacterised acid

Structure of major componentsOH

R

O

OHOH

O

R

OOH

OH

OOH

OH

R

O

OHO R=

CH3CH3

CH3

CH3

CH3

CH3

CH3CH3

CH3

α acid iso α acid

β acid

Composition of hop extract versus time

Gardner, D.S.: Commercial scale extraction of alpha-acids and hop oils with compressed CO2 inKing, M.B., Bott, T.R. (Eds.):Extraction of Natural Products Using Near-Critical Solvents, Chapman&Hall, Glasgow, 1993.

Chaste tree: Vitex agnus castus L.

p: 100-275-450 bar

T: 40-50-60 °C

Vitex agnus castus L.

Rotundifurane

Diterpene


β-amyrinTriterpenes

β-sitosterol


Casticin

Flavonoid


Minor compounds’ concentration of the extracts [g/kg]:

~3.5x ~3.5x >2.5x >3xsc-CO2/EtOH

Alkaloids: decaffeination of coffee

• Solubilityisobars of caffeine


Decaffeination of green coffee beans

Nearly continuous SFE of coffee

Extraction of nicotine from tobacco

Aroma removal

H2O

Tobacco Nicotine removal

Drying

Adsorbent regeneration

Aroma distribution CO2

Conditioning Nicotine reduced tobacco

CO2

Aroma transfer

Aroma

Nicotine by-product

Extraction of alkaloids used in medicine

Plant Alkaloid Recoverypaprika(Capsicum annuum L.)

capsaicin good

pepper(Piper nigrum L.)

piperine good

poppy(Papaver somniferum L.)

opium(codein,thebaine,

papaverin)

reasonable

strychnos(Strychnos nux-vomica L.)

strychnine reasonable

hemlock fruit(Conium maculatum L.)

coniine reasonable

ipecacuanhae radix(Cephyaelis ipecacuanha Will.)

emetin,caphaelin

poor

Fatty oils and waxes

• Solubility isotherms of soybean oil


Effect of alcohol-entrainer on extraction rate

0

10

20

30

40

50

60

0 20 40 60 80 100 120 140 160kg CO2 / kg dry corn germ

0 % alcohol

2.5 % alcohol

5 % alcohol

7.5 % alcohol

10 % alcohol

g oil / 100 g dry corn germ

List of raw materials for γ-linolenic acid

Raw material Yield CO2extraction

[%]

C18:3 in the extract(γ-linolenic acid)

[%]Oenothera biennis L. 22 8 - 12Borago officinalis L. 18 18 - 22Humulus lupulus L. 6 3 - 4Cannabis sativa L. 35 3 - 6Ribes rubrum L. 14 4 - 6Ribes negrum L. 18 16 - 19

Ribes grosullaria L. 15 10 – 12Rosa canina L. 12 24 - 31

Hippophae rhamnoides L. 12 26 - 30

Partition coefficient and selectivity of fish oil ethyl esters

Component K=y/x KA/K22

C16 0.090 7.5

C18 0.050 4.2

C20 0.027 2.3

C22 0.012 1.0

Two-column process for separation of EPA and DHA

Krukonis, V.J.: Processing with Supercritical Fluids in Charpentier, B.A., Sevemants, M.R. (Eds.): ACS Symposium Series 366, 26-43, ACS, Washington, 1988.

Extraction from lyophilised fermentation broths

Compound Micro-organism Methanol in CO2

1mevinolin Aspergilus terreus 3 %2chaetoglobosin Penicillium expansum 20 %2mycolutein unidentified 20 %2luteoreticulin unidentified 20 %2C16H12O7 Aspergillus fumigatus 0 %2sydowinin B Aspergillus fumigatus 20 %2elaiophylin Streptomyces sp. 20 %

1Larson and King, 19862Cocks et al., 1995

Partition coefficient of different compounds

Compounds K = y/x

N-alcohols C1 – C7 0.4…..31

Organic acids (wt basis) acetic acidbutyric acid

0.030.08

Esthersethyl formiateisoamyl acetate

16850

Continuous extraction broths

cells

cell separation unitCO2

productpressure reduction valve

fermenter

nutrients

extractor

Example: Extraction of pleuromutilin

Aqueous flow rate

ml/min

CO2 flowg/min

Ratio of CO2 to aqueous

w/v

Average extractor

residence timemin

Average extraction

%

1 10 10/1 130 78.4

2 10 5/1 65 65.0

1 20 20/1 130 88.8

2 20 10/1 65 76.9

Walford et al., 2000

Solubilization of proteins in supercritical fluids

H2O

H2O

CO2Head (hydrophilic)

Tail (CO2-philic)

*O

OO O

O

*n

m CF3 CF2 CF O CF2

CF3

COO-NH4+3

Flouorocarbons Hydrocarbon polymers

Modelling of SFE: Brunner’s equation

This image cannot currently be displayed.

)]exp(1[ ktYY EE −−= ∞

)]exp(1[)]exp(1[ 2211 tkYtkYY EEE −−+−−= ∞∞

•x concentration of soluble component(s) in solid phase (kg/kg )

•t time (s)

•K kinetic parameter (1/s)

•YE extraktion yield (kg/kg)

•YE∞ maximal extraction yield (kg/kg)

kxdt

dx −=

Fractionation of Greek sage oil

0 50 100 150 200 250 300 350 400 450 500

Idő (min)

0

1

2

3

4

5

6

Kih

ozat

al (

%)

összes 1. szeparátor 2. szeparátor

Parameters YE∞∞∞∞, % k, 1/min R

PE = 300 bar, TE = 45 °C 4,90 0,012 0,999

PSZ-1 = 90 bar, TSZ-1 = 30-40°C

YE1∞ = 2,30 k1 = 0,010 0,999

PSZ-2 = 20 bar, TSZ-1 = 20 °C YE2∞ = 2,59 k2 = 0,013 0,994

t (min)

Ytotal yield

2nd separator

1st separator

Component balances for solid and liquid phases

VyxJt

xVs d),(d)1( =

∂∂−− ερ VyxJ

h

yVu

t

yV ff d),(dd =

∂∂+

∂∂ ρερ

•x concentratin of soluble component(s) in solid phase (kg/kg)

•y concentation of soluble component(s) in supercritical fluid (kg/kg)

•J specific extraction rate (flux) (kg/(m3s))

•h length coordinate (at CO2 inlet h = 0) (m)

•t time (s)

•u fluid velocity (m/s)

ε voidage of packing (m3/m2)

ρf density of supercritical fluid (kg/m3)

ρs density of solid matrix (kg/m3)

Sovová’s model2The Sovová’s model describes the yield in function of the time using 3 equations:

τϑ t=

[ ])exp(10

Qx

Y−−= ϑ

Q

q<ϑ

)]1(exp[0

−−= kZQQ

q

x

Y ϑ kQ

q ϑϑ <≤

[ ]

−

−−+−= )1(exp1)exp(1ln

11

0

qQ

qSS

Sx

Yϑ kϑϑ ≥

kϑ

• Q, S dimensionless modelparameters [-]

• τ minimal extraction time [s]

• q soluble material fraction on the surface of the particles [kg/kg]

dimensionless time [-]

Zk, dimensionless length and time where and when the soluble material runs out from the surface [-]

[2] SOVOVÁ, H., Chemical Engineering Science, Vol. 49, 1994, p. 409

Parameters of the Sovová’s model

fm&

∗=ym

xm

f

s

&

0τsf

fpfs

m

akmQ

ρερ)1( −

=&

∗−=

ym

xakmS

f

pss

)1(0

ε&

•y* solubility at extraction pressure and temperature [kg/kg]xo initial concentration of soluble material [kg/kg] ρf, ρs density of the fluid and solid phase [kg/m3] ap specific surface area of the solid particles [m2/m3]

•kf, ks mass transfer coefficient in the fluid and the solid phase [m/s]

•ms amount of the raw material [kg]

• mass flow of the fluid [kg/s]

•ε void fraction in bed [m3/m3]

[ ]{ })exp(11ln1

SqSQ

qk −−+=ϑ

−

−+= 1exp

11ln

1Q

qS

qSZk ϑ







•Solubility was estimated from an empirical equationor

from the initial slope of the extraction curves

∗=ym

xm

f

s

&

0τsf

fpfs

m

akmQ

ρερ)1( −

=&

∗−=

ym

xakmS

f

pss

)1(0

ε&

[ ]{ })exp(11ln1

SqSQ

qk −−+=ϑ

−

−+= 1exp

11ln

1Q

qS

qSZk ϑ







•The initial concentration of soluble material wasestimated by traditional solvent Soxhlet-extraction

using n-hexane as solvent

∗=ym

xm

f

s

&

0τsf

fpfs

m

akmQ

ρερ)1( −

=&

∗−=

ym

xakmS

f

pss

)1(0

ε&

[ ]{ })exp(11ln1

SqSQ

qk −−+=ϑ

−

−+= 1exp

11ln

1Q

qS

qSZk ϑ


•y* solubility at extraction pressure and temperature [kg/kg]xo initial concentration of soluble material [kg/kg] ρf, ρs density of the fluid and solidphase [kg/m3] ap specific surface area of the solid particles [m2/m3]





•The density of the carbon dioxide was calculated fromthe Bender equation of state at the pressure and

temperature of the extraction.

∗=ym

xm

f

s

&

0τsf

fpfs

m

akmQ

ρερ)1( −

=&

∗−=

ym

xakmS

f

pss

)1(0

ε&

[ ]{ })exp(11ln1

SqSQ

qk −−+=ϑ

−

−+= 1exp

11ln

1Q

qS

qSZk ϑ


•y* solubility at extraction pressure and temperature [kg/kg]xo initial concentration of soluble material [kg/kg] ρf, ρs density of thefluid and solid phase [kg/m3] ap specific surface area of the solid particles [m2/m3]





•The density of the raw material was measured byHofsäss air pycnometer.

∗=ym

xm

f

s

&

0τsf

fpfs

m

akmQ

ρερ)1( −

=&

∗−=

ym

xakmS

f

pss

)1(0

ε&

[ ]{ })exp(11ln1

SqSQ

qk −−+=ϑ

−

−+= 1exp

11ln

1Q

qS

qSZk ϑ







•The specific surface are was estimated from the surface-volume diameterof the particles, what was determined from sieve experiments using the

Rosin–Rammler–Bennett particle size distribution model.

∗=ym

xm

f

s

&

0τsf

fpfs

m

akmQ

ρερ)1( −

=&

∗−=

ym

xakmS

f

pss

)1(0

ε&

[ ]{ })exp(11ln1

SqSQ

qk −−+=ϑ

−

−+= 1exp

11ln

1Q

qS

qSZk ϑ



•kf, ks mass transfer coefficient in the fluidand the solidphase [m/s]




•The mass transfer coefficient in the fluid phase wasestimated from a Sherwood correlation.

∗=ym

xm

f

s

&

0τsf

fpfs

m

akmQ

ρερ)1( −

=&

∗−=

ym

xakmS

f

pss

)1(0

ε&

[ ]{ })exp(11ln1

SqSQ

qk −−+=ϑ

−

−+= 1exp

11ln

1Q

qS

qSZk ϑ







•The void fraction in bed was calculated from thedensity of the raw material and the bulk density.

∗=ym

xm

f

s

&

0τsf

fpfs

m

akmQ

ρερ)1( −

=&

∗−=

ym

xakmS

f

pss

)1(0

ε&

[ ]{ })exp(11ln1

SqSQ

qk −−+=ϑ

−

−+= 1exp

11ln

1Q

qS

qSZk ϑ

The experimental yield points and the calculated yield curves

•Extraction of marigold: pE=350 bar, TE= 40°C

0

1

2

3

4

5

6

7

0 100 200 300 400idő (min)

Y (%)

Kísérletileg meghatározott extrakciós hozamokEgyszerűsített modellel számított hozamSovová modellel számitott hozam

•time (min)

0

1

2

3

4

5

6

7

0 100 200 300 400idő (min)

Y (%)


•The experimental yield points

•time (min)



0

1

2

3

4

5

6

7

0 100 200 300 400idő (min)

Y (%)


•time (min)



0

5

10

15

20

25

0 200 400 600 800 1000idő (min)

Y (%)

11 kg/h,mért7 kg/h, mért4 kg/h, mért11 kg/h, számított7 kg/h, számított4 kg/h, számított

•Effect of the solvent flow rate on the extraction of corn germ, Sovová modell, pE=450 bar, TE=40°C

•time (min)

•11 kg/h, experimental



•11 kg/h, calculated



0.1

1

10

100

0 5 10

SFE SFI SFE/SFF PA

Invesment costs of high pressure units

I=A(V TW)0.25

VTW

Perrut, M.: Ind. Eng. Chem. Res., 39, 4531 (2000).

Production costs of SFE(NATEX, Austria)

Comparison of solvent extraction with SFE

Plant capacity

(t/year)

Production cost

(EUR/kg raw material)

Notes

SE SFE

300-400 3-8 4-10 medicinal plants,

cosmetics, spices

1000-1200 1-3.5 2-5 food additives,

specific vegetable

oils

10000-12000 0.5-1.2 0.75-1.2 coffee, hop

100000-120000 0.2-0.4 ? oilseeds

CONCLUSIONS

• There is a growing interest in new natural products that act very specifically without any side-effects.

• Supercritical fluid extraction is ideal for preparation of medicinal plant extracts.

• Carbon dioxide extracts are different from the traditional preparations not only in chemical composition, but in phytotherapeutical activity as well.

• There is an urgent need for more detailed phytochemical analysis of extracts.

CONCLUSIONS

• Commercial supercritical fluid processes have been still limited to few areas.

• Large capacity plants, with optimised design and operation lead to prices that are very often of the same order of magnitude as those related to classical processes.

• Moreover, there are also cases were supercritical fluids permit to make products or operations, that cannot be realised by any other methods.

Thank you for your kind attention!

Supercritical Fluid Extraction and Separation of ...kbkf.kkft.bme.hu/EBCP_SF.pdfSupercritical Fluid...

Documents

Transcript of Supercritical Fluid Extraction and Separation of ...kbkf.kkft.bme.hu/EBCP_SF.pdfSupercritical Fluid...