Escolar Documentos
Profissional Documentos
Cultura Documentos
Review
Recovery and separation of organic acids by membrane-based solvent
extraction and pertraction
S. Schlosser
Presented in part at the membrane science and technology conference PERMEA 2003, Tatransk e Matliare, Slovakia, September
711, 2003 (http://sschi.chtf.stuba.sk/permea/).
Corresponding author. Tel.: +421 2 524 96743; fax: +421 2 524 96920.
E-mail address: schlosser@cvt.stuba.sk (
S. Schlosser).
1383-5866/$ see front matter 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.seppur.2004.07.019
238
S. Schlosser et al. / Separation and Purication Technology 41 (2005) 237266
3.3. Ionic liquids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
3.4. Toxicity and compatibility of solvent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
3.5. Co-transport and competitive transport (selectivity) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
4. Membrane contactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
4.1. Factors affecting the performance of contactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
4.2. Modelling and mass-transfer characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
5. Case studyrecovery of MPCA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
6. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
Nomenclature
A surface area (m
2
)
A
w
arithmetic mean value of the inner and outer
geometric surface areas of bers (m
2
)
c molar concentration of the solute (undissoci-
ated acid or acid in the complex) (mol m
3
)
D distribution coefcient ()
k individual mass-transfer coefcient (ms
1
)
K
e
overall mass-transfer coefcient in the extrac-
tor (ms
1
)
K
s
overall mass-transfer coefcient in the stripper
(ms
1
)
n molar ux (mol s
1
)
N
cs
number of contactors in series ()
N
ct
total number of contactors (both in parallel and
series) ()
r
e
rate constant of the extraction reaction, Eq. (3)
(ms
1
)
r
s
rate constant of the stripping reaction, Eq. (4)
(ms
1
)
R overall mass-transfer resistance (s m
1
)
Re Reynolds number ()
u linear velocity of the ow (ms
1
)
V volumetric owrate (m
3
s
1
)
Y
MA/OA
ratio of mineral acid to organic acid ux ()
Z concentration factor of the solute in the con-
centrate (output from the stripper) dened by
relation Z = c
R,n+1
/c
F1
()
n+1
concentration ratio
n+1
= c
S,n+1
/c
S,n+1
(ap-
proach to an equilibrium on the rafnate end
of the contactor in MBSE, c
S,n+1
= D
F
c
F,n+1
)
()
porosity of the wall ()
e
yield of the solute in extraction ()
conv
conversion of the reagent in the stripping solu-
tion ()
Subscripts
0 initial value
1 feed or stripping solution inlet end of a HF con-
tactor or a series of contactors
2 rafnate or stripping solution outlet end of a
HF contactor or a series of contactors
b boundary layer in the bulk phase
e extractor (MBSE)
F feed phase, feed boundary layer
i inner surface of the ber wall
n number of the contactor segments
o outside surface of the ber wall
R stripping solution; stripping interface
s stripper (MBSS)
S solvent phase, boundary layer in the solvent
w ber wall
Abbreviations
6-APA 6-aminopenicillanic acid
AOT sodium di(2-ethylhexyl)sulfosuccinate (an-
ionic surfactant)
BLM bulk liquid membrane
CF HF cross-ow hollow ber contactor
D2EHPA di(2-ethylhexyl)phosphoric acid
DLC double Lewis cell with layered BLM
DNNSA dinonylnaphtalenesulfonic acid
EC equilibrium cell for contacting two liquids
ELM emulsion liquid membrane
EXT solvent extraction
FSC at sheet contactor
HF hollow ber
MBSE membrane-based solvent extraction
MBSS membrane-based solvent stripping
MHS multimembrane hybrid system (LM between
two ion-exchange membranes)
MIBK methylisobutylketone
S
.
S
c
h
l
o
s
s
e
r
e
t
a
l
.
/
S
e
p
a
r
a
t
i
o
n
a
n
d
P
u
r
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
4
1
(
2
0
0
5
)
2
3
7
2
6
6
Table 10
Biotransformations of organic acids integrated with extractive separation of the product or realized in hybrid systems
Product Carbon source or reactant(s) (micro-
organism or biocatalyst)
Process for separation of the product Solvent Contactor type
a
Literature
Itaconic acid Citric acid (Aspergillus terreus) Hybrid bioreactor with two SLM (D2EHPA, TOA)/(octanol;
dichloromethane)
FSC [226]
Fumaric acid l-Malic acid (Sacharomyces cerevisiae) D2EHPA/dichloromethane FSC [227]
-Hydroxyisobutyric acid 2-Methyl-1,3-propanediol (Acetobacter
alei)
EXT TOPO/isooctane screening of
solvents
EC [228]
MBSE TOPO/isooctane PF HF [229]
N-(Benzyloxycarbonyl)-l-aspartyl-
l-phenylalanine methyl ester
N-(Benzyloxycarbonyl)-l-aspartic
acid and l-phenylalanine methyl ester
(thermolysin)
MBSE tert-Amylacohol PF HF [230,231]
EXT into dispersion +MF Hexanol, heptanol FSC [19]
EXT in situ Butylacetate agitated vessel [232]
N-Formyl-l-aspartyl-l-
phenylalanine methyl ester
N-Formyl-l-aspartic acid and l-phenyl-
alanine methyl ester (thermolysin)
EXT in situ 1-Butanol agitated vessel [233]
Dipeptide aspartam (ZalaPheOMe) ZAlaOH and PheOMe (thermolysin) EXT in situ Ethyl acetate agitated vessel [234]
MBSE PF HF [235]
Isovaleraldehyde Isoamyl alcohol (Gluconobacter oxy-
dans)
MBSE Isooctane PF HF [236]
S-Phenylethanol Acetophenone (dehydrogenase fromCan-
dida boidinii)
MBSE Isooctane HF [237]
2-Phenylethanol Phenylalanine (Sacharomyces cerevisiae) EXT Dibutyl sebacate uidised bed with capsules [13]
EXT in situ Oleic acid agitated vessel [238]
3-Cyanobenzoic acid 1,3-Dicyanobenzene (hydratase from
Rhodococcus)
EXT Ionic liquid agitated vessel [239]
Abbreviations used are explained in Nomenclature.
C, whichis 87.0 g m
3
and
at pH 1.98 is 52.7 g m
3
. Solubility of D2EHPA decreases
with increasing concentration of chlorides and HCl. Rela-
tively high solubility of D2EHPA results in a low lifetime of
SLM [26].
3.2. Diluents and modiers
To nd a suitable extractant (carrier) forming a com-
plex with the target solute is only half of the success. The
soluteextractant complex has to be soluble in the diluent
forming with the extractant a solvent. The achievement of
this goal requires in many cases selection of an appropriate
diluent or addition of a modier to the solvent, e.g. isodecanol
to n-alkanes and trioctylamine (TOA) in the solvent for lactic
acid [115]. The diluent type may have a great inuence on the
solvent performance. In extraction of MPCA from solutions
with higher content of mineral salts with solvents contain-
ing TOA, the value of the distribution coefcient is much
higher for xylene as a diluent than for n-alkanes [271]. When
considering separation integrated with bioprocess toxicity as-
pects should be taken into account, as discussed in Section
3.4. Lower alcohols, which are popular modiers, should be
avoided. Isotridecanol, available as a technical product (Uni-
par, NL), is a good compromise [272].
The diluent effect in extraction of organic acids is ad-
dressed in papers [94,96,126,178,247,268]. The diluent ef-
fect in systems with enzymatic reactions studied Miyako et
al. [46,47]. In extraction of lactic acid with tertiary amines,
the value of the distribution coefcient of a 50% mixture of
Alamine 336 with oleyl alcohol was found to be higher than
that of pure amine [212].
Addition of 30 wt.% of isodecanol to the solvent with
0.4 mol dm
3
of Hostarex A327 changed completely course
of concentration dependence of the distribution coefcient,
which value was increasing with decreasing concentration
252
S. Schlosser et al. / Separation and Purication Technology 41 (2005) 237266
of DMCCA in aqueous phase [29,87]. For the solvent with
0.4 mol dm
3
of TOA in n-alkanes the value of the distribu-
tion coefcient of DMCCA was decreasing with decreasing
concentration of acid.
3.3. Ionic liquids
Ionic liquids (IL) are composed of organic cations and ei-
ther organic or inorganic anions that remain liquid over a wide
temperature range, including room temperature [273,274].
ILs are a newgroup of solvents or extractants of great interest
recently studied as potential green solvents [273,275277].
Practically zero vapour pressure of IL and temperature sta-
bility makes them attractive solvents in many applications
as reaction medium. A higher viscosity at room tempera-
tures could be their less favourable property. Solvent extrac-
tion of organic acids by ionic liquids was studied in papers
[278281], erythromycin by Cull et al. [239] and chlorophe-
nols by Bekou et al. [282]. Transport of amines and neutral
organic substances through liquid membranes from IL is of
concern in papers of Fortunato et al. [283] and Branco et
al. [284]. Pertraction of organic acids through liquid mem-
branes facilitated by enzymatic reactions on L/L interfaces
using IL as a liquid membrane was studied by Miyako et al.
[46,174]. Tailoring of ionic liquids to achieve good partition-
ing of target solutes and acceptable viscosity of IL may lead
to interesting results.
Experiments with developmental IL show promising
results in extraction of organic acids, especially of lactic
acid [279281]. In extraction of butyric acid, lactic acid
and phenol fairly higher distribution coefcients were found
for solvents with tested developmental ionic liquid IL-A
compared to the solvents containing tertiary amines. The
value of the distribution coefcient of lactic acid is up to
30 at lower acid concentrations, what is promising. IL-A
acts as an extractant forming undissociated lactic acid/IL-A
complexes 1:1 and 2:1 (lactate anions are not extracted)
[281]. In the pertraction of LA through SLM the value of the
overall mass-transfer coefcient increases with decreasing
concentration of lactic acid in the aqueous phase what cor-
relates with increasing value of its distribution coefcient.
Increased concentration of the carrier IL-A did not change
the value of the mass-transfer coefcient in pertraction
of LA contrary to the increased value of the distribution
coefcient. This may indicates that the slower kinetics of the
interfacial reaction in decomposition of the complex plays a
role or higher viscosity of membrane is responsible of this.
Separation of taurine (2-aminoethanesulfonic acid) and
sodium sulfate by leaching a solid mixture by ionic liquids
is another example of their application [159]. Dialkylimida-
zolium chloride ionic liquid as leaching agent and organic
solvent as precipitating agent (lower alcohols, like ethanol,
are effective) were developed. Selective separation of taurine
from a solid mixture containing a large amount of sodium
sulfate could be realized with 6798.5% yield in a single
separation step.
3.4. Toxicity and compatibility of solvent
Toxicity of the solvent for microorganisms or its inhibitory
effect on the catalytic activity of enzymes or the biologic ac-
tivity of the product are important properties of the solvent,
which have to be considered in their evaluation. Papers con-
cerning these items for the solvents or their components are
listed in Table 12.
The parafn (probably parafnic oil, not specied)
severely reduces the biological activity of the extracted bac-
teriocin nisin, while toluene, kerosene and isooctane were
found suitable [153]. Strategies for reducing solvent toxicity
in extractive fermentations are discussed in paper [213]. Co-
immobilization of soybean oil with cells in carrageenan gel
beads decreased greatly toxic effect of Alamine 336 in lactic
acid fermentation. Toxicity of tertiary amine Alamine 336 to
Lactobacillus delbreucki was suppressed by the second stage
extraction of amines with oleyl alcohol before returning the
broth to the fermentation [212].
Logarithm of the distribution coefcient of a given com-
pound in the n-octanol/water system, log P, is traditionally
used as a simple criterion to guide the choice of solvents for
biphasic enzymatic reactions from point of view of their bio-
compatibility with enzyme [200,285]. However recently, it
was found that log P is not a very reliable parameter for the
solvent choice for enzymatic transformations [286].
Wider screening of 17 solvents and their components for
their toxicity to Rhizopus arrhizus showed that solvents with
tertiaryamines (HostarexA327, TOA), secondaryamine with
isotridecanol as modier and pure trihexylphosphate (THP)
are suitable from point of view of biomass compatibility
[272]. Toxicity of alcohols decreased with increasing molec-
ular mass. Octanol and isodecanol and solvents containing
them are toxic. Isotridecanol is medium toxic. Earlier study
[287] showed that TBP is practically lethal for fungi. Siebold
et al. [127] have found that except kerosene and oleyl alcohol,
the biocompatibility of other chemicals tested was unsatisfac-
tory for various Lactobacilli studied. Demirci et al. [39] tested
toxicity of various solvents in lactic acid biolm fermenta-
tions with Lactobacillus casei performed in solvent-saturated
media and found solvent with 5 vol.% of TBP in isooctane
acceptable. This result compared with TBP toxicity to fungi,
where TBP was found lethal, shows that toxicity tests should
be done for every microorganism of interest and cannot be
extrapolated.
Tong et al. [264] suggest to use a solvent with relatively
toxic extractant TOMAC in oleyl alcohol and remove dis-
solved TOMAC in adsorption column with ion-exchange
resin Amberlite IR-120B before returning the broth to fer-
menter. This resulted in satisfactory extractive fermentation
of lactic acid. A similar approach, but with L/L extraction of
secondary amine dissolved in the broth by oleyl alcohol in
the second stage before its returning to fermenter, was used
by Jin and Yang [209] in propionic acid production. Immo-
bilization of cells into calcium alginate gell reduces toxicity
of the solvent to microorganisms [198].
e
A
o,e
k
Sb
D
+
1
r
e
(1)
where the overall mass-transfer resistance is composed of
four individual resistances. The overall mass-transfer coef-
cient is dened for concentrations in the aqueous phase and
the effective surface area of L/L interface
n
e
= K
e
A
ei
c
F
c
S
D
S
(2)
The kinetics of formation and decomposition of the
permeantextractant complex(es) via interfacial reactions
can be in the rst approximation described by the rst-order
rate equations [12,29,90,100]
n
e
= r
e
A
i,e
e
c
FS
(3)
n
s
= r
s
A
i,s
s
c
SR
(4)
s
A
o,s
k
Sb
+
A
i,s
A
w,s
k
Sw
+
1
r
s
(5)
Eqs. (1) and (5) represent the diffusion-reaction models of
MBSE and MBSS, and without the last terms for resistance
based on the reaction kinetics, they represent the diffusion
models for the estimation of the overall mass-transfer resis-
tance or the overall mass-transfer coefcients in these pro-
cesses. The estimation of the individual mass-transfer coef-
cients is discussed in papers [5,10,11,100].
Mass-transfer characteristics of two-phase HF contactors
in several systems are presented in Table 13. Data for contac-
tors Liqui Cel (Celgard) with cross-ow of phases of labora-
tory size 2.5 in. 8 in. and pilot plant contactors 4 in. 13 in.
and 4 in. 28 in. (which can be used also for smaller produc-
tion plants) are available for some systems. The values of
the lumped rate constants of extraction and stripping reac-
tions, dened by Eqs. (3) and (4), for some organic acids and
solvents are listed in Table 14.
It is evident from Table 13 that in most systems the
value of the overall mass-transfer coefcient in stripping is
lower, not seldom one order of magnitude, comparing with
MBSE. This is connected with a slower decomposition of the
extractantacid complex on stripping interface. The values of
the lumped rate constants on extraction interface are higher
than the rate constants for stripping reactions, as documented
in Table 14. The only exception among systems shown in
Table 13 is lactic acid with the low value of the distribution
coefcient reecting weaker interaction between lactic acid
and the extractant.
The analysis of mass-transfer resistances showed, that the
kinetics of the acidcarrier complexes formation and/or de-
composition should be taken into account in most systems.
The contribution of the mass-transfer resistances based on
reaction kinetics to the overall resistances in MBSE and
MBSS of MPCAwith the solvent with TOAis up to about 50
and 60%, respectively [90]. The proposed reaction-diffusion
model gives a much better t to the experimental values than
a pure diffusion model. The mean values of the lumped rate
constant of the extraction and stripping reactions for MPCA
are shown in Table 14. The contribution of the diffusion resis-
tance in the feed and the solvent boundary layer is small and is
less than 10% (under reasonable hydrodynamic conditions).
Juang et al. [302] estimated that the mass-transfer resis-
tances based on reaction kinetics participate in the overall
resistance in pertraction of lactic acid through SLM by 30 to
80% depending on acid and carrier concentrations. Lazarova
et al. [145] did analysis of the mass-transfer resistances in
MBSE and MBSS of penicillin G with the solvent contain-
ing the secondary amine Amberlite LA2 and concluded that
the mass-transfer resistance based on reaction kinetics can
participate in the overall resistance in MBSE up to 32% and
in MBSS up to 79% depending on the process parameters.
A short-cut method for the design and simulation of two-
phase HFcontactors with variable mass-transfer and distribu-
tion coefcients taking into account reaction kinetics in the
stripping interface is presented in paper [100] in this issue
and method considering reaction kinetics in both interfaces
in paper [299].
5. Case studyrecovery of MPCA
5-Methyl-2-pyrazinecarboxylic acid (MPCA) is a valu-
able acid of industrial importance. The waste solution of
MPCA resulting from downstream processing of the enzy-
matic resolution technology contains much mineral salts and
their pHis below2. MPCAhas to be concentrated fromabout
16 kg m
3
by a factor of 10, at least. From this concentrate
pure MPCA is recovered as an organic phase formed after
decreasing its pH<2 (low water solubility of undissociated
MPCA, the same refers also to DMCCA).
Mass-transfer characteristics of HF contactors in MBSE
and MBSSof MPCAare presented in paper [90]. The concen-
tration dependence of the distribution coefcient of MPCA
in the solvent used has a favourable course with increasing
value of D with decreasing acid concentration [89,271]. This
is an opposite situation comparing with butyric acid, where
the value of the distribution coefcient decreases with de-
creasing concentration [253]. This has a positive effect on
the length of bers in HF modules needed to achieve required
yield.
Compositions of streams in laboratory tests and supposed
also in a pilot unit for recovery of MPCA from process solu-
tion were following:
Feed: aqueous solution containing of about 0.12 kmol m
3
MPCA, 1 kmol m
3
Na
2
SO
4
; with pH
F
kept constant at 2.5
by the addition of H
2
SO
4
.
Solvent: 0.4 kmol m
3
TOA in xylene.
2
5
6
S
.
S
c
h
l
o
s
s
e
r
e
t
a
l
.
/
S
e
p
a
r
a
t
i
o
n
a
n
d
P
u
r
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
4
1
(
2
0
0
5
)
2
3
7
2
6
6
Table 13
Mass-transfer characteristics of HF contactors in MBSE and MBSS of different organic acids based on experimental data
Solute Solvent Contactor type c
F0
(mol m
3
) D at c
F0
u
F
(cms
1
) K
e
10
6
(ms
1
) c
S1
(mol m
3
) u
S
(cms
1
) K
s
10
6
(ms
1
) Literature
Butyric acid 0.4 kmol m
3
TOA in n-alkanes CF HF
a
100 3.0 1.78 6.0 150 0.084 2.4
a,e
[87,98100]
300 3.5 1.78 9.1
a,d
DMCCA S1 CF HF
a
127 31.7 0.37 10.5 678.9 0.074 0.81
a
[87,305]
MPCA 0.4 kmol m
3
TOA/xylene CF HF
a
100 3.9 1.34 8.56
c
0.084 1.7
a,d
[89,90,111]
Lactic acid S2 CF HF
a
543 0.79 0.14 0.30 102 0.27 1.0
a
[115]
S3 CF HF
a
545 0.52 0.14 0.32 70 0.11 1.3
a
S4 PF HF 280 0.50 1.33 0.53 0.66 [116]
10% TOPO in kerosene PF HF
+
0.330.5
+
[117]
S5 PF HF 100 0.24 0.5 1.02 [122]
TOMAC in decanol CF HF
a
154 0.032
f
1 10
4f
[124]
Valeric acid S6 49 1.9 1.81 15.7 0.47 [106]
4-Methyl-thiazole Benzene 354 21.8 1.2 15.0 0.32 [155]
Cyano-thiazole Benzene 118 6.0 1.2 4.3 0.32 [155]
Phenyl-alanine 0.75 kmol m
3
D2EHPA in n-alkanes CF HF
a
50 2.5 3 2.5
a,d
10 0.08 0.60
a
[137]
30 0.89
a
S7 CF HF
c
0.288
c
0.077
c
[199]
40 mol m
3
Aliquat 336 in xylene 40 1.3 0.100.23 [136]
Penicillin G S8 CF HF
b
13.3 1.11
b
0.13
b
[145]
S9 PF HF 10 15 1.14 [147]
Phenol MIBK PF HF 0.18 [168]
Isopropyl-benzene CF HF
c
320 1.25
c+
3.5
c+
Based on data [86]
80 1.25
c+
9.8
c+
S10 PF HF 2.77 [162]
30% TBP in kerosene 7 6.3
x
[306]
7 14.5
xx
1-Decanol CF HF
a
31.9 25.4 4.8
f
2.2 10
3f
[124]
p-Nitro-phenol 1-Octanol PF HF 0.72 89.2
e
6.5 [91]
0.2 [92]
Sulfanilic acid S11 PF HF 6.14 >26 0.33 1.6 0.073 [158]
0.20 1.1 0.073
Phenyl-glycine TOMAC in decanol CF HF
a
16.4 0.33 0.5
f
0.053
f
[124]
Abbreviations used are explained in Nomenclature and under this table. Contactors: Liqui Cel Extra Flow (Celgard) with cross-ow of phases, size: a: 2.5 in. 8 in.; b: 4 in. 13 in.; c: 4 in. 28 in.; d: circulation
method; e: concentration independent value; f: specic formulation of the model equations; +: the organic phase ows in bers, x: straight bers; xx: coiled bers. Solvents: S1: 0.4 kmol m
3
Hostarex A327,
30 wt.% isodecanol in n-alkanes; S2: 0.4 kmol m
3
Hostarex A327, 0.8 kmol m
3
isotridecanol in n-alkanes; S3: 0.4 kmol m
3
Hostarex A327, 0.8 kmol m
3
isodecanol in n-alkanes; S4: 30% Aliquat 336 in
Shellsol; S5: 0.5 kmol m
3
TOMAC in oleyl alcohol; S6: 10% Amberlite LA2 in toluene; S7: 10 vol.% D2EHPA in kerosene; S8: 0.115 kmol m
3
Amberlite LA-2, 10% isodecanol in kerosene; S9: 60 mol m
3
Amberlite LA-2 in butylacetate; S10: Cyanex 923 in kerosene; S11: 20% TOA, 30% octanol in kerosene.
conv
=0.70 follows
V
R2
= 9.0 Lh
1
, c
R2
=1.16 kmol m
3
.
The number of contactors Liqui Cel 4 in. 28 in. (Cel-
gard, with an effective length of bers of 0.6 m) [295]
in series, as resulted from simulations for the mentioned
process data, was found 2 for both MBSE and MBSS,
what are reasonable numbers. The technological owsheet
of a pilot plant unit for recovery of MPCA is shown in
Fig. 9. These results document potential of application of
HF contactors in recovery of MPCA from waste solu-
tions.
6. Conclusions
Several processes based on partitioning of components on
one or two L/Linterfaces have been designed to achieve sepa-
ration of mixtures, and numerous papers on recovery and sep-
aration of organic acids have been published. Not so much of
this is reected in industrial applications of MBSEor pertrac-
tion. It is hoped that potential for further progress exists. As
shown, some applications are matured enough to be applied
in future. This is especially probable for more hydropho-
bic acids like butyric and higher acids, DMCCA, MPCA,
phenylalanine, etc. Development in extractive fermentation
process for production of propionic acid shows its potential,
as well.