1-s2.0-S1226086X14001117-main

8/10/2019 1-s2.0-S1226086X14001117-main

1/6

Liquidliquid extraction of succinic acid using a hollow fibermembrane contactor

Luciana de Souza Moraes *, Frederico de Araujo Kronemberger,Helen Conceicao Ferraz, Alberto Claudio Habert

Chemical Engineering Program COPPE, Federal University of Rio de Janeiro, P.O. Box 68502, CEP 21941-972 Rio de Janeiro, RJ, Brazil

1. Introduction

The demand for organic acids produced by fermentation has

grown due to their use in the manufacture of biodegradable

polymers, being poly(lactic acid) the most prominent [1].

Separation is still a major cost factor: the main issue in removing

organic acidsfrom aqueous streams as in fermentationbroths is

their low concentration and the variety of substances present in

the broth[2]. As an alternative to the highly expensive precipita-

tion method currently used, several processes have been proposed,

such as adsorption, ion exchange, liquidliquid extraction and

membrane separation[36]. Among them, membrane processes

seem to be technically and economically promising. Non-disper-

sive solvent extraction using hollow fiber membrane contactors

(also referred as membrane contactor extraction), liquid mem-

branes and facilitated transport have been extensively studied for

organic acids separation from fermentation broths [710].Extraction with membrane contactors, using an organic

compound as extraction agent, has many advantages over

conventional liquidliquid extraction. Membrane acts as a barrier

which avoids dispersion of the liquid phases and also enhances

their contact area. Moreover, the process is modular and has low

energy consumption[11]. Despite all potential advantages of this

process, a good extraction also relies on selecting a suitable

extractant. It needs to be highlyselective for the component andto

have a good extraction capacity. Also, it is desirableto combine low

cost with low viscosity, a moderate interfacial tension, good

biological degradability and low toxicity[12]. Many works report

the evaluation of hydrocarbons, alcohols, amines and phosphorous

compounds as extractants for organic acids extraction. For the first

two, the extraction is physical, and the distributioncoefficients are

small[13]. Amines form complexes with organic acids, therefore

the extraction is reactive. Phosphorous compounds have been

combined with amines to enhance the distributioncoefficients, but

high concentrations are required, increasing costs [14]. To improve

the solubility of the complexes, hydrocarbons and alcohols are

added as diluents to theextractantmixture.They also modify some

properties of the extractants, such as viscosity [15]. Novelextractants, especially ionic liquids, have been evaluated, although

only a few have presented results as good as trioctylamine (TOA),

the most common extractant for organic acids extraction[1618].

For liquidliquid membrane extraction, porous membranes are

often used, due to the lower mass transfer resistance compared to

dense membranes. However, in this case, a rigorous pressure

control is necessary to avoid disruption of the interface, and

partially soluble extractants (as for example, primary amines)

cannot be used [19]. To overcome eventual limitations of mass

transfer in dense membranes, an appropriate swelling may be

Journal of Industrial and Engineering Chemistry 21 (2015) 206211

A R T I C L E I N F O

Article history:

Received 21 October 2013

Accepted 16 February 2014

Available online 22 February 2014

Keywords:

Succinic acid

Membrane contactor

Hollow fiber

Extraction

Amine

A B S T R A C T

Extraction with membrane contactors is an attractive technological alternative for organic acidseparation from fermentation broths. This work aims at studying the recovery process of succinic acid,

integrating both extraction and re-extraction steps. It was proposed to use a non-porous membrane,

which allows more flexibility of operation without loss of efficiency. For the integrated process, a 5-fold

increase in the recovery of acid from the feed stream was observed when compared to the single

extraction process, leading to an overall recovery close to 50%.

2014 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights

reserved.

Theauthorswouldlike to dedicate this paper to ProfessorAlberto Luiz Coimbra,

in the 50th anniversary of COPPE (19632013), the Graduate School of Engineering

of the Federal University of Rio de Janeiro.

* Corresponding author. Tel.: +55 21 2562 8354.

E-mail address: [email protected] (L.d.S. Moraes).

Contents lists available at ScienceDirect

Journal of Industrial and Engineering Chemistry

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

http://dx.doi.org/10.1016/j.jiec.2014.02.026

1226-086X/ 2014 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jiec.2014.02.026mailto:[email protected]://www.sciencedirect.com/science/journal/1226086Xhttp://www.elsevier.com/locate/jiechttp://dx.doi.org/10.1016/j.jiec.2014.02.026http://dx.doi.org/10.1016/j.jiec.2014.02.026http://www.elsevier.com/locate/jiechttp://www.sciencedirect.com/science/journal/1226086Xmailto:[email protected]://dx.doi.org/10.1016/j.jiec.2014.02.026http://crossmark.crossref.org/dialog/?doi=10.1016/j.jiec.2014.02.026&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1016/j.jiec.2014.02.026&domain=pdf

8/10/2019 1-s2.0-S1226086X14001117-main

2/6

promoted, in addition to a reduction in the membrane thickness.

Lee et al.[20]already reported that extraction of solutes in liquid

liquidsystems using dense membranesis feasible,but thechoice of

a solvent with a good interaction with the membrane material is

important.

This work aims at studying the extraction of succinic acid,

investigating both the extraction to the organic phase and the re-

extraction into a new aqueous phase using a hollow fiber dense

membrane module as a contactor. The objective is to obtain a

purified acid, improving extraction efficiency. Capacity and

selectivity of mixtures of 1-octanol and several amines were

evaluated and the overall recovery of acid was measured for acirculating batch feed in a 2-step extraction system.

2. Experimental

To recover succinic acid (Sigma Aldrich, 99%), the extractants

used were 1-octanol (Vetec), Primene JM-T1 (a C16C22 primary

amine, Rohm and Haas, 80%), Primene TOA1 (C8 primary amine,

Rohm and Haas, 99%), n-Butyldiethanolamine (n-BDEA, Sigma

Aldrich, 98.6%) and a trioctylamine (tris(2-ethylhexyl)amine,

TEHA, TCI America, 93%). Because amines are viscous and

corrosive, 1-octanol was used as diluent. The amine concentration

in the organic phase was set in 10% w/w. All products were used as

received from the manufacturers. For the membrane extraction, a

cellulose diacetate membrane module (Dicea 1701

, BaxterHealthcare) was selected. It was packed with dense hollow fibers,

with 200mm of inner diameter, 15 mm of wall thickness and

1.7 m2 of contact area.

Succinic acid solutions were prepared with concentrations

ranging from 0.01 to 0.5 mol/L (160 kg/m3), which include the

typical values found in fermentation broths [21]. Organic acid

concentration in the aqueous phase was determined by HPLC

(Shimadzu, SCR 102H column, H3PO4 as mobile phase, flow rate

0.6 mL/min, 80 8C, RI detector).

2.1. Equilibrium experiments

Equilibrium tests were performed at 25, 40 and 60 8C. Equal

volumes of succinic acid aqueous solution and extractant were

charged in 50 mL polypropylene vessels and stirred at 120 rpm for

1 h. The mixture was then centrifuged at 5000gfor 30 min to

allow complete separation of the organic and aqueous phases.

Succinic acid concentration in the aqueous phase was determined

by chromatography and the concentration in the organic phase

was calculated from a mass balance. Distribution coefficients (KD)

were calculated, according to Eq. (1).

KD

H2Sucorg;eq

H2Suc aq;eq (1)

where [H2Suc]org,eqand [H2Suc]aq,eqare, respectively, the equilib-

rium concentration in organic and aqueous phases.

The influence of the pH of the aqueous phase used for re-

extraction on the recovery efficiency was also evaluated. Equal

volumes of aqueous and organic phases (500 mL each) were

charged in a vessel with controlled stirring (120 rpm) and

temperature (25 8C). Succinic acid concentration in aqueous phase

was 0.17 mol/L (20 kg/m3) and the amine concentration in 1-

octanol was 10%, w/w. Allowing 3 h for the equilibrium establish-

ment, the phases were then separated by centrifugation (5000g

for 30 min), and the organic phase was collected and stored.

Samples were taken from the aqueous phase before and after

extraction to determine thepercentage of acid extracted. The pH ofthe re-extraction was changed by adding 0.1 mol/L hydrochloric

acid or 0.1 mol/L sodium hydroxide solutions. Equal volumes of the

stored organic phase and the re-extraction water were added to

50 mL polypropylene vessels. The procedurewas thesame as in the

extraction experiments. The re-extraction efficiency (RE) was

calculated as the ratio between the concentrations of acid in the

aqueous phase after re-extraction and in the organic phase, as

shown in Eq.(2).

RE % H2Sucaq;eqH2Sucorg;o

100 (2)

2.2. Extraction experiments in the 2 ways membrane contactor

system

For membrane contactor extraction (MCE), a system containing

two coupled hollow fiber modules was set (Fig. 1), allowing

simultaneously extraction of the acid from an aqueous phase (feed)

to an organic phase (extractant) and re-extraction from this organic

phase into a new aqueous phase (recovery), and therefore, succinic

acid extraction could be achieved. This approach is expected to be

very effective when the acid is extracted from a fermentation broth

(a multicomponent system).In this work, tests wereperformed with

a solution of pure succinic acid, with the purpose of evaluating the

performance of the selected membrane and extractants, without

interference of other solutes. The system configuration allows its

operation in extraction or coupled (extraction + re-extraction)

modes. Therefore, experiments were performed in both operationalmodes, and their results were compared.

For these tests, operation was performed with batches of

succinic acid (0.17 mol/L) and extractant (10%, w/w amine in 1-

octanol) solutions, and pure water for re-extraction. All solutions

were allowed to recirculate. The extractant flows inside the fibers,

at a flow rate of 105 m3/s, corresponding to a Reynolds number

(Re) of 0.4, while the aqueous phases flow externally to the fibers

(succinic acid solution flows in the first module, and pure water in

the second one), at a flow rate of 3105 m3/s (Re= 19). The

efficiency of this process is calculated in terms of the percentage of

recovery, Recov (Eq.(3)):

Recov % mH2Suc;rext

mH2Suc;F 100 (3)

Nomenclature

[Ext]T total amine concentration in the organic phase

(mol/L)

[H2Suc]aq succinic acid concentration in the aqueous phase

(mol/L)

[H2Suc]org succinic acid concentration in the organic phase

(mol/L)KD distribution coefficient, defined by Eq.(1)

mH2Suc succinic acid mass in aqueous phase (kg)

RE re-extraction efficiency (%), defined by Eq.(2)

Recov succinic acid recovery in membrane contactor (%),

defined by Eq.(3)

Z loading, defined by Eq.(4)

Subscripts

eq equilibrium

ext extraction

F feed

o initial

rext re-extraction

L.S. Moraes et al. / Journal of Industrial and Engineering Chemistry 21 (2015) 206211 207

8/10/2019 1-s2.0-S1226086X14001117-main

3/6

3. Results and discussion

3.1. Liquidliquid extraction (LLE)

Succinic acid concentrations after equilibrium in aqueous and

organic phases were obtained for several initial concentrations in

the aqueous phase.Table 1shows the percentage of acid extracted

foreachmixture used, fora 0.17 mol/Lsuccinic acid solution. For n-

BDEA and Primene TOA1 extractants, extraction was negligible

due to emulsion formation. The best extractant was Primene JM-

T1, which extracted 86.0% of the acid.

The temperature effect on equilibrium isotherms was evaluated

for all extractants, despite the observed emulsion formation at

25 8C for Primene TOA1 and n-BDEA amines. Results are presented

inFigs. 24. For 1-octanol, the relationship between succinic acid

concentration in organic and aqueous phases is linear through the

entire concentration range, and extraction is not influenced by

temperature, since it is a physical extraction. For reactive

extraction with amines, the relation between concentrations in

aqueous and organic phases, at equilibrium, is not linear,

particularly at high concentrations. This is clearly a deviation

from a Henrys law type isotherm (typical for physical extraction),

as described by Wasewar et al. [22]. Moreover, these results appear

to indicate that at higher temperatures, the reaction mechanism

may be different, and that another type of complex between

succinic acid and amine could be formed, as also suggested by

Tamada et al. [15] and Tamada and King [3]. Increasing thetemperature from 25 to 40 8C did not improve extraction efficiency

when primary amines (Fig. 3) were used as extractants. Also, for

diluted succinic acid aqueous solutions extracted by Primene JM-

T1, temperature did not affect extraction at all (Fig. 4).

In the case of extraction with tertiary amines, the effect of

temperature on extraction is more pronounced. For n-BDEA,

extraction was significant only at the highest temperature

investigated (60 8C). At this temperature, the relationship between

the equilibrium concentrations in aqueous and organic phases was

linear, and the distribution coefficient obtained is close to 1. For all

other conditions, the behavior is very close to that obtained for 1-

octanol, an indication that the acid in the organic phase was

extracted by the diluent, being the exception the highest

concentration, at 40 8

C. The TEHA isotherms did not show a

[

Fig. 2. Effect of temperature on succinic acid LLE with 1-octanol.

[

Fig. 3.Effect of temperature on succinic acid LLE with the primary amines Primene

JM-T1 (a) and Primene TOA1 (b). Amine concentration in 1-octanol: 10% (w/w).

[

Fig. 1.Extraction system representation.

Table 1

Succinic acid removal from aqueous phase by LLE.

Extractant 1-Octanol Primene JM-T1 Primene TOA1 n-BDEA TEHA

Removal (%) 15.9 86.0 4.8 3.7 32.0

Amines were diluted in 1-octanol, in a proportion of 10% (w/w). Acid concentration

in aqueous phase: 0.17 mol/L. Temperature: 258C.

L.S. Moraes et al. / Journal of Industrial and Engineering Chemistry 21 (2015) 206211208

8/10/2019 1-s2.0-S1226086X14001117-main

4/6

particular behavior, which may be interpreted as a change in the

complexes stoichiometry, and this hypothesis should be further

investigated.

Values of distribution coefficients (KD) obtained for the

extractants are summarized in Fig. 5. For comparison, a fixed

initial concentration (0.17 mol/L) was assumed for calculations. It

can be clearly seen that Primene JM-T1was the best extractant for

succinic acid LLE. Its KD value was more than five times higher than

the second best extractant, TEHA, at 25 8C. It is also possible to

check that at the highest temperature, the efficiency of all amines

increased considerably. For the case of Primene TOA1 andn-BDEA

amines, this behavior may be due to the fact that reactionmechanism changes with increasing temperature. The KD values

for 1-octanol and TEHA amines are similar to those found in the

literature. Li et al. [13] obtained a distribution of 0.26 when 1-

octanol was used for succinic acid extraction, at 25 8C, and Huh

et al. [21] and Song et al. [23] reported a KD of 0.3 with a mixture of

6.5% (w/w) of tri-n-octylamine in octanol.

Fig. 6 shows the dependence of KD on the succinic acid

concentration in aqueous phase and temperature for the two more

suitable extractants evaluated: Primene JM-T1 and TEHA. For the

primary amine, KD value decreases with increasing acid concen-

tration in the aqueous phase. Additionally, the results for 25 and

40 8C seem to indicate that its extraction capacity approaches a

limit. One way to identify this behavior is to obtain the loading

curves. According to Wasewar et al. [22], loading (Z) is defined as

shown in Eq. (4), in which [H2Suc]org is the succinic acid molar

concentration in organic phase after extraction and [Ext]T is the

total amine concentration (kmol/m3). Loading curves for Primene

JM-T1 and TEHA are shown inFig. 7.

ZH2Sucorg

ExtT(4)

For the primary amine (Fig. 7a), loading curves for 25 and 40 8C

tendto reach a plateau atZ= 1, confirming that the amine achieves

its limit (saturation). Thesame behavior was not observed at 60 8C,

when loading exceeds 1 at [H2Suc]aq,o= 0.5 mol/L. This is an

indicative that complexes with more than one acid molecule per

amine molecule are formed. Wasewar et al. [22] states that for

concentrated acid solutions, when loading is higher than 0.5, it is

possible an acid/amine ratio different from 1. Therefore, it couldbe

inferred from this study that n:1 (with n >1) stoichiometries, in

which n is the number of acid molecules in the complex, are

formed in the extraction with Primene JM-T1 amine, while for

[

Fig. 4.Effect of temperature on succinic acid LLE with the tertiary amines n-BDEA

(a) and TEHA (b). Amine concentration in 1-octanol: 10% (w/w).

[

Fig. 5.Temperature dependence on distribution coefficient for succinic acid LLE for

all extractants evaluated. Amine concentration in 1-octanol: 10% (w/w).

[

Fig. 6. Temperature dependence on distribution coefficient for succinic acid LLE.

Amine concentration in 1-octanol: 10% (w/w). Succinic acid initial concentration in

aqueous phase varies from 0.01 to 0.5 mol/L.


8/10/2019 1-s2.0-S1226086X14001117-main

5/6

tertiary amine TEHA (Fig. 7b) this is improbable, although this

possibility could not be entirely ruled out.

For Primene JM-T1 diluted in 1-octanol, re-extraction was

evaluated as a function of the pH of the aqueous phase. As a first

step, the liquidliquid extraction resulted in an efficiency of 86.3%.

The organic phase loaded with succinic acid was then used in the

re-extraction experiments. Results are presented in Fig. 8.

Observing the pH of the aqueous phase after extraction, it was

possible to identify a buffering region in which the recovery

efficiency was extremely low. When the initial pH was very basic

(above 13), it was observed a significant increase in recovery of the

acid, with an efficiency of about 60%. Some authors, such Lee et al.[24], Keshav and Wasewar [25], have obtained similar results

studying the effect of the addition of sodium ions for the re-

extraction of succinic and propionic acids, respectively, from an

organic phase containing trioctylamine (TOA) in alcohols. The

former pointed out that in a basic environment, when the pH is

higher than pKa1 and pKa2 of succinic acid, it is possible to back

extract the acid by destroying the complex and forming a sodium

succinate salt. The high solubility of this salt in the aqueous phase

explains the increase in the efficiency of recovery.

3.2. Extraction experiments with the membrane contactor

In this work, a non-porous membrane was selected since the

membrane provides a barrier which prevents contact between

liquid phases, and consequently, emulsion formation. Thus, a

variety of extractants can be evaluated.Fig. 9shows the efficiencyof acid recovery with the membrane contactor for the investigated

extractants. This recovery is defined as the ratio between succinic

acid mass in the aqueous re-extraction phase and the mass in the

initial aqueous phase (feed), as shown in Eq. (3). For n-BDEA amine,

the presence of succinic acid in aqueous re-extraction phase was

observed only after 6 h, which makes n-BDEA not suitable for

succinic acid extraction, even when a non-dispersive method, such

as the extraction with a membrane contactor, is adopted.

It is possible to observe that equilibrium was reached in

approximately 2 h, the only exception being Primene TOA1 amine.

The initial permeate fluxes obtained for succinic acid are close to

3106 kg/(m2 s). Water passage to the organic phase was only

perceived after long-term extraction. Although the membrane

employed in this study is hydrophilic, its swelling degree in theaqueous phase is low (about 8% in mass). This swelling resulted in

insignificant water content in organic phase at the end of the

experiments, and for this reason, it was not quantified. Organic

phase by-passing through the membrane was not observed.

[

Fig. 7. Loading curve for succinic acid LLE with (a) Primene JM-T1

and (b) TEHA,both diluted in 1-octanol, in a proportion of 10% in mass.

[

Fig. 8. Effect of the pH of the aqueous re-extraction phase on the efficiency of

succinic acid recovery from an organic phase consisting of 10% (w/w) Primene JM-

T1 in 1-octanol.

[

Fig. 9. Succinic acid recovery in MCE. Amine concentration in organic phase: 10%

(w/w). Succinic acid initial concentration in aqueous phase: 0.17 mol/L.

L.S. Moraes et al. / Journal of Industrial and Engineering Chemistry 21 (2015) 206211210

8/10/2019 1-s2.0-S1226086X14001117-main

6/6

Three extractants presented a recovery of about 45%: 1-octanol,Primene TOA1 and TEHA. To better understand this result, the

extraction was analyzed separately, and the result is shown in

Fig. 10. Primene JM-T1 extracted 77% of the total acid in the feed.

For the other amines chosen, the results are similar to that

obtained for the 1-octanol. This couldbe explainedby the fact that,

as there is no chemical reaction, the step of re-extraction is more

effective for 1-octanol. In the case of the amines, as acid

accumulationin the organic phase occurs, the reaction equilibrium

is reached, and consequently the amount of acid removed from the

feed is lower. To improve the re-extraction efficiency for amines, it

would be necessary to add sodium ions to the aqueous re-

extraction phase, as shown in the experiments using direct contact

(Fig. 8). However, this would lead to the production of succinate

salts, which is not desired in most applications, since an additionalprocess would be necessary to recover the acid. Another way to

improve the recovery is exploring the variation of temperature

(either in extraction or re-extraction), for example.

At the experimental conditions of these tests, TEHA perfor-

mance was similar to that exhibited by 1-octanol. However, it is

important to emphasize that the alcohol alone is not selective

toward organic acids. The use of amines is justified by the

possibility of extracting the organic acid directly from fermenta-

tion broths. For this reason, this extractant was chosen to evaluate

the gain in efficiency when both extraction and re-extraction steps

are coupled,comparing results with the efficiency of the extraction

itself. The advantage of coupling extraction and re-extraction steps

is to shiftthe extraction equilibrium, increasing the total amount of

acid recovered from the feed. Results in Fig. 11shown that, whenre-extraction is coupled, the total acid removed from the feed is

almost 5 times higher than the single extraction.

4. Conclusions

Extraction of succinic acid with five different extractants was

successfully accomplished using a non-porous cellulose diacetate

membrane as a contactor. Liquidliquid extraction tests indicated

that the influence of temperature on the distribution coefficients

leads to different behaviors when 1-octanol and amines are

compared. Primene JM-T1 amine was the best succinic acid

extractant among those evaluated, removing 86.0% of the succinic

acid in a 0.17 mol/L aqueous solution. For membrane contactor

extraction, 45% of acid recovery was achieved, using water as re-

extraction agent. This value can be improved, especially forPrimene JM-T1 amine, by using a sodium solution in re-extraction

or by changing the temperature. Results presented in this work

indicate that extraction with non-porous membrane contactor for

succinic acid extraction is feasible and advantageous since there

are fewer restrictions in the extractant choice.

Acknowledgments

The authors would like to thank to Rohm and Haas, who kindly

donated Primene JM-T1 and Primene TOA1 amines, and to the

Brazilian Council for Scientific Development, CNPq (Conselho

Nacional de Desenvolvimento Cientfico e Tecnologico), for

financial and research support.

References

[1] M. Matsumoto, T. Otono, K. Kondo, Sep. Purif. Technol. 24 (2001) 337 .[2] M. Rodriguez, M.J. Gonzalez-Munoz, S. Luque, J.R. Alvarez, J. Coca, J. Membr. Sci.

274 (2006) 209.[3] J.A. Tamada, C.J. King, Ind. Eng. Chem. Res. 29 (1990) 1333 .[4] C. Abels, F. Carstensen, M. Wessling, J. Membr. Sci. 444 (2013) 285.[5] T. Kurzrock, D. Weuster-Botz, Biotechnol. Lett. 32 (2010) 331.[6] P. Pal, J. Sikder, S. Roy, L. Giorno, Chem. Eng. Process. 48 (2009) 1549 .[7] R. Prasad, K.K. Sirkar, AIChE J. 34 (2) (1988) 177.[8] S. Alex, F. Biasoto, G. Aroca, Desalination 241 (2009) 337 .[9] A.K. Pabby, A.M. Sastre, J. Membr. Sci. 430 (2013) 263 .

[10] A.-I. Galaction, M. Postaru, D. Cascaval, L. Kloetzer, Solvent Extr. Ion Exch. 31(2013) 171.

[11] R. Klaassen, P. Feron, A. Jansen, Desalination 224 (2008) 81 .[12] K.L. Wasewar, Chem. Biochem. Eng. Q. 19 (2) (2005) 159.[13] Z. Li, W. Qin, M. Wang, Y. Huang, Y. Dai, Chin. J. Chem. Eng. 10 (3) (2002) 281 .

[14] K.L. Wasewar, A.A. Yawalkar, J.A. Moulijn, V.G. Pangarkar, Ind. Eng. Chem. Res. 43(2004) 5969.

[15] J.A. Tamada, A.S. Kertes, C.J. King, Ind. Eng. Chem. Res. 29 (1990) 1319.[16] A. Krzyzaniak, B. Schuur, M. Sukumaran, H. Zuilhof, A.B. de Haan, Chem. Eng.

Trans. 24 (2011) 709.[17] M. Blahusiak, S. Schlosser, J. Martak, Sep. Purif. Technol. 119 (2013) 102.[18] A. Krzyzaniak, M. Leeman, F. Vossebeld, T.J. Visser, B. Schur, A.B. de Haan, Sep.

Purif. Technol. 111 (2013) 82.[19] S. Schlosser, R. Kertesz, J. Martak, Recovery, Sep. Purif. Technol. 41 (2005) 237.[20] L.T.C. Lee, W.-S. Ho, K.-J. Liu, Membrane solvent extraction, US Patent No.

3,956,112 (1976).[21] Y.S. Huh, Y. Jun, Y.K. Hong, H. Song, S.Y. Lee, W.H. Hong, Process Biochem. 41

(2006) 1461.[22] K.L. Wasewar, A.B.M. Heesink, G.F. Versteeg, V.G. Pangarkar, J. Chem. Techn.

Biotechnol. 77 (2002) 1068.[23] H. Song, Y.S. Huh, S.Y. Lee, W.H. Hong, Y.K. Hong, J. Biotechnol. 132 (2007) 445.[24] E.Z.Lee, Y.S.Huh, Y.-S. Jun,H.J. Won, Y.K.Hong,W.H. Hong, Biotechnol.Bioprocess

Eng. 13 (2008) 342.

[25] A. Keshav, K.L. Wasewar, Chem. Eng. Sci. 65 (2010) 2751.

[

Fig. 10. Efficiency in succinic acid extraction in a membrane contactor. Amine

concentration in organic phase: 10% (w/w). Succinic acid initial concentration in

aqueous phase: 0.17 mol/L.

[

Fig. 11. Comparison between extraction and coupled process in a membrane

contactor, for TEHA/1-octanol extractant. Amine concentration in organic phase:

10% (w/w). Succinic acid initial concentration in aqueous phase: 0.17 mol/L.

http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0005http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0010http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0010http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0010http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0010http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0010http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0010http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0015http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0020http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0025http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0030http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0035http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0040http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0045http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0050http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0050http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0055http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0060http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0065http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0070http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0070http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0075http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0080http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0080http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0080http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0080http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0085http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0085http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0085http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0085http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0085http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0085http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0085http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0090http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0090http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0090http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0090http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0095http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0095http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0095http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0095http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0095http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0095http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0095http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0105http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0105http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0110http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0110http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0115http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0120http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0120http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0125http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0125http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0120http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0120http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0115http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0110http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0110http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0105http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0105http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0095http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0090http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0090http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0085http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0080http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0080http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0075http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0070http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0070http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0065http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0060http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0055http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0050http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0050http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0045http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0040http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0035http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0030http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0025http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0020http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0015http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0010http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0010http://refhub.elsevier.com/S1226-086X(14)00111-7/sbref0005

Documents

1-s2.0-S1226086X14001117-main