**7. Results**

Typical results obtained in the extraction experiments are shown in Figs 12 and 13, in the first case the aqueous solution contained vanillin and ferulic acid, in the second one vanillin and vanillylamine.

Apparently vanillin was rapidly removed and reached the concentrations corresponding to the partition coefficient measured in batch experiments, whereas the ferulic acid or vanillylamine concentrations in the aqueous phase remained almost constant. The membrane

Recovery of Biosynthetic Products Using Membrane Contactors 639

0 60 120 180

Fig. 14. Vanillin concentration in the organic and aqueous phases vs. time in the counter extraction from buthyl-acetate to water at pH = 12. Solvent tube side, solvent volume 0.5 L,

water volume 0.3 L, solvent flow rate 25 L/h, feed flow rate 45 L/h

**Time** *(min.)*

**Time (***min* **.)**

**Cw Cs**

y = -0,0448x

= 0,9989

R2

0 60 120 180

Fig 14 refers to a counter-extraction experiment at pH 12, and reports the time course of the vanillin concentration in the organic and in the aqueous phases. Due to the favourable partition coefficient vanillin effectively moved towards the aqueous phase. The mass transfer was somewhat slower than in the extraction runs performed in similar conditions,

Fig. 15. Evaluation of the module efficiency through Eq. (20); same data as Fig. 13

0









**ln [...]**

0

1000

2000

3000

**C (***mg/L* **)** 4000

based solvent extraction is thus a good technique for the selective removal of vanillin without removing the substrates of the bioconversion.

Fig. 12. Vanillin and ferulic acid concentrations vs. time in the extraction with buthylacetate. Feed shell side, feed volume 0.5 L, solvent volume 0.3 L, feed flow rate 45 L/h, solvent flow rate 25 L/h, pH 7

Fig. 13. Vanillin and vanillylamine concentrations vs. time in the extraction with buthylacetate. Feed shell side, feed volume 0.5 L, solvent volume 0.3 L, feed flow rate 45 L/h, solvent flow rate 25 L/h, pH 7

based solvent extraction is thus a good technique for the selective removal of vanillin

0 60 120 180

Fig. 12. Vanillin and ferulic acid concentrations vs. time in the extraction with buthylacetate. Feed shell side, feed volume 0.5 L, solvent volume 0.3 L, feed flow rate 45 L/h,

0 60 120 180

Fig. 13. Vanillin and vanillylamine concentrations vs. time in the extraction with buthylacetate. Feed shell side, feed volume 0.5 L, solvent volume 0.3 L, feed flow rate 45 L/h,

**Time (***min* **.)**

Cw, van Cs, van Cw VanNH2

**Time (***min.* **)**

Cw, van Cs, van Cw, fer

without removing the substrates of the bioconversion.

0

solvent flow rate 25 L/h, pH 7

1500

**C (***mg/L* **)**

0

solvent flow rate 25 L/h, pH 7

500

1000

500

1000

**C (***mg/L* **)**

1500

Fig. 14. Vanillin concentration in the organic and aqueous phases vs. time in the counter extraction from buthyl-acetate to water at pH = 12. Solvent tube side, solvent volume 0.5 L, water volume 0.3 L, solvent flow rate 25 L/h, feed flow rate 45 L/h

Fig. 15. Evaluation of the module efficiency through Eq. (20); same data as Fig. 13

Fig 14 refers to a counter-extraction experiment at pH 12, and reports the time course of the vanillin concentration in the organic and in the aqueous phases. Due to the favourable partition coefficient vanillin effectively moved towards the aqueous phase. The mass transfer was somewhat slower than in the extraction runs performed in similar conditions,

Recovery of Biosynthetic Products Using Membrane Contactors 641

(up to 180 min) enzymatic conversion was performed in a conventional stirred tank reactor, the reactor was then connected to the hollow fibre module (as in Fig. 6). Apparently the vanillin was quickly transferred to the solvent and the reaction (vanNH2 → vanillin)

The results were not equally successful in the case of microbial conversion of ferulic acid to vanillin. After the start up of the extraction, the microbial activity seemed to stop almost at all. May be the solvent had a strong inhibitory effect, or the solvent itself was used as a

> Cw van Cw vanNH2 Cs van

0 60 120 180 240 300 360

Asimakopoulou, A.G. & Karabelas, A.J., (2006a). Mass transfer in liquid–liquid membrane-

Asimakopoulou, A.G. & Karabelas, A.J., (2006b). A study of mass transfer in hollow-fiber

Bao, L. & Lipscomb G.G., (2002). Mass transfer in axial flows through randomly packed fiber bundles with constant wall concentration, *Journal of Membrane Science* 204, 207–220 Bao, L. & Lipscomb G. G. (2003). Mass transfer in axial flows through randomly packed fiber

Bird, R.B., Stewart, W.E. & Lightfoot, E.N. (2007). *Transport Phenomena*. John Wiley & Sons.

Elsevier Science B.V., ISBN: 0-444-51175-X, Amsterdam.

based extraction at small fiber packing fractions, *Journal of Membrane Science*

membrane contactors - The effect of fiber packing fraction. *Journal of Membrane* 

bundles, In: *New Insights into Membrane Science and Technology: Polymeric and Biofunctional Membranes,* Bhattacharyya D. & Butterfield D.A. (Editors), pp. 5 – 26,

*Time(min)*

substrate instead of ferulic acid; this point require further investigations.

proceeded with the same kinetics.

0

271,151–162.

*Science* 282, 430-441

**8. References** 

Fig. 16. Coupling extraction and enzymatic conversion

ISBN: 978-0-470-11539-8, New York.

40

80

120

*C (mg/L)*

160

this behaviour can be clearly explained considering that, in this case, due to the partition coefficient value, the mass transfer resistance through the membrane was quite large.

The module efficiency was calculated from the vanillin concentration vs. time data as explained in the theoretical section (paragraph 3.1), i.e. plotting the left side of Eq. (20) vs. time. As shown in Fig. 15, which reports to the same data of Fig. 13, the data were well fitted by a straight line, of course only the points sufficiently apart from the equilibrium were considered, up to 90 minutes of experiment in this case.

From the slope of the straight (*0.0448*) the efficiency was calculated to be *Eff = 2.8 %,* finally the overall mass transfer coefficient, calculated from Eq. (15) was *Kw = 1.3 10-5 m/s*.

Tab. 3 summarizes the results obtained in the same way for all the extraction experiments. There is a pretty good agreement between the experimental and theoretical values of the mass transfer coefficients, which of course refers to the actual interface, the inner surface for aqueous feed flowing through the lumen, and the outer surface for aqueous feed in the shell.


Table 3. Module efficiency, experimental and theoretical values of the mass transfer coefficients in the extraction runs at different aqueous phase and solvent flow rates

The mass transfer coefficients were lower for feed flowing shell side with respect to the feed flowing through the lumen; however, it has to be considered that the interfacial area is substantially larger in the latter case, being the external surface of the fibres. As a result the mass transfer rates were comparable in the two cases.

The relative role played by the various mass transfer resistances involved were estimated from Eqs (25) or (26) based on the theoretical values of the individual mass transfer coefficients. In all cases the main resistance was associated to the aqueous phase boundary layer (70-80%). The membrane resistance too was appreciable (20-25%) while the solvent boundary layer played a minor role.

As expected, behaviour is different in counter-extraction. From the data of Fig. 14, the module efficiency was calculated to be nearly 0.9 and the membrane is responsible for 95% of the overall resistance; the solvent boundary layer played a minor role (nearly 5 %), whereas the aqueous phase mass transfer resistance was completely negligible. Of course in the counter-extraction hydrophilic membranes would be more convenient than the hydrophobic membranes used in this work.

Extraction experiments with the whole bioconversion broth gave nearly the same results, showing that fouling did not represent a serious problem. The coupling of the membrane contactor with the bioreactor exhibited good results for the enzymatic conversion of vanillylamine to vanillin. Fig. 16 reports one test performed as follows: in the first period (up to 180 min) enzymatic conversion was performed in a conventional stirred tank reactor, the reactor was then connected to the hollow fibre module (as in Fig. 6). Apparently the vanillin was quickly transferred to the solvent and the reaction (vanNH2 → vanillin) proceeded with the same kinetics.

The results were not equally successful in the case of microbial conversion of ferulic acid to vanillin. After the start up of the extraction, the microbial activity seemed to stop almost at all. May be the solvent had a strong inhibitory effect, or the solvent itself was used as a substrate instead of ferulic acid; this point require further investigations.

Fig. 16. Coupling extraction and enzymatic conversion

#### **8. References**

640 Mass Transfer - Advanced Aspects

this behaviour can be clearly explained considering that, in this case, due to the partition

From the slope of the straight (*0.0448*) the efficiency was calculated to be *Eff = 2.8 %,* finally

Tab. 3 summarizes the results obtained in the same way for all the extraction experiments. There is a pretty good agreement between the experimental and theoretical values of the mass transfer coefficients, which of course refers to the actual interface, the inner surface for aqueous feed flowing through the lumen, and the outer surface for aqueous feed in the shell.

> Feed tube side 12 10 4.76 1.4 10-5 1.45 10-5 20 20 5.36 1.6 10-5 1.62 10-5 45 25 2.0 1.5 10-5 1.98 10-5 Feed shell side 20 35 4.52 9.7 10-6 7.6 10-6 45 25 2.77 1.3 10-5 9.2 10-6

Kw *(m/s)* Kw theor *(m/s)*

coefficient value, the mass transfer resistance through the membrane was quite large. The module efficiency was calculated from the vanillin concentration vs. time data as explained in the theoretical section (paragraph 3.1), i.e. plotting the left side of Eq. (20) vs. time. As shown in Fig. 15, which reports to the same data of Fig. 13, the data were well fitted by a straight line, of course only the points sufficiently apart from the equilibrium were

the overall mass transfer coefficient, calculated from Eq. (15) was *Kw = 1.3 10-5 m/s*.

η *%*

Table 3. Module efficiency, experimental and theoretical values of the mass transfer coefficients in the extraction runs at different aqueous phase and solvent flow rates

The mass transfer coefficients were lower for feed flowing shell side with respect to the feed flowing through the lumen; however, it has to be considered that the interfacial area is substantially larger in the latter case, being the external surface of the fibres. As a result the

The relative role played by the various mass transfer resistances involved were estimated from Eqs (25) or (26) based on the theoretical values of the individual mass transfer coefficients. In all cases the main resistance was associated to the aqueous phase boundary layer (70-80%). The membrane resistance too was appreciable (20-25%) while the solvent

As expected, behaviour is different in counter-extraction. From the data of Fig. 14, the module efficiency was calculated to be nearly 0.9 and the membrane is responsible for 95% of the overall resistance; the solvent boundary layer played a minor role (nearly 5 %), whereas the aqueous phase mass transfer resistance was completely negligible. Of course in the counter-extraction hydrophilic membranes would be more convenient than the

Extraction experiments with the whole bioconversion broth gave nearly the same results, showing that fouling did not represent a serious problem. The coupling of the membrane contactor with the bioreactor exhibited good results for the enzymatic conversion of vanillylamine to vanillin. Fig. 16 reports one test performed as follows: in the first period

considered, up to 90 minutes of experiment in this case.

S *(L/h)*

mass transfer rates were comparable in the two cases.

boundary layer played a minor role.

hydrophobic membranes used in this work.

W *(L/h)* 


Recovery of Biosynthetic Products Using Membrane Contactors 643

Lazarova, Z., Syska, B. & Schugerl, K. (2002). Application of large scale hollow fiber

Li, K. (2007). Ceramic membranes for separation and reaction. *John Wiley,* ISBN 978-470-

Lemanski, J. & Lipscomb, G.G. (1001). Effect of shell-side flows on the performance of hollow-fiber gas separation modules. Journal of Membrane Science 195, 215–228. Liang, T. & Long, R.L. (2005). Corrections to Correlations for Shell-Side Mass-Transfer

Lipnizki, F. & Field, R.W. (2001) Mass transfer performance for hollow fibre modules with

Lv, Y.; Yu, X.; Jia J.; Tu, S.T.; Yan, J. & Dahlquist, E. (2011). Fabrication and characterization

Miyatake, O. & Iwashita, H. (1990). Laminar-flow heat transfer to a fluid flowing axially

Nunge, R.J. & Gill W.N. (1966). An analytical study of laminar counter flow of double-pipe

Ramachandra Rao S. & Ravishankar G.A. (2000). Vanilla flavour: production by conventional and biotechnological routes. *J Sci Food Agric;* 80, 289-34 Reid R.C.; Prausnitz J.M. & Poling B.E. (1978) "*The properties of gas & liquids*", McGraw-Hill.

Seara, J.F.; Uhìa F.J.; Jaime Sieres, J.S. & Campo, A. (2007). A general review of the Wilson

Schlosser, S.; Kertész, R. & J. Martàk J. (2005). Recovery and separation of organic acids by

Sparrow, E.M.; Loeffler, A.L. & Hubbard, H.A. (1961). Heat transfer in longitudinal laminar flow between cylinders. *Trans ASME, Journal of. Heat Transfer* 83, 415 – 422. Trébounet, D.; Burgard, M. & Loureiro, J.M. (2006). Guidelines for the application of

fiber membrane contactor. *Separation and purification Technology* 50, 97-106. van den Heuvel R.H.H., Fraaije M.W., Laane C., van Berkel W.J.H., (2001) Enzymatic

Walton N.J., Narbad A, Faulds C.G. & Williamson G. (2000). Novel approaches to the

on recovery of MPCA. *Separation and Purification Technology* 41, 237–266. Siegel, R.; Sparrow, E.M. & Hallman T.M. (1958). Steady laminar heat transfer in circular tube with prescribed wall heat flux, *Applied Sci. Research* A7(5), 386 – 392. Sparrow, E.M. & Loeffler, A.L. (1959). Longitudinal laminar flow between cylinders

plot method and its modifications to determine convection coefficients in heat

membrane-based solvent extraction and pertraction. An overview with a case study

stationary model in the prediction of the overall mass transfer coefficient in hollow

absorption. *Appl Energy* doi:10.1016/j.apenergy.2010.12.038

exchange devices. *Applied Thermal Engineering* 27 2745–2757.

arranged in regular array. *AIChE Journal* 5, 325 – 330.

synthesis of vanillin, *J. Agric. Food Chem.* 49, 2954-2958.

biosynthesis of vanillin. *Curr. Opinion Biotechnol.* 11, 490-496

penicillin G. *Journal of Membrane Science* 202, 151 – 164.

*Journal of Membrane. Science* 193, 195–208.

heat exchanger, *AIChE Journal* 12, 279 – 289.

ISBN 0-07-100284-7, New York.

01440-0, Chichester.

*44,* 7835-7843.

(3) 417–425

membrane contactors for simultaneous extractive removal and stripping of

Coefficients in the Hollow-Fiber Membrane (HFM) Modules. *Ind. Eng. Chem. Res.*

shell-side axial feed flow: using an engineering approach to develop a framework,

of superhydrophobic polypropylene hollow fiber membranes for carbon dioxide

between cylinders with a uniform surface temperature, *Int. J. Heat Mass Transfer* 33


Boddeker, K.W; Gatfield, I.L.; Jahnig J. & Schorm C. (1997). Pervaporation at the vapor

D'Elia, N.A., Dahuron, L., Cussler, E.L., (1986). Liquid-Liquid extraction with microporous

Devis H.R. & Parkinson G.V. (1970). Mass transfer from small capillaries with wall resistance in laminar flow regime, *Applied Sci. Research* 22(1), 20 – 30. Ding,W.; He, L.; Zhao, G.; Zhang,H; Shu, Z.; Gao, D. (2004). Double porous media model for

Dongliang H.; Cuiqing, M.; Song ,L.; Lin, S.; Zang, Z.; Deng, Z. & Xu, P. (2007). Enhanced

Drioli, E.; Criscuoli, A. & Curcio E. (2006)*. Membrane contactors: fundamentals, applications and* 

Frank, G.T. & Sirkar, K.K. (1987). An integrated bioreactor-separator: in situ recovery of

Gostoli, C. & Gatta A. (1980). Massa transfer in a hollow fiber dialyser, *Journal of Membrane* 

Guzman, C. (2004). Vanilla, In *Handbook of herbs and spices, Vol. 2*, Peter K.V. (Ed.), pp. 322 –

Koonaphapdeelert, S.; Li, K. (2007). Preparation and characterization of hydrophobic ceramic hollow fibre membrane. *Journal of Membrane Science* 291 (1–2), 70–76. Koonaphapdeelert, S.; Zhentao Wu & Li, K. (2009). Carbon dioxide stripping in ceramic hollow fibre membrane contactors, *Chemical Engineering. Science*. 64, 1 - 8 Kosaraju, P.B. & Sirkar, K.K. (2007). Novel solvent-resistant hydrophilic hollow fiber

Koschilowski, J.; Wieghaus M. & Rommel M. (2003). Solar thermal driven desalination

Kreulen, H.; Smolders, G.F.; Versteeg, G.F. &van Swaaij W.P.M. (1993). Microporous hollow

plants based on membrane distillation. *Desalination* 156, 295-304.

membranes for efficient membrane solvent back extraction. *Journal of Membrane* 

fibre membrane modules as gas-liquid contactors. *Journal of Membrane Science* 78,

354, Woodhead Publishing, ISBN: 1-85573-721-3, Cambridge, UK. Happel, J. (1959). Viscous flow relative to arrays of cylinders. *AIChE Journal* 5, 174 – 177. Hatton A.P. & Quarmby (1962). Heat transfer in the thermal entry length with laminar flow in annulus, *International Journal of heat and mass transfer* 5(19), 973 -980. Knudsen, J.G. & Katz, D.L. ( 1958).*Fluid dynamics and heat transfer*, McGraw-Hill, ISBN, New

extraction technique. *Biotechnology and Bioengineering Symp*. 17, 303 - 316 Gaeta, S.N, 2003. Membrane contactors in industrial applications. Proc. of 1st Italy-Russia Workshop on membrane and membrane processes, Cetraro (I), 51-55 Gawronski, R. & Wrzesinska , B. (2000). Kinetics of solvent extraction in hollow-fiber

mass transfer of hemodialyzers. *International Journal of Heat and Mass Transfer* 47,

vanillin production from ferulic acid using adsorbent resin, *Appl. Microbiol* 

fermentation products by a novel membrane-based dispersion free solvent

pressure limit: Vanillin, *Journal of Membrane Science* 137, 155-158.

hollw fibres, *Journal of Membrane Science* 29, 309-319.

*potentialities*, Elsevier, ISBN 0-444 52203 4, Amsterdam

contactors. *Journal of Membrane Science* 168 , 213–222

4849–4855.

*Biotechnol* 74, 783-790

*Science* 6, 133 – 148.

*Science* 288, 41–50.

197-216.

York.


**28** 

*Brazil* 

Marco Aurelio Cremasco

*Chemical Engineering School/ University of Campinas* 

**Taxol® Separation in a Simulated Moving Bed** 

Cancer is a public health concern worldwide that must be considered in various area of the knowledge. In USA, appear a million cases each year. In the South East of the Brazil, cancer is the second largest cause of death. The introduction of chemotherapy to combat the cancer results in significant tumors cure that didn't control with success by exclusive use of surgery and/or radiotherapy (Bonadonna, 1990). Researchers around the world have particular attention in the study of natural products as possible source of antineoplasic agents. Due to the diversity of the chemical structures founded in theses products, there are big chances to identify news molecules with anti-tumor activities (Cremasco & Starquit, 2010). The Taxol® (commercial name for Paclitaxel) discovery offers good points for this reasoning (Holanda et

According to Rhoads (1995), while many drugs act to disrupt the cancer cells, Taxol® paralyzes the internal structure. In the metaphase stage of cell replication, chromosome pairs split and move to the opposite ends of the cell and wait to become part of a daughter cell. Rhoads (1995) reported that these chromosomes were guided by the microtubules made of tubulin. These bundles of microtubules must be dismantled before the cell be divided. However, Taxol® prevents the cells from dividing further and essentially halts the cancer growth. It has been approved by the FDA, in the USA, for the treatment of advanced breast cancer, lung cancer, and refractory ovarian cancer. Taxol® is a diterpene compound (Figure

**1. Introduction** 

al., 2008).

1).

Fig. 1. Chemical structure of Taxol®

