1-s2.0-S0255270110002059-main

8/13/2019 1-s2.0-S0255270110002059-main

1/6

Chemical Engineering and Processing 49 (2010) 1101110 6

Contents lists available at ScienceDirect

Chemical Engineering and Processing:Process Intensication

j o u rn a l h o m ep a g e : www.e l sev i e r. co m/ l o ca t e / cep

Enzymatic hydrolysis of canola oil with hydrodynamic cavitation

Romain Sainte Beuve a , Ken R. Morison b,a ENSAIA, Vandoeuvre Les Nancy, Franceb Chemical and Process Engineering, University of Canterbury, Private Bag 4800, Christchurch 8140, New Zealand

a r t i c l e i n f o

Article history:Received 24 April 2009Received in revised form 12 August 2010Accepted 12 August 2010Available online 20 August 2010

Keywords:Canola oilEnzyme hydrolysisCandida rugosaHydrodynamic cavitation

a b s t r a c t

A hydrodynamic cavitation system based on a venturi was used to test the effectiveness of cavitationfor enhancing the enzymatic hydrolysis of canola oil using lipase from Candida rugosa . Cavitation ledto the production of ne oil-in-water and water-in-oil emulsions with the enzyme in the water phase.Using venturi inlet pressures of up to 8bar, the yield of fatty acids was only about 60% of the maximumpossible. In contrast, a simple stirred batch reactor produced over 90% of the maximum possible yieldwith reaction rates equal to, or better than, those obtained in a cavitating system. It was concluded thatcavitation inhibited the reaction in some way and is not effective for intensication of hydrolysis.

2010 Elsevier B.V. All rights reserved.

1. Introduction

Edible oil hydrolysis (lipolysis) is a primary unit process whichis used by industries such as dairy,detergent, cosmetic, oleochemi-

cal,petroleum,andwaste processing.Sustainable waysof obtainingfree fatty acids and glycerol from triglycerides have been inves-tigated by many researchers [13] . Traditional processes requireoperating conditions with high pressures and temperatures, oralkaline/acid conditions, and have highenergy consumption. How-ever they have short reaction times and get yields close to 100%.The most common of methods has been the Colgate Emery processwhich needs operating temperatures of 250 C and up to 60bar of pressure [4,5] .

Alternative systems using enzymes can eliminate some of thedisadvantages of these methods. Use of an aqueous enzyme as acatalyst speeds the reaction up. This intensication allows it to becarried out at milder conditions and to get more specic productswith an equivalent yield [6] . However, because of the immiscibility

of the oil and aqueous enzyme phases, the reactions to producefree fatty acids, glycerol andothers hydrolysis productsoccur at theinterface of these twophases. Therefore, increases in the interfacialarea between the phases can enhance the extent of reaction [4,7] .

The kinetics of oil hydrolysis catalysed by enzymes has beenwidely studied for several years [3,4,710] , and the main parame-tersaffecting thisreaction have beenidentied.Al-Zuhair et al. [7,9]showed that agitation, oil fraction, and temperature inuenced

Corresponding author. Tel.: +64 3 364 2578; fax: +64 3 364 2063.E-mail address: [email protected] (K.R. Morison).

the surface area between palm oil and water in a stirred reactor.Theystated that, due to the adsorptiondesorption dynamicsof theenzyme at the interface, the rate of the reaction did not increaselinearly with enzyme concentration and reaches a threshold at

a critical enzyme concentration. Any further increase in enzymeconcentration did not change the rate signicantly.Recently, numerous researchers have investigated process

intensication of different applications, in particular with cavita-tion. Hydrodynamic cavitation has been proposed as a good wayto intensify chemical processes, mass transfer and biological dis-ruption [8,11] . Cavitation is dened as the generation, growth andthe subsequent collapse of vapour bubbles in a liquid. The collapseof these bubbles creates high energy release, with high local tem-peratures and pressures, at a large number of reaction sites eventhough theoverall reactionis carried out at ambient conditions [8] .In addition, the dissociation of water molecules within the bub-bles can lead to the generation of free radicals. These substanceshave been shown to be very effective for intensication for some

reactions such as the oxidation of potassium iodide [12] .Cavitation can occur when high liquid velocities cause the localpressure to drop below the vapour pressure of the liquid. The pres-sure, P , inside a device such as a venturi can be estimated from therelevant parts of Bernoullis equation:

P = P 0 1 A20 / A

2v

v 2v

2 (1)

where P 0 is the upstream pressure, v v is the uid velocity in theventuri, is the uid density and A0 and Av are the cross-sectionalareas of the upstream pipe and venturi respectively. If the valueobtainedis sufciently lower thanthe vapour pressure of the liquid,

0255-2701/$ see front matter 2010 Elsevier B.V. All rights reserved.

doi: 10.1016/j.cep.2010.08.012
http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.cep.2010.08.012http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.cep.2010.08.012http://www.sciencedirect.com/science/journal/02552701http://www.elsevier.com/locate/cepmailto:[email protected]://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.cep.2010.08.012http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.cep.2010.08.012mailto:[email protected]://www.elsevier.com/locate/cephttp://www.sciencedirect.com/science/journal/02552701http://localhost/var/www/apps/conversion/tmp/scratch_4/dx.doi.org/10.1016/j.cep.2010.08.012

8/13/2019 1-s2.0-S0255270110002059-main

2/6

1102 R. Sainte Beuve, K.R. Morison / Chemical Engineering and Processing 49 (2010) 11011106

bubble formation is likely within the region of high velocity. Oncevapourisation occurs, predictions of P from Eq. (1) using the liquiddensity are no longer correct.Any downstream increase in pressureabove the liquid vapour pressure leads to collapse of the bubbles.

For heterogeneous liquid/liquid reactions, the collapse of bub-bles can take place near the interface between the two liquids. Itresults in mixing and disruption of the solution which scatters onephase inside the other. Thus an emulsion is generated raising thesurface area between the phases, and causing faster mass transfer[4,7] . Recently, Patist and Bates [11] showed that ultrasonic pro-cesses canproducea very ne andstable emulsion with lowenergyinput, increasing the surface area between two immiscible liquids.

Normally enhanced reaction rates in cavitation systems areassociated with free radicals, locally high temperatures and pres-sures [13] , enhanced mass transfer rates, or enhanced interfacialareas. Pandit and Joshi [1] compared acoustic and hydrodynamiccavitation systems for the enhancement of oil hydrolysis. For thehydrodynamic system, they used 200 L of oilwater mixture (with110% oil) which was passed through a throttled cavitating valvewithout any enzyme, at 34bar for about 40 h. They showed thathydrolysis can occur using hydrodynamic cavitation or ultrason-ics (at 20 kHz for 10 h) at room temperature, and concluded thatcavitation provided localisedhightemperatures and pressures thatenabled the reaction to occur. This work has been cited in over 25papers since butno other reports of hydrolysis with hydrodynamiccavitation have been found. However, ultrasonic cavitation hasbeen used to enhance hydrolysis. Talukder et al. [14] f ound an opti-mal power for ultrasonic enhancement of hydrolysis of olive oil byChromobacterium viscosum lipase in a two-phase water/isooctanesystem. They stated that this was associated with increased inter-facial area, but they also noted a more rapid loss of lipase activitywith sonication. Lee et al. [15] used ultrasound to enhance lipaseactivity in ionic liquids and concluded that it increased the masstransfer rates without causing loss of enzyme stability. Yachmenevetal. [16] concluded thatit enhancedenzymetransportand openedup the surface of the solid substrate. No research was found thatcombined cavitation with enzyme hydrolysis.

In another enhanced hydrolysis process, Weatherley andRooney [3] f ound that an electrostatic system used with enzymescould be as efcient as steam splitting performed at 240 C and33 bar. Giorno et al. [17] used microltration membranes to createemulsions for the study of the distribution of the Candida rugosaenzyme on the oil/water interface.

The aim of this study was to determine the effect of cavita-tion on the rate and yield of enzymatic hydrolysis of vegetable oil.The effect of pressure, enzyme concentration, emulsion type, andprocess operation were considered to enable a comparison withhydrolysis in a stirred reactor.

2. Materials and methods

Supermarket house brand canola oil (Pams Products Ltd., NewZealand) was used for all experiments carried out in this study.All the oil used had the same manufacture identication andwas assumed to be from the same batch. Typical canola oil con-tains triglycerides with a typical molecular mass of 882.1 g/mol(94.498.1%), phospholipids(up to 3.2%), freefatty acids (0.41.2%),an unsaponiable oil part (0.51.2%) and other compounds (toco-pherols, chlorophylls, sulphur, iron) [18] . The proportions of different types of lipid chains are: monounsaturated chains 67.6%(mostly oleic acid), polyunsaturated 27.2%, saturated chains 7.4%,others (trans saturated) less than 1%.

The lipase enzyme used was from C. rugosa (Sigma Chemical, Japan), with a specied concentration of 901 units per microgram

of solid. It presents neither regioselectivity nor specicity for type

Fig. 1. Hydrodynamic cavitation apparatus.

of chain. However, it shows a level of discrimination against longerchains offattyacids(C18C22,mainlyomega 3 fat) and a slow selec-tivity for unsaturated acids [6] . It was added to reverse osmosiswater at a concentration of 1g/L (unless otherwise stated) to form

the aqueous phase. Optimal conditions of its activity are a temper-ature range of 3339 C,andpH of7 but the enzymeis activein thepH range 5.59 [19] . All enzyme concentrations reported here arewithin the aqueous phase only.

A cavitation loop ( Fig. 1) consisted of a 500mL stainless steelbeaker, a variable-speed magnetically coupled gear pump (GD-M35, Micropump Inc., WA, USA) and a venturi device ( Fig. 2). Thesame device had been used previously to study the Weissler reac-tion [12] . For each run using the cavitation, 500mL of the aqueousphase was added to the empty beaker and apparatus. Initially theaqueous phase was circulated at less than 1 bar gauge pressure(to prevent cavitation) while increasing the temperature to 36 C.Indeed, it hadpreviously been found [12] that cavitation of a diluteKI solution in the same system occurred from 1.8bar gauge. Then100mL of the oilwas added atabout1 mL/s tothe loop immediatelyupstream of the gear pump by a variable-speed peristaltic pump(Masterex Console Drive). Once the oil had been added, the gearpump speed was quickly increased to obtain the desired pressure(0.68bar, upstream of the venturi) and the timer started. Duringthe runs the solution temperature was controlled to within 1 Cusing a stainless steel cooling coil in the beaker through whichwater was circulated from controlled water bath with a temper-ature stability of 0.1 C. The water bath temperature set pointwas adjusted depending on the pressure, and hence power input,from the pump during a run. The pH of the solution in the beakerwas measured continuously throughout all the runs but it was notcontrolled.

To compare the efciency of the cavitation system, a 470mL stirred batch reactor was set up using the same beaker and coolingcoil, with a 36-mm axial impeller with 4 blades, and inserted athalf the liquid depth. A volume of 392 mL of aqueous solution waspouredintothe beaker, the temperature wasbrought to thedesiredvalue, and 78mL ofoil was added directlyto the beaker. Atthe starttime the speed rate of impeller was quickly raised to 1850rpm for

Fig. 2. Venturi device.

8/13/2019 1-s2.0-S0255270110002059-main

3/6

R. Sainte Beuve, K.R. Morison / Chemical Engineering and Processing 49 (2010) 11011106 1103

Fig. 3. Small oil droplets and surrounding aqueous phase (water and enzyme) generated (a) by cavitation (35 min) and (b) in the stirred reactor (30min). Both images areabout 2mm wide.

oil-in-water emulsions (or 1600rpm for water-in-oil emulsions).At these conditions, the impeller tip speed was too low to causecavitation. The temperature was controlled and the pH measuredin the same manner as for the cavitation run.

To determinethe emulsionsize,a samplewas removed from thebeaker and a droplet was immediately placed between two micro-scopeslides. Theemulsion was viewedat 5 magnicationusinganOlympus BX-60F3 microscope (Olympus Optical Ltd, Japan) ttedwith a camera (ProgresCFscan, Jenoptik, Germany).

At different intervals, a sample (about 3 mL) of solution wasremoved from the beaker using a syringe. It was poured into aconical ask containing 30 mL each of acetone and ethanol [6].This solvent broke the emulsion and the mixture separated intoan aqueous phase containing the enzyme and free fatty acids, anda solvent phase containing the remaining oil. The enzyme reactionwas thus stopped and titration of the aqueous phase was enabled.

Then sample was titrated using 0.05M NaOH solution until neu-tral. Previously, phenolphthalein droplets had been added to thissolvent mixture after which it was neutralized before adding thesample. The same solvent mixture was kept for several titrationsand changed after 4 or 5 successive samples.

The extent of hydrolysis, X , was calculated from the titrationvolume. It was dened as the quantity of free fatty acid moleculesin a sample, divided by 3 times the initial quantity of oil molecules.

Fig. 4. Effect of pressure upon extent of reaction: curves represent the best t

exponential model (3.4bar dotted, 6bar dashed, 8bar solid).

It was calculated by:

X =V

NaOHC

NaOHMW

oil3woil m s (2)

where V NaOH is the volume of NaOH solution required to neutral-ize the free fatty acid, C NaOH is the molar concentration of solution,MW oil is the molar mass of oil (triglycerides), woil is the mass frac-tion of oil in the initial solution, and m s is the mass of the sample.Theuncertaintyfrom theanalysisin thedeterminationof theextentof hydrolysis was estimated from titration uncertainties to be 2%hydrolysis.

The experimental data were tted by non-linear least squaresusing Microsoft Excel and Solver to an exponential model (Eq. (3) ):

X = X max (1 e kt ) + X 0 (3)

where X max is maximum extent of hydrolysis for each run, k is the

rate constant and X 0 was the initial extent of hydrolysis of the oilmeasured by the same titration method.

3. Results and discussion

The pressured used, ow rates obtained and calculated valuesof P (using Eq. (1) ) are summarised in Table 1 . During the experi-ments the temperature of the mixture was maintained in therange3537 C. The pH dropped, without adjustment, from the initialpH of 7.0 to a typical value of 5.2 during a run; the minimumpH recorded was 4.8. The hydrodynamic cavitation system wasvery effective at producing a stable emulsion. Indeed some of theemulsion samples that were left to sit remained suspended after anumber of weeks. The microscopic examination of the oil-in-water

emulsion conrmed the presence of numerous small oil droplets inthe aqueous phase, proving the efciency of both the venturi sys-tem and stirred system in creating a ne emulsion with a resultantlarge surface area ( Fig. 3).

The hydrolysis results showeda typical rst order response, butcomplete hydrolysis was never achieved ( Fig. 4). The cavitation

Table 1Experimental conditions and calculated pressure for the cavitation system.

Pressure (bar) Flow rate (mL/s) Velocity (m/s) P (bar abs.)

0.6 12.9 0.1 19 0.22.0 21.9 0.2 32.2 2.13.4 26.6 0.9 39.1 3.16.0 34.1 1.0 50.2 5.47.4 37.7 0.9 55.6 6.8

8/13/2019 1-s2.0-S0255270110002059-main

4/6


Fig. 5. Efciency of different systems upon oil hydrolysis. Stirred reactor value isnoted on the y-axis at zero pressure.

timer was started a few minutes after the oil and aqueous phaseswere rst in contact, so some hydrolysis reaction had already

occurred and free fatty acids were detected in the mixture at thestart of cavitation or rapid stirring.Uncertainties in the calculated extent of hydrolysis were esti-

mated from three runs that were carried out at 6 bar and 1 g/L enzyme. The uncertainty was estimated from the standard devi-ation which was almost the same as half the range in values.Uncertainties ranged from 3% hydrolysis initially to 6% hydrol-ysis. These same uncertainties were found to adequately describethe discrepancy between data points and an exponential model inalmost all cases. For clarity typical uncertainties are shown in Fig. 4only.

3.1. Effect of pressure

The hydrodynamic cavitation experiments were carried outat several inlet pressures from 0.6 to 8 bar. The pressure waslimited by the gear pump to 9 bar. All the ow rates tested(from 0.6 to 8 bar) gave rise to calculated pressures ( P ) thatwere well below the vapour pressure of water at 36 C (Table 1 ),but there was only audible evidence of cavitation at 2bar andhigher.

The results did not show any clear effect of inlet pressureon the initial stages of hydrolysis, and for pressures from 3.4 to8 bar there seemed to be no pressure effect at all ( Fig. 5). Belowthis pressure the maximum extent of hydrolysis was dependenton pressure. The model curves (Eq. (3)) tted to experimentalpoints conrmed that the trends were not distinguishable forpressures higher than 3 bar. It is tempting to claim that some

of the effects seen were due to cavitation. However turbulencecauses mixing of the two phases and this might be solely respon-sible for the increase in hydrolysis observed. Once turbulentmixing was adequate (at 3.4 bar) no further improvement wasobtained.

The initial rate reaction obtained from the curve tting showedno clear trend (not shown). Over a range in pressures, the initialconversion rate was between 0.011 and 0.017 min 1 with much of this variation due to experimental variability.

3.2. Stirred batch reactor

To evaluate the effect of cavitation upon lipolysis, a compar-ison was made with a stirred batch reactor. It was performed

with the same conditions as for cavitation. The reaction curve

Fig.6. Testsof inhibition ofhydrolysis:withcavitationat 6.0barup to150 minthenthe stirred batch reactor; 1 g/L enzyme up to 165 min then 2 g/L at 6 bar; 0.6bar upto 120min then 6.0bar; no enzyme at 6bar.

again showed a typical rst order response but a maximum extentof hydrolysis of nearly 90% was achieved ( Fig. 4). The remain-

ing 10% could be explained by the presence of unsaponiablematter (0.51.2% of crude oil [17] ) and acylglycerin compounds(mono- and di-acylglycerin) which did not react with NaOH.No further analysis was carried out to check this possibil-ity.

The yield of nal titrated products in the stirred reactorwas greater than those obtained using the hydrodynamic cav-itation process. The tip speed of the stirrer was too low tocause cavitation so these greater yields further support thecase for the effect of turbulence. There was no evidence thatcavitation in the venturi enhanced the reaction and indeedthe poor performance, as compared with the stirred reactor,offers evidence of some form of inhibition or back reac-tion.

The overall efciency of both systems was compared by cal-culating the energy specic conversion. Fig. 5 conrms that thestirred system was more efcient than the cavitation system. Also,it shows that high pressures needed relatively more energy for thereaction.

Emulsion size was assessed to conrm previous results. Asample picture was taken from each process ( Fig. 3). The num-ber of droplets became higher and their diameter smaller withtime spent under cavitation (not shown). A ner emulsion wascreated in the cavitation system, which was likely to enhancethe interfacial area available for the hydrolysis reaction. Simi-lar development of droplets was found in the stirred reactor,but a ne emulsion took longer to form than when gener-ated with the cavitation system. Thus it seems very unlikelythat the difference in conversion was related to interfacialarea.

The initial rate of conversion for stirred reactor were calculatedand compared to the values from the cavitation system. The values(0.015 and 0.016 min 1 ) are within the same range and indicatethat the interfacial area was sufcient in all cases and for bothsystems.

3.3. Enzyme concentration and inactivation

The relatively low maximum extent of reaction observed ledto concerns of possible enzyme inactivation by cavitation. Theeffect of this and enzyme concentration was checked by usingno enzyme, 1 g/L, and also by adding an extra 1 g/L after 165 min

of reaction. The results in Fig. 6 clearly show that no hydrolysis

8/13/2019 1-s2.0-S0255270110002059-main

5/6

R. Sainte Beuve, K.R. Morison / Chemical Engineering and Processing 49 (2010) 11011106 1105

occurred without enzyme but that extra enzyme did not have asignicant effect, indicating that inactivation was unlikely to be aproblem. If the initial enzyme had been inactivated over the timeof cavitation, the extra enzyme would have produced a greaterresponse.

Some previous researchers, working on stirred processesdemonstrated that, for higher enzyme concentrations, hydrolysiswas more efcient. Rooney and Weatherley [4] used water-in-oil emulsion and found after 100min 88% of hydrolysis isachieved with an enzyme concentration of 0.4% of total masstotal, against 50% with enzyme concentration of 0.1%. In con-trast, Al-Zuhair et al. [7] noted a decrease and stabilizationof initial reaction rates for higher enzyme concentrations inan oil-in-water mixture. This was thought to be mostly dueto saturation of enzyme molecules at the oilwater inter-face.

A run to check enzyme inactivation was carried out with twodifferent pressures: a rst period with 0.6 bar (no cavitation), thena second period after 120 min at 6 bar. The results ( Fig. 6) weresimilar to those shown in Fig. 4. At 0.6 bar, the extent of hydroly-sis reached a steady value of around 25% after about 30 min. Withthe increase in pressure up to 6 bar, the extent of hydrolysis roseto values found in earlier experiments ( 60%). These results showthat the low extent of hydrolysis at 0.6 bar was not due to enzymeinactivation.

Ina subsequent trial,6 barcavitationwas followed bythe stirredbatch reactionafter 150min. During therst period, shown in Fig.6,hydrolysis proceeded up to 60% as previously found ( Fig. 4) andthen, in the stirred reaction, continued up to 8090%. Thus therewas no evidence of enzyme inactivation. Further, given that therewas no adjustment of pH andthat the stirred reaction continued tohydrolyse the triglyceride, it is unlikely that low pH was limitingthe reaction.

Lee and Choo [20] investigated that effect of shear stress oninactivation of lipase from Candida cylindracea (now known asC. rugosa ) and found that its activity was affected by shear ratebut not shear stress. They concluded that shear induced airwater

interfacial effects were likely to be the cause of inactivation. Insome experiments Lee and Choo found recovery of enzyme activ-ity while being sheared, but they did not determine the reason forthis. Mohanty et al. [21] also investigated the inactivation of lipase(from Aspergillus oryzae ) by rapid stirring in a bafedvessel. At highspeed (4000 rpm), inactivation to about 20% of the initial activityoccurred in the rst 15 min followed by recovery to 60% over 4 h.Theauthorsfoundthata gasliquid interface was required forinac-tivationand thatloss of tryptophan uorescence intensity was veryclosely related to loss of activity. They did not offer an explanationfor the recovery of activity.

In the cavitation device shear rates due to liquid ow were esti-matedtobeuptoabout5 10 5 s 1 (much higherthan150 s 1 usedby Lee and Choo [20] ) with shear stresses up to about 500Pa (at

least 3 times higher than Lee and Choo). In addition there are sig-nicant stresses caused by bubble collapse. Given the results inFig. 6, there is no evidence from the cavitation results to indicatepermanent enzyme inactivation by shear. However, the possibilityexists of recovery(over minutes)of lipase activityafter inactivationin a high shear environment.

In a reverse micelle system C. rugosa lipase was found to beinhibited by the oleic acid product of olive oil hydrolysis [22] . Theresults in Fig. 6 indicate that this is not likely in the current work.

Theresultsdiscussedsofarinthissectioncouldalsobetheresultof a competing product reaction. Ferreira-Dias and Fonseca [23]used the immobilised C. rugosa lipase with olive oil, hexane, andexcess glycerol to produce monoglycerides. They found thatmono-glycerides were formedas well as fatty acids even when there were

small amounts of water present. There is a chance that hydrody-

Fig. 7. Comparisoncavitationprocess (6.0bar) and stirred process (1600rpm) withwater-in-oil emulsion (ratio 2:1, oil:water v/v) compared with oil-in-water (1:5)emulsion. Curves represent the closest exponential model for both runs (Eq. (3) ).

namic cavitation enables a reaction to produce monoglycerides.Clearly further experimentation is required to determine the

mechanisms for low conversions during hydrodynamic cavitationof this system.

3.4. Effect of emulsion type

Most of the past research on hydrolysis has used water-in-oilemulsions. Hence, some experiments were carried out to test forany effect from the emulsion type. Both the cavitation and stirredsystems were tested with 2:1 ratio (oil:water) and with the sameenzyme concentration in the aqueous phase.

Results in Fig. 7 show similar evolution, particularly for the rst120 min for the two different reactor types. It was very difcult tomeasure the pH in the emulsion as the continuous phase was oil.Further the creamy emulsion generated after only 2 h become tooviscous to enable high velocities. After 1 h and 45 min and with apressure of 6 bar the typical cavitation sound was no longer heard.

Cavitation might nothaveoccurred after this time. These runs werenot repeated so the steady state extent of reaction was not deter-mined.

A comparison between the different oil:water ratios in Fig. 7shows also that there is a clear difference of yield between twotypes of emulsion. With 2.5 times less enzyme (i.e., water) and 4times more oil (in mass), the proportion of hydrolysis achieved inthe water-in-oil emulsion was more than 2 times less than oil-in-water emulsion at any given time. Initial conversion rates are notthe same showing a slow reaction when the water-in-oil emul-sion is used. It could be accounted for by restricted dispersion of enzyme at the oilwater interfacial area. Enzymes are trapped insmall water droplets (because water is the dispersed phase), whenagitation is limited [7].

The efciency of water-in-oil system was quite similar to theoil-in-water emulsion. A quantity of 2 10 6 mol/J was producedin the cavitation system at 6bar and 2.8 10 5 mol/J in the stirredsystem. This compared with about 1 10 6 and 3 10 5 mol/Jrespectively for the oil-in-water emulsion ( Fig. 5), i.e., there wasno signicant inuence of emulsion type upon oil hydrolysis ef-ciency in the stirred systemand a slight difference in the cavitationsystem.

4. Conclusion

Hydrodynamic cavitation was found to be very effective inproducing both oil-in-water and water-in-oil emulsions. Howeverhydrodynamiccavitation was not effective in enhancing enzymatic

hydrolysis of oils, and lower yields were obtained. A stirred batch

8/13/2019 1-s2.0-S0255270110002059-main

6/6


system produced superior yields and energy efciency. Permanentinactivation of the enzyme, pH effects and product inhibition wererejected as causes of thepoorer performance of hydrodynamic cav-itation. The cause of the difference was not determined.

References

[1] A.B. Pandit, J.B. Joshi, Hydrolysis of fatty oils: effect of cavitation, Chem. Eng.Sci. 48 (1993) 34403442.

[2] Q. Gan, H. Rahmat, L.R. Weatherley, Simultaneous reaction and separation inenzymatic hydrolysis of high oleate sunowerevaluation of ultraltrationperformance and process synergy, Chem. Eng. J. 71 (1998) 8796.

[3] L.R. Weatherley, D. Rooney, Enzymatic catalysis and electrostatic processintensication for processing of natural oils, Chem. Eng. J. 135 (2008)2532.

[4] D. Rooney, L.R. Weatherley, Theeffect of reaction conditions upon lipasecatal-ysed hydrolysis of high oleate sunower oil in a stirred liquidliquid reactor,Process. Biochem. 36 (2001) 947953.

[5] J.A.Kent (Ed.), RiegelsHandbookof IndustrialChemistry,9th edition,Chapmanand Hall, New York, 1992.

[6] D. Agrawal, Enhanced enzymatic hydrolysis of oils and fats. M.E. Thesis, Uni-versity of Canterbury, New Zealand, 2004.

[7] S.Al-Zuhair, K.B.Ramachandran, M. Hasan, Kineticsof theenzymatic hydrolysisof palm oil by lipase, Process. Biochem. 38 (2002) 11551163.

[8] P.Gogate,Cavitational reactors for process intensicationof chemical process-ing applications: a critical review, Chem. Eng. Process. 47 (2008) 427515.

[9] S. Al-Zuhair, K.B. Ramachandran, M. Hasan, Effect of enzyme molecules cover-ing of oilwater interfacial on the kinetic of oil hydrolysis, Chem. Eng. J. 139(2008) 540548.

[10] C. Albasi, N. Bertrand, J.-P. Riba, Enzymatic hydrolysis of sunower oil in astandardized agitated tank reactor, Bioprocess. Eng. 20 (1999) 7781.

[11] A. Patist, D. Bates, Ultrasonic innovations in the food industry: from the labo-ratory to commercial production, Innov. Food Sci. Emerg. 9 (2008) 147154.

[12] K.R. Morison, C.A. Hutchinson, Limitations of the Weissler reaction as a modelreaction for measuring the efciency of hydrodynamic cavitation, Ultrason.Sonochem. 16 (2009) 176183.

[13] S. Arrojo, C. Nern, Y. Benito, Application of salicylic acid dosimetry to eval-uate hydrodynamic cavitation as an advanced oxidation process, Ultrason.Sonochem. 14 (2007) 343349.

[14] M.M.R.Talukder,M.M. Zaman, Y.Hayashi,J.C. Wu,T. Kawanishi,Ultrasonicationenhanced hydrolytic activity of lipase in water/isooctane two-phase systems,Biocatal. Biotransfor. 24 (2006) 189194.

[15] S.H. Lee, H.M. Nguyen, Y.-M. Koo, S.H. Ha, Ultrasound-enhanced lipase activityin the synthesis of sugar ester using ionic liquids, Process. Biochem. 43 (2008)

10091012.[16] V. Yachmenev, B.D. Condon, K.T. Klasson, A.H. Lambert, Acceleration of theenzymatic hydrolysis of corn stover and sugar cane bagasse celluloses by lowintensity uniform ultrasound, J. Biobased Mater. Bio. 3 (2009) 2531.

[17] L. Giorno, E.Piacentini,R. Mazzei, E.Drioli,Membraneemulsicationas a novelmethod to distribute phase-transfer biocatalysts at the oil/water interface inbioorganic reactions, J. Membr. Sci. 317 (2008) 1925.

[18] Edibleoil andfat products,in: F. Shahidi (Ed.),in: BaileysIndustrial Oiland FatProducts, vols. 36, 6th edition, John Wiley & Sons, Hoboken, NJ, 2005.

[19] E.B. Pereira, H.F. de Castro, F.F. de Moraes, G.M. Zanin, Kinetic studies of lipase from candida rugosa : a comparative study between free and immobi-lized enzyme onto porous chitosan beads, Appl. Biochem. Biotechnol. 9193(2001) 739752.

[20] Y.-K.Lee, C.-L.Choo, Thekineticsand mechanismof shear inactivation on lipaseCandida cylindracea , Biotech. Bioeng. 33 (1989) 183190.

[21] M. Mohanty,R.S. Ghadge, N.S.Patil, S.B.Sawant,J.B. Joshi, A.V.Deshpande, Inac-tivation of lipase at gasliquid interface in stirred vessel, Chem. Eng. Sci. 56(2001) 34013408.

[22] S.W. Tsai, C.L. Chiang, Kinetics, mechanism and time course analysis of lipase-catalyzed hydrolysis of high concentration olive oil in AOT-isooctane reversemicelles, Biotechnol. Bioeng. 38 (1991) 206211.

[23] S.Ferreira-Dias,M.M.R.da Fonseca,Productionof monoglyceridesby glyceroly-sisof oliveoil withimmobilized lipases:effect of thewater activity, Bioprocess.Eng. 12 (327) (1995) 33.

1-s2.0-S0255270110002059-main

Documents

Transcript of 1-s2.0-S0255270110002059-main