Lipase-catalyzed process in an anhydrous medium with enzyme reutilization to produce biodiesel with low acid value

Journal of Bioscience and BioengineeringVOL. 112 No. 6, 583–589, 2011

www.elsevier.com/locate/jbiosc

Lipase-catalyzed process in an anhydrous medium with enzyme reutilization toproduce biodiesel with low acid value

Laura Azócar,1,⁎ Gustavo Ciudad,1 Hermann J. Heipieper,2 Robinson Muñoz,1 and Rodrigo Navia 1,3

⁎ CorrespondE-mail add

[email protected]

1389-1723/$doi:10.1016/j

Scientific and Technological Bioresource Nucleus, Universidad de La Frontera, Casilla 54-D, Temuco, Chile,1 Department of Environmental Biotechnology,Helmholtz Centre for Environmental Research-UFZ, Permoserstr. 15, 04318 Leipzig, Germany,2 and Departamento de Ingeniería Química, Universidad de La

Frontera, Casilla 54-D, Temuco, Chile3

Received 21 February 2011; accepted 1 August 2011Available online 1 September 2011

One major problem in the lipase-catalyzed production of biodiesel or fatty acid methyl esters (FAME) is the high acidity ofthe product, mainly caused by water presence, which produces parallel hydrolysis and esterification reactions instead oftransesterification to FAME. Therefore, the use of reaction medium in absence of water (anhydrous medium) was investigatedin a lipase-catalyzed process to improve FAME yield and final product quality. FAME production catalyzed by Novozym 435was carried out using waste frying oil (WFO) as raw material, methanol as acyl acceptor, and 3 Å molecular sieves to extractthe water. The anhydrous conditions allowed the esterification of free fatty acids (FFA) from feedstock at the initial reactiontime. However, after the initial esterification process, water absence avoided the consecutives reactions of hydrolysis andesterification, producing FAME mainly by transesterification. Using this anhydrous medium, a decreasing in both the acidvalue and the diglycerides content in the product were observed, simultaneously improving FAME yield. Enzyme reuse in theanhydrous medium was also studied. The use of the moderate polar solvent tert-butanol as a co-solvent led to a stablecatalysis using Novozym 435 even after 17 successive cycles of FAME production under anhydrous conditions. These resultsindicate that a lipase-catalyzed process in an anhydrous medium coupled with enzyme reuse would be suitable for biodieselproduction, promoting the use of oils of different origin as raw materials.

© 2011, The Society for Biotechnology, Japan. All rights reserved.

[Key words: Biodiesel; Transesterification; Waste frying oils; Molecular sieves; Novozym 435; Lipase reuse; tert-Butanol]

Lipase-catalyzed processes have been widely investigated inbiodiesel or fatty acid methyl ester (FAME) production (1–5), but amajor problem is the high acidity of the final product, mainly causedby water presence. In fact, water favors the hydrolysis reactionproducing FAME with a high acid value. Therefore, to accomplishwith international biofuel standards, additional treatments areneeded (6). In addition, this problem could be enhanced by usingalternative feedstock characterized by high acid value, such as wastefrying oils (WFO), as well as jatropha, microalgae and crude palm oil(7).

To solve these drawbacks the addition of water adsorbents to thelipase-catalyzed reaction has been recently investigated (2,8). Somereported absorbents are blue silica gel, maceo-pored silica gel, fine-pored silica gel, and molecular sieves of 3 Å, 4 Å and 5 Å (2,8). Amongthe adsorbents examined, blue silica gel has been reported as the bestone, increasing FAME yield when immobilized lipase from Penicilliumexpansum was used as catalyst (2). However, high dosage of silica gelcould provoke low biodiesel yield. In fact, as the pore size of silica gel

ing author. Tel.: +56 45 592128; fax: +56 45 325053.resses: [email protected] (L. Azócar), [email protected] (G. Ciudad),[email protected] (H.J. Heipieper), [email protected] (R. Muñoz),(R. Navia).

- see front matter © 2011, The Society for Biotechnology, Japan. All.jbiosc.2011.08.001

is much larger than the methanol molecule size, silica adsorbsmethanol negatively affecting the reaction performance (8).

The possible advantages of using water adsorbents in lipase-catalyzed processes to improve FAME production yield may be incontradiction with reports showing that lipases activation may needwater presence. In fact, lipases activation involves the active siterestructuration, which requires the presence of an oil–water interface(9). As a result, small dosages of water in the reaction have beenproposed to achieve an effective process (9). Candida antarctica lipaseB immobilized on acrylic resin (Novozym 435) has been shown tocontain enough water in its support medium to preserve its catalyticactivity in an anhydrous medium (4,10). In a response surfacemethodology study, Organovic et al. (4) correlated Novozym 435concentration with water concentration, achieving the highestbiodiesel yield of 92% using the medium with the lowest watercontent level (0%) and the highest enzyme concentration (5% basedon oil weight). In another study, Novozym 435 showed low activity inthe presence of water and therefore an anhydrous medium wasproposed for efficient biodiesel production (10).

Novozym 435 has been widely reported as an effective biocatalystfor promoting high FAME production yields (7). Optimal conditions toreach 100% FAME yield using WFO as feedstock were determined tobe the following: Methanol-to-oil molar ratio 3.8:1 (wt), 15% (wt)Novozym 435 and incubation at 44.5°C for 12 h with agitation at

rights reserved.

584 AZÓCAR ET AL. J. BIOSCI. BIOENG.,

200 rpm (11). Novozym 435 activity loss and regeneration have beenalso investigated. It has been established that the immobilizationmaterial (acrylic resin) could adsorb polar components such asmethanol and glycerol, provoking enzyme inactivation. Therefore,Novozym 435 reutilization by means of washing processes usingacetone, soybean oil, tert-butanol, isopropanol and 2-butanol hasbeen investigated (12,13). In addition, the incorporation of a co-solvent in the reaction has been recently proposed (5,9).

According to previous research, Novozym 435 has been proved to bean effective catalyst in FAMEproduction and has been shown to preserveits catalytic activity in an anhydrous medium. However, Novozym 435has not been investigated yet regarding catalyzed-processes whenadsorbent materials are added to the reaction in order to obtain highquality biodiesel with low acid value. In this sense, it is necessary toevaluate the behavior of the enzymatic reaction in anhydrous medium.In addition, it is necessary to determine the effect of water absence in theenzyme activity during consecutive reactions and to evaluate alterna-tives for enzyme recovery during successive reactions.

The aim of this work is to test an anhydrous medium in Novozym435 catalyzed-process to produce FAME with a low acid value. Inaddition, the behavior of Novozym 435 in the anhydrous mediumwasdiscussed and enzyme recovery alternatives were proposed.

MATERIALS AND METHODS

Materials Filtered WFO collected from restaurants and crude rapeseed oil froma local factory in Southern Chile was used as feedstock. The WFO was characterized byan acid value of 5.61 mg KOH/g oil, 2.8% FFA, 1.3% (v/v) water content, 0.1% (wt)monoacylglycerols (MG), 3.3% (wt) diacylglycerols (DG) and 925 kg/m3 of specificgravity. The rapeseed oil was characterized by an acid value of 0.8 mg KOH/g oil, 1.6%FFA and 927 kg/m3 of specific gravity. C. antarctica lipase immobilized on acrylic resin(Novozym 435) donated by Novo Industries (Denmark) was used as catalyst. 3 Åmolecular sieves (Sigma-Aldrich) were used to generate the anhydrous medium.Linoleic acid with a 99% purity grade was acquired from Sigma Aldrich and used toanalyze the FFA's effect on the reaction rate. Chromatographically pure methylheptadecanoate, 1,2,3-butanetriol and 1,2,3-tricaprinoylglycerol were used as internalstandards. All other chemicals were of analytical grade.

Biocatalysis in an anhydrous medium FAME production using Novozym 435in an anhydrousmediumwas studied by adding 3 Åmolecular sieves to the reaction. Allreactions were carried out in flasks containing 1 mL of WFO (0.925 g) and 15%Novozym 435 (% wt based on oil weight) as the biocatalyst. A methanol-to-oil molarratio of 4:1 was used. Methanol was added in two steps in order to avoid lipaseinhibition. In the first step, one-third of the total molar ratio was added at the initialreaction time and in the second step, the remaining two-thirds of the total methanolmolar ratio were added at 8 h of reaction time (11). The flasks were incubated in ashaker at 35°C and stirred at 200 rpm. Each flask was managed as a destructive sampleat 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, and 12 h of reaction time and each condition was carriedout in triplicate. Controls were carried out under the same conditions but withoutmolecular sieves.

To calculate the dosage of molecular sieves, first the maximal theoretical watercontent in the reaction was estimated. This value was established by taking the initialwater content in the WFO and the amount of water produced during the reaction byesterification of FFA contained in the WFO. The water produced by esterification of FFAwas estimated by using the mass balance, considering that 1 mol of esterified FFAproduces 1 mol of water. With an FFA content of 2.8% and water content of 1.3% in theWFO described above, a total water content of 4.1% was used to estimate the molecularsieve dosage in the reaction, as shown in Eq. 1.

Molecular sieves ¼Voild ρoild H2OWAC

≈ 1d 0:925d 4:120

= 0:19 g ð1Þ

Where Voil (mL) is the volume of oil added, ρoil (g/mL) is the density of the oil, H2Ois the total water content during the reaction (water content in the oil more waterproduced by esterification of FFA) (% v/v) andWAC (%) is the water absorption capacityof the sieves. At the end of the reaction period, the samples were stored at 4°C to stopthe reaction. The samples were centrifuged and the upper layer was extracted foranalysis. FAME yield, acid values and MG and DG content were analyzed in accordancewith the analytical procedures.

The effect of free fatty acid on the reaction yield As the used WFO wasmainly composed by linoleic acid (49.5%), this acid was chosen to investigate the effectof FFA addition on the yield when FAME is produced by biocatalysis. Five experimentswere carried out, each of which was repeated four times. For each experiment, asynthetic substrate was prepared in a flask by mixing synthetic linoleic acid and the

vegetable oil. The concentrations in each experiment were 0, 2000, 5000, 8000 and10,000 mg of linoleic acid per liter of rapeseed oil (the total volume in each experimentwas 1 mL). The used rapeseed oil was a typical virgin vegetable oil characterized by avery low FFA content. Therefore, the FFA content in rapeseed oil was not considered inthe FFA concentration calculate. The reaction conditions were the following: Methanol-to-oil ratio of 3:1 (mol/mol); 15% Novozym 435 (% wt based on oil weight); 35°C; andstirring at 200 rpm during 12 h of reaction time. Like the previous experiment,methanol was added in two steps. At the end of the reaction period, the samples werestored at 4°C to stop the reaction. The upper layer of the centrifuged samples wasanalyzed to quantify the FAME yield according to the analytical procedures.

The results were compared using two-way ANOVA analysis. Prior to carrying outthis test, the normality of the data was analyzed using the Shapiro–Wilk test. Variancehomogeneity was analyzed using the Levene statistic (PN0.05). The results obtainedshowed that the data were both normal and homogeneous, which enabled theparametric analysis to be conducted.

Treatments for enzyme reutilization in an anhydrous medium Differenttreatments to recover enzyme activity for further reutilization were investigated. Priorto the enzyme recovery treatments, reactions of FAME production were carried out inflasks which were incubated in a shaker under the same operating conditions. Theoperational conditions were the following: Methanol-to-oil ratio of 3.8:1 (mol/mol);15% Novozym 435 (% wt based on oil weight); 44.5°C; stirring at 200 rpm; and 12 h ofreaction time. Methanol was added in two steps as described previously (11).

For each reaction, 8 mL of WFO (7.4 g) with an acid value of 5.61 mg KOH/g oilwere used. To maintain an anhydrous medium, 0.95 g of 3 Å molecular sieves wereadded to each reaction flask, which were calculated according to Eq. 1. Each reactionwas carried out in triplicate. Samples of 70 μL were taken consecutively at 3, 6, 9 and12 h of reaction time. The samples were centrifuged and the upper layer was extractedto analyze the FAME yield, according to the analytical procedures.

For enzyme recovery, the molecular sieves were separated with a strainer after thetransesterification reaction. A paper filter over a vacuum system was placed underneaththe strainer containing the molecular sieves to retain the enzymes. The product of thereaction was passed through the two filters: Themolecular sieves captured in the strainerwere eliminated, whereas the enzymes captured in the paper filter were placed in a newflask for the recovery treatment.

Three treatments to reutilize the enzymeswere performed according to the referencestudies (12,13,9) and previous experiments (data not shown). A control experimentreusing the enzymes without treatment was also carried out. The treatments were thefollowing: (i) Acetone washing: The enzymes were washed by adding a small dosage ofacetone to the flask containing the enzymes. The flask was shaken and the liquid residuewas eliminated. Thiswas repeated successively until the liquidwas clear. The total volumeused in this processwas about10 mLof acetoneper gramof enzyme. Subsequently, awashwith 10 mL of WFO per gram of enzyme was carried out to eliminate residual acetone inthe enzymes. (ii) Waste frying oil washing: The enzymes were washed in a sufficientquantity of WFO to maintain the enzymes submerged in the flask. The flask was shakenand the liquid residuewaseliminated. Thewashingwas repeated 3 times. (iii) tert-Butanolwashing: The washing was carried out by adding a specific dosage of tert-butanol to theflask containing the enzymes. Then, the mixture was shaken and the residual liquid waseliminated. This sequencewas repeated consecutively until the liquid was clear. The totalvolume used in this process was about 10 mL of tert-butanol per gram of enzyme. Afterthis, a wash with about 10 mL of WFO per gram of enzyme was carried out to eliminateresidual tert-butanol in the enzymes.

After eachwashing step, a known dosage ofWFOwas added to the flasks containingthe enzymes, which were incubated in oil overnight at ambient temperature. Afterabout 10 h, both methanol and 3 Å molecular sieves were added to carry out a newFAME production reaction in the same incubation medium. The control was carried outby transferring the enzymes directly to a new reaction for FAME production. Cycles thatconsisted of both a reaction for FAME production and a treatment for enzyme recoverywere repeated 4 times.

FAME production in an anhydrous medium using tert-butanol as a co-solvent As an alternative to enzyme reuse for FAME production using an anhydrousmedium, the addition of tert-butanol as a co-solvent in the reaction was investigated. Fiveexperiments of enzymatic FAME production were performed in triplicate to define theoptimal tert-butanol dosage. The experiments were carried out by adding differentdosages of tert-butanol to the reaction: 0, 0.5, 0.75, 1 and 1.25 (% v/v based on oil volume).The reactions were carried out in an anhydrous medium with a methanol-to-oil ratio of3.8:1 (mol/mol) and 15% Novozym 435 (%wt based on oil weight) (11). The experimentsalso included the dosage of 3 Å molecular sieves and the steps addition of methanol.Under these conditions, 1.9 mL of WFO (1.76 g) was added to the flask which wasincubated at 44.5°C and 200 rpm during 4 h of reaction time. After reaction, sampleswerecentrifuged and the upper layer was maintained at 85°C for 20 min to eliminate residualtert-butanol from the samples. This upper layer was analyzed using gaseous chromatog-raphy to quantify the FAME content, according to the analytical procedures. The optimaldosage of tert-butanol was obtained by analyzing the reaction productivity. The resultswere compared using one-way ANOVA analysis and Duncan test.

The reutilization of the enzyme in the FAME production reaction was studied byusing the selected dosage of tert-butanol. To carry out the experiments, consecutivecycles of enzymatic FAME production using the same enzymes (reutilized) were carriedout in an anhydrous medium and in a tert-butanol system. Each experiment wascarried out in flasks by adding 5 mL ofWFO (4.6 g), 15% of Novozym 435 (%wt based on

acid

val

ue [

mg

KO

H/g

]

0

1

2

3

4

5

6

7ControlAnhydrous medium

time [h]0 2 4 6 8 10 12

FAM

E y

ield

[%

]

0

20

40

60

80

100

A

B

FIG. 1. Acid values (A) and FAME yield (B) during the reaction in an anhydrous mediumwith a methanol-to-oil ratio of 4:1 (mol/mol), 15% (wt) Novozym 435 (based on theweight of oil), 35°C, 200 rpm and 12 h of reaction time.

ENZYMATIC BIODIESEL PRODUCTION IN ANHYDROUS MEDIUM 585VOL. 112, 2011

oil weight) and a methanol-to-oil ratio of 3.8:1 (mol/mol). Methanol was added to theflask in two steps, similar to the previous experiment. In addition, both the selecteddosage of tert-butanol and an amount of 3 Å molecular sieves (estimated according toEq. 1) were added to the flasks. Under these conditions, the flasks were incubated at44.5°C and 200 rpm during 12 h of reaction time. All the experiments were carried outin triplicate and samples were taken throughout the reaction time. The upper layer ofthe centrifuged samples was analyzed to determine the FAME yield, according to theanalytical procedures.

Analytical procedures The acid value and FFA content in both the feedstockand throughout the reaction time were determined by titration with a KOH solution ofknown concentration and using phenolphthalein as indicator. Water and sedimentswere measured according to ASTM Standard Method D 1976–97 (14).

FAME yield was determined by FAME quantification from the upper layer of thesamples obtained after each reaction, previously centrifuged. The quantification wascarried out using a Clarus 600 chromatograph coupled with a Clarus 500T massspectrometer from Perkin Elmer (GC-MS). An Elite-5 ms capillary column with alength of 30 m, thickness of 0.1 μm and internal diameter of 0.25 mm was used. Thevials were prepared by adding 3 μg of the sample to 100 μL methyl heptadecanoate asan internal standard (initial concentration of 1300 mg/L). The following temperatureprogram was used: 50°C for 1 min and then increasing temperature at a rate of1.1°C/min up to 187°C. Both the injector and detector temperatures were 250°C andHe was used as the carrier gas.

MG and DG in feedstock and reaction products were quantified using an HP 6890series gas chromatography system (GC-MS) with an adaptation of the EN-14214methodology (15). A 50-m long BPX-5 column with a thickness of 0.5 μm and aninternal diameter of 0.32 mmwas used. The vials were prepared by combining 10 mg ofthe sample with 0.8 μL of 1,2,3-butanetriol and 10 μL of 1,2,3-tricaprinoylglyceroldissolved in pyridine as internal standards. 2,2,2-trifluoro-N-methyl-N-trimethylsilylacetamide (MSTFA) was added to the vials, which were then shaken and incubated for15 min to transform the sample into more volatile siliade components. Subsequently,the preparation was dissolved in 0.8 mL of n-heptane. The following temperatureprogram was used: 15°C for 1 min and three consecutive ramps of 15°C/min to 180°C,7°C/min to 230°C and 10°C/min to 320°C and 320°C for 15 min. The detectortemperature was 380°C and a split rate 10 was used for injection of 1 μL of sample.

RESULTS AND DISCUSSION

Biocatalysis in an anhydrous medium In order to reduce theacid value of the produced FAME, the alternative of carrying outFAME production reaction in an anhydrous medium was studied.Adsorbent materials such as blue silica gel and molecular sieves ofdifferent sizes have been shown to be effective in water removalduring biodiesel production (2). However, in this study, blue silicagel was discarded due to results from previous experiments showingpossible methanol adsorption, negatively interfering with thereaction (data not shown). The adsorption of methanol could alsooccur when molecular sieves with large pore size are used. Thus, 3 Åmolecular sieves were chosen to produce an anhydrous mediumthrough water absorption.

The acid value and FAME yield results obtained using an anhydrousmedium and a control run are shown in Fig. 1. The acid value obtainedfrom the reaction using the anhydrous medium was maintained inabout 1 mgKOH/g oil throughout the entire reaction time,whereas thecontrol showed a value higher than 3 mg KOH/g oil by the end of thereaction (Fig. 1A). The values obtained using the anhydrous mediumare close to those established by the biodiesel norm (15). As FFApresent inWFOwere esterified producingwater and FAME at the startto the reaction (Fig. 1A), water was removed from the reaction byadsorption onto the molecular sieves, avoiding hydrolysis reactionsand further FFA production. Therefore, the biocatalysis in ananhydrous medium produces a higher quality product compared to atypical standard reaction using Novozym 435.

FAMEyieldwasalso analyzed in ananhydrousmedium(Fig. 1B). Theresults obtained show that water removal during FAME productiongenerateda significant increase in FAMEyieldusingNovozym435 as thecatalyst. These results are in agreement with those obtained by Li et al.(16) who established that using 3 Å molecular sieves as adsorbent toremove excessive water could significantly increase the FAME yieldwhen 100% FFA is used as rawmaterial for biodiesel production. In thisstudy, WFO containing only 2.8% FFA and 1.3% water was employed asthe raw material, but FAME yield increased drastically. In addition, the

control run only using molecular sieves indicates that this material didnot catalyze the reaction (data not shown). Therefore, the increment inFAME yield is only related to the anhydrous medium. This corroboratesthe results obtained by Tamalampudi et al. (10) who suggest thatNovozym 435 transesterification activity is inhibited by the presence ofadded water and that it needs a nearly anhydrous medium for anefficient performance.

A reasonable explanation for the high FAME yield obtained may berelated to the work of Cabrera et al. (17). They suggest that lipase mayexist in two different structural forms; in one form, the site of the lipaseis isolated from themediumwhereas in the other form, the active site isexposed to the reaction medium. In a homogeneous aqueous mediumthe lipase is in equilibrium between these two structures. In the case ofNovozym 435, it is immobilized in acrylic resin with hydrophilicproperties. Therefore, the higher yield in the reactionmayoccurbecauseof interactions with the hydrophobic medium, where the lipase shiftstowards the open structure form, increasing its activity. Theseconformational changes enable lipases to be greatly altered bycontrolled immobilization of the support material properties. This is inaccordance with the results obtained by Samukawa et al. (13) whoestablished that preincubation of the enzyme in oil prior to the reactionimproves the yield. This is because water adsorbed during the reactionprevented oil penetration, whereas this did not occur with thepreincubated enzyme.

The MG and DG content were also quantified in the same reactionwith the abovementioned anhydrous medium. MG and DG areintermediary products in biodiesel production which have hygro-scopic characteristics that can cause damage to the engine. Fukuda et

586 AZÓCAR ET AL. J. BIOSCI. BIOENG.,

al. (6) explain that despite the high FAME yield which can be achievedby regiospecific lipases, considerable amounts of MG and DG remainafter the process. In this study, it was speculated that MG and DGcontent in the product could be affected by the different performanceof the enzyme in the anhydrous medium. Therefore, MG and DGcontent were measured throughout the reaction in the anhydrousmedium in order to compare them with a control run (Fig. 2).

Fig. 2A shows that MG content throughout the reaction was lowand quite similar in the anhydrous medium and in the control with0.6±0.4% and 0.7±0.1% at the end of the reaction, respectively. Thesevalues are in accordance with those established by the norm for MGcontent in biodiesel (b0.8%).

The DG content was quickly reduced in the beginning of thereaction (after 2 h) in the anhydrous medium and in the control run(Fig. 2B). This result may occur because intermediary products such asDG are more available substrates for the enzymatic catalysis,according to results reported recently (11). However, higher valuesof DG content were detected until the end of the reaction time in thecontrol. On the contrary, when the reaction was carried out inanhydrous medium, the DG content reached a maximum value of 2%,which declined to 0.3% by the end of reaction (Fig. 2B). The valuereached is close to the biodiesel norm (EN-14214) which establishes amaximum limit of 0.2% DG content in the FAME. The reason for thislower DG content may be related to the molecular sieves added to thereaction. Other studies have established that the use of materials suchas silica gel could produce acyl migration, which in turn increasesFAME yield when lipases with specificity are used (18). Thus, theaddition of molecular sieves could reduce DG content because it leadsto acyl migration in the reaction. Therefore, the sieves added to the

mon

ogly

ceri

de c

onte

nt in

pro

duct

[%

]

0

1

2

3

4

5ControlAnhydrous medium

time [h]0 2 4 6 8 10 12

digl

icer

yde

cont

ent i

n pr

oduc

t [%

]

0

1

2

3

4

5

A

B

FIG. 2. Monoacylglycerol (A) and diacylglycerol (B) content in the product during thereaction in an anhydrous medium with a methanol-to-oil ratio of 4:1 (mol/mol), 15%(wt) Novozym 435, 35°C, 200 rpm and 12 h of reaction time.

reaction could produce twopositive effects in the biocatalytic reaction:an improvement in FAME yield by generating an anhydrous mediumand an improvement in FAME properties through a reduction in boththe acid value and the DG content.

The effect of free fatty acids on the reaction yield Theperformance of lipases in carrying out hydrolysis, esterification andtransesterification reactions may be affected in an anhydrous medium.When an anhydrous medium was used in the reaction, FFA contentdiminished throughout the reaction, probably by inhibition of esterifi-cation andhydrolysis of triacylglycerides (TG). It was therefore assumedthat the higher FAME yield reached in anhydrous medium could berelated to the lower FFA presence on the reaction. To analyze the effectof FFA presence on the reaction yield, an experiment was carried out byadding a known dosage of synthetic linoleic acid to the rapeseed oil(Fig. 3).

Fig. 3 shows a significant reduction in FAME yield when differentdosages of FFA were added to the reaction. In these cases, the enzymeesterified these acids to produce FAME and water. The producedwater reacted with TG producing more FFA and glycerol. Therefore,the enzyme carried out three reactions in parallel: hydrolysis,esterification and transesterification, which could generate a lesseffective and slower process. In addition, it has been reported thatexcess water reduces the methanolysis as it acts as a competitiveinhibitor for lipase-catalyzed esterification or transesterification(18,10). As a result, water produced by esterification could increaseinhibition of the transesterification reaction, diminishing the reactionrate. Therefore, the higher FAME yield of the biocatalyst in ananhydrous medium could be related to the avoidance of side reactionsof hydrolysis and esterification, being FAME mainly produced bytransesterification.

Treatments for enzyme reutilization in an anhydrous mediumThemain advantage of using immobilized lipases is that the enzyme

can be used repeatedly in a semi-continuous process. However, thisobjective is not always reached under the optimized conditions as shortchain acyl acceptors can produce a loss of enzyme activity in successivereactions (4). In this study, the stability of the enzyme and treatmentsfor its reutilization in an anhydrous medium were examined. Fig. 4shows the results obtained by different treatments to maintain lipaseactivity after each FAME production reaction in an anhydrous medium.Fig. 4A shows the reutilization of Novozym 435without treatment afterreaction. According to the results obtained, FAME yield decreasedconsiderably in the second cycle. This tendency was maintained insuccessive cycles, with values close to 0% of FAME yield in the fourth

Linolenic acid concentration [mg/L]

FAM

E c

onte

nt in

pro

duct

[%

]

30

35

40

45

50

55a

b b

ab

b

2000 50000 8000 10000

FIG. 3. FAME yield to different linoleic acid concentration on feedstock. Operationalconditions: methanol-to-oil molar ratio of 3:1, 15% (wt) Novozym 435, 35°C, 200 rpmand 12 h of reaction time. The letters “a” and “b” show significant differences (Pb0.05).

0

20

40

60

80

100 3 h6 h9 h12 h

FAM

E y

ield

[%

]

0

20

40

60

80

100

0

20

40

60

80

100

cycle [Nº]1 2 3 4

0

20

40

60

80

100

A

B

C

D

FIG. 4. Comparison of reutilization treatments of Novozym 435 after the reactions toFAME production with a methanol-to-oil ratio of 3.8:1 (mol/mol), 15% (wt) Novozym435 (based on the weight of oil), 44.5°C, 200 rpm and 12 h of reaction time. (A) Control,(B) acetone washing, (C) waste frying oil washing, (D) tert-butanol washing.

ENZYMATIC BIODIESEL PRODUCTION IN ANHYDROUS MEDIUM 587VOL. 112, 2011

cycle. These results are in accordancewith the findings of Ognjanovic etal. (4), who reported 0% FAME yield in the fourth cycle using Novozym435 and methanol as the acyl acceptor in an organic medium. Thus,although the activity of the enzyme increased in the first cycle using ananhydrous medium compared to the control (Fig. 1), the loss of activityseems to be similar in both types ofmediumwhen successive cycles arecarried out (Fig. 4A). Lipases have shown high synthesis activity andstability in hydrophobic solvents, but alcohol and glycerol areimmiscible in those solvents (19). This situation could produce poorsolubilization of polar compounds in the medium, leading to theiradsorption onto the lipases hydrophilic support and therefore provok-ing a low transesterification rate (19). So far, an alternative process isneeded to allow the reuse of the enzyme achieving a low-cost processwhich can feasibly be implemented at an industrial scale.

The results of the first treatment studied for reusing the enzymeare shown in Fig. 4B. Applying an acetonewashing step a higher FAMEyield was achieved in the second cycle compared to the control(Fig. 4B and A, respectively). This result is caused by the fact thatacetone is a hydrophilic solvent that could remove the glycerol andthe methanol adsorbed onto the hydrophilic support material of theenzyme. However, in the third cycle the FAME yield diminisheddrastically (Fig. 4B). Acetone is a hydrophilic solvent with a very lowlog P (partition coefficient in a standard octanol-water two-phasesystem) value of −0.24. According to Yu et al. (9) water has higheraffinity to these hydrophilic solvents rather than to the enzyme. It hasbeen reported that Novozym 435 can contain a sufficient quantity ofwater to preserve its catalytic activity in an anhydrousmedium (4,10).

Therefore, the enzyme activitymay be reducedwhen using an acetonewashing step, because the enzyme might lose its flexibility confor-mation due to the lack of bound water (9). The enzyme washing stepusingWFO followed by overnight incubation inWFOwas also studied(Fig. 4C). The activity of Novozym 435 was higher in the third cycle incomparison to the two previous alternatives. Samukawa et al. (13)reported that Novozym 435 preincubated in methyl oleate andsubsequently in soybean oil could enhance the rate of FAMEproduction through impregnation of these compounds into theenzyme support material. In this study, it was assumed that theenzyme could contain methyl ester residues when incubated in WFOand therefore, the treatment should be similar to that carried out bySamukawa et al. (13).

This treatment using WFO is advantageous compared to the acetonewashing process because it is a cheaper andmore environmental friendlyprocess. In addition, the sameWFO that was used for the incubation wasused subsequently for the reaction of FAMEproduction, and therefore lessequipment is needed to carry out this process. In another study, Chen andWu (12) achieved a FAME yield five times higher when the enzyme wasincubatedovernight in soybeanoil.However,when theyused the sameoilin incubation experiments to recover the enzyme after the reaction, theactivity began to decay after the fourth reutilization. Although theadvantages of usingWFO in enzyme recovery, FAMEyield also declined inthis study, as reported by Chen andWu (12). The reason may be the lowsolubility of alcohol and glycerol in the oil (hydrophobic wash), whichimpedes the efficient washing of the enzyme by oil.

As thewashingprocesswithhydrophobic andhydrophilic solventsdidnot allow an efficient recovery of the enzyme activity bymore than threesuccessive reactions in an anhydrous medium, a moderate polar solventwas also investigated. Tert-butanol, a tertiary alcohol with a log P value of0.35 has been shown to improve enzyme activity more than linearalcohols. This fact could be attributed to the differences inmiscibilitywithtriglycerides, as compared to alcohols with the same carbon numbers,branched ones have better miscibility with triglycerides compared tolinear isomers (9). Therefore, a washing treatment of the enzymes withtert-butanol, followed by incubation in WFO, was conducted. The resultsshowed that this enzyme pretreatment achieved the best results,maintaining a higher activity over the time compared to the previousexperiments (Fig. 4D). It is likely that tert-butanol recovered the enzymeactivity because it has the advantages of both hydrophilic andhydrophobic solvents but none of the drawbacks. Therefore, tert-butanolshould promote the removal of both glycerol and methanol from thelipase support material because of its hydrophilic properties, while itshydrophobic properties should help maintain a high level of lipaseactivity. This is in accordance to Chen and Wu (12) who found that tert-butanol can be even used to regenerate a deactivated, immobilizedenzyme such as Novozym 435. Therefore, tert-butanol is a promisingalternative for enzyme recovery after reaction in an anhydrous medium,and is also highly stable and less reactive than other butanol isomers.Based on these results, a new experimentwas carried out which includedtert-butanol as a co-solvent in the reaction in an anhydrous medium.

Successive reactions to FAME production in an anhydrousmedium using tert-butanol as a co-solvent A previous studyfound that tert-butanol is inert in the methanolysis system, whereas it isalso a potential co-solvent that could maintain enzymatic activity (20).According to this, the performance of tert-butanol as a co-solvent wasstudied in an anhydrous medium to determine its ability to maintainNovozym 435 activity in successive reactions. A preliminary study todetermine thedosage of tert-butanol that should be added to the reactionwas carried out. The results showed that the highest FAME productivitywas reached when 0.75% (v/v) and 1% (v/v) of co-solvent was added tothe reaction (Fig. 5). These resultswere similar to those obtained in otherstudies (20). The reasonmaybe that the presence of tert-butanol at 0.75%(v/v) and 1% (v/v) could increase the solubility between oil andmethanol, avoiding enzyme inhibition by methanol. For lower dosages,

time [h]0 20 40 60 80 100 120 140 160 180 200

FAM

E y

ield

[%

]

0

20

40

60

80

100

Control Anhydrous medium with tert-butanol

FIG. 6. Time course of FAME yield during 17 batch reaction cycles with a methanol-to-oil ratio of 3.8:1 (mol/mol), 15% (wt) Novozym 435 (based on the weight of oil), 44.5°C,0.75% v/v of tert-butanol, 0.9 g of molecular sieves, 200 rpm and 12 h of reaction time.Symbols: in tert- butanol system (closed circles), control (open circles).

588 AZÓCAR ET AL. J. BIOSCI. BIOENG.,

tert-butanol is not able to prevent the enzyme inhibition produced bymethanol. On the other hand, when tert-butanol dosagewas higher than1% (v/v), FAME yield decreased gradually. According to Li et al. (20) thiscould be caused by the dilution effect of reactantswith high tert-butanoldosage, diminishing mass transfer and negatively affecting FAME yield.Thus, considering the necessity of solvent recovery (5), in the followingexperiments a 0.75% (v/v) of tert-butanol was utilized.

Successive reactions for FAME production were carried out in ananhydrousmediumunder previously optimized operational conditions,using the selected dosage of tert-butanol (11) (Fig. 6). A controlwithout tert-butanol as co-solvent was also carried out, showing thatFAME yield was drastically reduced in the second cycle (b20% FAMEyield). The best results were obtained in the system using the co-solvent, where FAME yield was maintained over 50% after 17 cycles(Fig. 6). According to similar findings from other studies, the enzyme isinhibited throughout the reaction (5). The inhibitory effect at thebeginning of the reaction is due to the presence of methanol which haspoor miscibility in oil. Subsequently, once methanol concentrationdecreases, inhibition is caused by a glycerol layer coating the catalyst.Therefore, the positive effect of tert-butanol in the reaction could befirst related to the increment in the oil-methanolmiscibility, preventingdirect contact between enzyme and alcohol at the beginning of thereaction and improving the reaction yield. Secondly, due to tert-butanol's hydrophilic properties, it is able to dissolve both methanoland glycerol during the reaction, preventing enzyme inhibitionthroughout the reaction. Thirdly, tert-butanol's hydrophobic propertiesmaintained the high lipase activity through the successive reactions.

According to Yu et al. (9), different properties such as viscosity,dielectric constant, solubility parameters and log P should be consideredwhen choosing a co-solvent to improve enzymeperformance in biodieselproduction. Another aspect which should be considered is the feasibilityof recovering the co-solvent through distillation, which is related to theboiling point. In this sense, tert-butanol has advantages compared toothers co-solvents reported, such as amyl alcohol, which has very similarproperties compared to the tert-butanol but a higher boiling point(102°C) close to the boiling point of thewater. As the boiling point of tert-butanol is 82°C and the distillation of methanol occurs at 65°C, therecovery of tert-butanol should not increase the amount of energy spentthat much in comparison to the advantages of using this co-solvent.Therefore, employing tert-butanol as a co-solvent could allow the use ofan anhydrous medium as an industrial alternative for FAME productionusing Novozym 435 as biocatalyst.

tert-butanol dosage [% V/V]0 0.5 0.75 1 1.25

Prod

uctiv

ity [

g FA

ME

/L/h

]

0

1

2

3

4

5

6

7

8

a a

bb

a

FIG. 5. Productivity of methanolysis reaction in an anhydrous medium for different tert-butanol doses. Operational conditions: methanol-to-oil ratio of 3.8/1 (mol/mol), 15%(wt) Novozym 435 at 44.5°C and 200 rpm during 12 h of reaction time. The letters “a”and “b” show significant differences (Pb0.05).

Conclusions The results show that the use of an anhydrousmedium (resulting from water extraction by using molecular sieves),FAME yield was improved by avoiding hydrolysis and esterificationreactions, producing FAME mainly through transesterification of TG. Inaddition, theuseof thismediumdecreasedboth the acid value and theDGcontent in the final product, reaching a value close to that established bythe biodiesel norm. The enzyme activity cannot be recovered successivelyneither using a hydrophilic washing step with acetone nor byhydrophobic washing with WFO. However, 17 successive cycles ofFAMEproductionusing tert-butanol as amoderate polar co-solvent, showthatNovozym435canbe reused inanhydrousmedium. These results alsoshow that the anhydrous medium could enable the implementation oflipase-catalyzed processes on an industrial scale for biodiesel productionmainly by transesterification reaction. Different raw materials could beused for this, achieving properties close to the norm and potentiallyavoiding post-treatments to refine the produced biodiesel.

ACKNOWLEDGMENTS

This work was supported by Chilean FONDECYT Project1090382;Chilean CONICYT Project79090009; and UFZ-Centre for Environmen-tal Research Leipzig-Halle.

References

1. Hernandez-Martin, E. and Otero, C.: Different enzyme requirements for thesynthesis of biodiesel: Novozym (R) 435 and Lipozyme (R) TL IM, Bioresour.Technol., 99, 277–286 (2008).

2. Li, N., Zong, M., andWu, H.: Highly efficient transformation of waste oil to biodiesel byimmobilized lipase from Penicillium expansum, Process Biochem., 44, 685–688 (2009).

3. Maceiras,R., Vega,M.,Costa, C.,Ramos,P., andMárquez,M.:Effect ofmethanol contenton enzymatic production of biodiesel fromwaste frying oil, Fuel, 88, 2130–2134 (2009).

4. Ognjanovic, N., Bezbradica, D., and Knezevic-Jugovic, Z.: Enzymatic conversionof sunflower oil to biodiesel in a solvent-free system: process optimization and theimmobilized system stability, Bioresour. Technol., 100, 5146–5154 (2009).

5. Royon, D., Daz, M., Ellenrieder, G., and Locatelli, S.: Enzymatic production ofbiodiesel from cotton seed oil using t-butanol as a solvent, Bioresour. Technol., 98,648–653 (2007).

6. Fukuda, H., Hama, S., Tamalampudi, S., and Noda, H.: Whole-cell biocatalysts forbiodiesel fuel production, Trends Biotechnol., 26, 668–673 (2008).

7. Azócar, L., Ciudad, G., Heipieper, H., and Navia, R.: Biotechnological processes forbiodiesel production using alternative oils, Appl. Microbiol. Biotechnol., 88,621–636 (2010).

8. Wang, L., Du, W., Liu, D., Li, L., and Dai, N.: Lipase-catalyzed biodiesel productionfrom soybean oil deodorizer distillate with absorbent present in tert-butanolsystem, J. Mol. Catal. B: Enzym., 43, 29–32 (2006).

ENZYMATIC BIODIESEL PRODUCTION IN ANHYDROUS MEDIUM 589VOL. 112, 2011

9. Yu, D., Tian, L., Wu, H., Wang, S., Wang, Y., Ma, D., and Fang, X.: Ultrasonicirradiation with vibration for biodiesel production from soybean oil by Novozym435, Process Biochem., 45, 519–525 (2010).

10. Tamalampudi, S., Talukder, M., Hama, S., Numata, T., Kondo, A., and Fukuda, H.:Enzymatic production of biodiesel from Jatropha oil: a comparative study ofimmobilized-whole cell and commercial lipases as a biocatalyst, Biochem. Eng. J.,39, 185–189 (2008).

11. Azócar, L., Ciudad, G., Heipieper, H., Muñoz, R., and Navia, R.: Improving fattyacid methyl ester production yield in a lipase-catalyzed process using waste fryingoils as feedstock, J. Biosci. Bioeng., 109, 609–614 (2010).

12. Chen, J. and Wu, W.: Regeneration of immobilized Candida antarctica lipase fortransesterification, J. Biosci. Bioeng., 95, 466–469 (2003).

13. Samukawa, T., Kaieda, M., Matsumoto, T., Ban, K., Kondo, A., Shimada, Y.,Noda, H., and Fukuda, H.: Pretreatment of immobilized Candida antarcticalipase for biodiesel fuel production from plant oil, J. Biosci. Bioeng., 90, 180–183(2000).

14. American Society for Testing and Materials (ASTM): Annual Book of ASTMStandards: Section 5 - petroleum products, lubricants, and fossil fuels. ASTMInternational, West Conshohocken (2008).

15. The British Standards Institution: BS EN 14214: Automotive fuels. Fatty acidmethylesters (FAME) for diesel engines. Requirements and test methods. BSI,London (2003).

16. Li, W., Du, W., and Liu, D.: Rhizopus oryzae IFO 4697 whole cell-catalyzedmethanolysis of crude and acidified rapeseed oils for biodiesel production in tert-butanol system, Process Biochem., 42, 1481–1485 (2007).

17. Cabrera, Z., Fernandez-Lorente, G., Fernandez-Lafuente, R., Palomo, J., andGuisan, J.:Novozym 435 displays very different selectivity compared to lipase fromCandida antarctica B adsorbed on other hydrophobic supports, J. Mol. Catal. B:Enzym., 57, 171–176 (2009).

18. Du, W., Xu, Y., Liu, D., and Li, Z.: Study on acyl migration in immobilized lipozymeTL-catalyzed transesterification of soybean oil for biodiesel production, J. Mol.Catal. B: Enzym., 37, 68–71 (2005).

19. Halim, S. and Kamaruddin, A.: Catalytic studies of lipase on FAME productionfrom waste cooking palm oil in a tert-butanol system, Process Biochem., 43,1436–1439 (2008).

20. Li, L., Du, W., Liu, D., Wang, L., and Li, Z.: Lipase-catalyzed transesterification ofrapeseed oils for biodiesel production with a novel organic solvent as the reactionmedium, J. Mol. Catal. B: Enzym., 43, 58–62 (2006).