Industrial Crops and Products 20 (2004) 275–279

Extraction of oil fromJatropha curcasL. seed kernels byenzyme assisted three phase partitioning

Shweta Shah, Aparna Sharma, M.N. Gupta∗

Department of Chemistry, Indian Institute of Technology, Delhi, Hauz Khas, New Delhi 110016, India

Received 15 April 2003; accepted 16 October 2003

Abstract

Three phase partitioning has been evaluated for extraction of oil fromJatropha curcasL. seeds. This process consisted ofsimultaneous addition oft-butanol (1:1, v/v) and 30% (w/v) ammonium sulphate to the slurry prepared fromJatrophaseedkernels. Combination of sonication and enzyme treatment with a commercial preparation of fungal proteases at pH 9, led to 97%oil yield within 2 h.© 2003 Elsevier B.V. All rights reserved.

Keywords:Enzyme assisted three phase partitioning;Jatrophaseed kernel oil; Three phase partitioning

1. Introduction

Recently, we have described a new approach forextraction of oil from plant materials, in general, andsoybean in particular (Sharma et al., 2002a). In this ap-proach, called three phase partitioning (TPP), appro-priate amounts of ammonium sulphate andt-butanolwere added to an aqueous suspension of soybean meal.This led to the formation of three distinct phases: up-per organic phase, lower aqueous phase and interfa-cial precipitate (consisting mostly of proteins). The oilwas found in the organic solvent phase and could berecovered by evaporating thet-butanol. TPP is simpleto carry out, has short processing time (about 1 h), andis easy to scale up. It may be also useful to compare

∗ Corresponding author. Tel.:+91-11-265-91503;fax: +91-11-265-81073.

E-mail address:mn gupta@hotmail.com (M.N. Gupta).

this process with solvent extraction. Hexane is the sol-vent which is generally used in that process.t-Butanolhas a higher boiling point (84◦C) than hexane (69◦C),hence addition of volatile organic solvents to the at-mosphere during TPP will be much lower even if opensystems are used (Rosenthal et al., 1996). Thus, sim-pler extraction designs will be possible with TPP.

Jatropha is a genus comprising 70 species grow-ing in tropical and subtropical countries (Openshaw,2000). The seed kernel contains 40–60% (w/w) oil(Makkar et al., 1997). The presence of some antinu-tritional factors renders this oil unsuitable for use incooking (Makkar et al., 1997; Gubitz et al., 1998).Some efforts are being made for removal of theseantinutritional factors (Haas and Mittelbach, 2000).Meanwhile, as such, it has been used for productionof biodiesel (Foidl et al., 1996).

The present work demonstrates the application ofTPP in extraction of oil from seed kernels ofJatropha

0926-6690/$ – see front matter © 2003 Elsevier B.V. All rights reserved.doi:10.1016/j.indcrop.2003.10.010

276 S. Shah et al. / Industrial Crops and Products 20 (2004) 275–279

curcas L. An important further step in the processdevelopment has been the pretreatment of the plantmaterial with enzymes before TPP.

2. Materials and methods

Jatrophaseeds were obtained from Dr. Zope, Col-lege of Forestry, Akola, India. ProtizymeTM waspurchased from Jaysons Agritech Pvt. Ltd., Mysore,India. It is reported to contain mostly acid (pH op-timum range 3–4), neutral (pH optimum range 5–7)and alkaline proteases (pH optimum range 7–10)from Aspergillus flavuswith a specific activity of2.1 U/mg protein with casein as substrate. All otherchemicals and solvents used were of analyticalgrade.

2.1. Three phase partitioning for extraction of oilfrom Jatropha curcas L.

Three phase partitioning was carried out as de-scribed earlier (Sharma et al., 2002a). Jatrophaseedswere cracked, the shells carefully removed and thekernels thus obtained were used for slurry prepara-tion. The slurry was prepared by grindingJatrophaseed kernels (5 g/30 ml) in distilled water. The pHof the slurry was adjusted to the desired value with0.1 N NaOH or 0.1 N HCl. The appropriate amountof ammonium sulphate was added and vortexed gen-tly, followed by addition of appropriate amount oft-butanol. The slurry was then incubated at 25◦C for1 h for the three phase formation. The three phaseswere then separated by centrifugation at 2000× g for10 min. The upper organic layer was collected andevaporated on rotary evaporater (under reduced pres-sure at 50◦C, for 5 min) to obtain oil. The amountsof oil recovered were calculated as percentages oftotal oil present inJatrophaseed kernels. The latterwas determined by soxhlet extraction using hexaneas a solvent as per the standard AOAC procedure(Horowitz, 1984). The FT-IR of oil obtained by TPPwas done by taking oil (10�l) on KBr disc, thespectra was recorded on Nicolet–Protégé-460 spec-trometer.

Each extraction by three phase partitioning was runin duplicate and the yields were found to agree be-tween duplicates within 3%.

3. Results and discussion

The solvent extraction ofJatrophaseed kernels gavea yield of 44 g oil/100 gJatropha seed kernels.Jat-rophaoil is reportedly present in the range of 40–60 goil/100 gJatrophaseed kernels (Makkar et al., 1997).A value of 44 g oil/100 gJatrophaseed kernels wastaken as 100% recovery of oil when calculating theoil recovery by TPP.

The effect of varying ammonium sulphate duringTPP is shown inFig. 1. About 83% (w/w) oil could beobtained by using 30% (w/v) ammonium sulphate. Inmany other cases as well, where TPP has been used forseparation of proteins/enzymes generally ammoniumsulphate 30% (w/v) has given best results (Dennisonand Lovrein, 1997; Sharma and Gupta, 2001a). Inthe case of extraction of soybean oil by TPP also,30% (w/v) ammonium sulphate was found to be op-timum (Sharma et al., 2002a). Fig. 2 shows the effectof varying the ratio of volumes oft-butanol toJat-rophaslurry. Decreasing thet-butanol volume by halfdecreases the oil yield whereas increasingt-butanolto twice the volume did not lead to any significantincrease. Thus, 1:1 ratio was employed in furtherstudies.

It is well established that oil bodies in plantsare trapped in the meshwork of proteins and cel-lulose/hemicellulose structures. Thus, the treatmentwith enzymes is an integral part of an approach calledaqueous enzymatic oil extraction (AEOE) (Rosenthalet al., 1996). AEOE has been successfully used in ex-traction of oils from rice bran (Sharma et al., 2001b)and peanut (Sharma et al., 2002b). Thus, it was con-sidered worthwhile to attempt pretreatment of theJatrophaslurry by enzymes before carrying out TPP.In order to distinguish it from other approaches, wewould like to call this approach ‘enzyme assistedthree phase partitioning (EATPP)’.

In an earlier case (Sharma et al., 2002b), a com-mercial preparation called ProtizymeTM had givengood results with AEOE. ProtizymeTM is a mixtureof enzymes, its three main components are pro-teases with different pH optimum range of 3–4, 5–7and 7–10. Hence, treatment ofJatrophaslurry withProtizymeTM before carrying out TPP was attempted(Fig. 3). Extraction of oil was carried out at three dif-ferent pH values: 4.0, 7.0 and 9.0, since the enzymepreparation has three proteases with these pH optima.

S. Shah et al. / Industrial Crops and Products 20 (2004) 275–279 277

Fig. 1. Effect of varying amount of ammonium sulphate in TPP on oil extraction. Slurry was prepared by dispersing groundJatrophaseeds (5 g/30 ml distilled water, pH 7.0). Varying amount of ammonium sulphate (%, w/v) was added to this slurry followed by additionof t-butanol (30 ml). The oil was recovered from uppert-butanol layer by following the procedure described in theSection 2.

By enzyme pretreatment at pH 9.0 followed by TPP,92% oil could be obtained (Fig. 3). It has been re-cently shown that the particle size of plant materialplays an important role in giving optimum oil yieldby AEOE. Smaller particle size was found to give

Fig. 2. Effect of varying the ratio oft-butanol to theJatrophaslurry in TPP on oil extraction. Ammonium sulphate (30%, w/v) was addedto the slurry and vortexed gently followed by addition oft-butanol to achieve different volumetric ratios of slurry tot-butanol. The oilwas recovered as described in theSection 2.

higher yields (Rosenthal et al., 2001). Sonication isan efficient way for breaking up bigger particles intosmaller ones (Szentmihalyi et al., 2002). It was foundthat using sonication before enzyme pretreatmentincreased the oil yield upto 97% (Fig. 4).

278 S. Shah et al. / Industrial Crops and Products 20 (2004) 275–279

Fig. 3. Effect of enzyme pretreatment in TPP on oil extraction. The pH of the slurry was adjusted to 4.0, 7.0 and 9.0 before addingProtizymeTM (250 mg). The slurry was incubated at 50◦C with constant stirring at 100 rpm for 1 h. Thereafter, ammonium sulphate (30%,w/v) and t-butanol in a ratio of 1:1 (v/v) were added to the slurry. The oil was recovered from uppert-butanol layer as described inthe Section 2. A control was also carried out by omitting the addition of enzymes. Control (dotted), enzyme treated TPP (verticallyhatched).

82

9297

0

20

40

60

80

100

TPP EA-TPP Sonication+EATPP

Process

Oil

Yie

ld (

%,w

/w)

Fig. 4. Effect of sonication in TPP on oil extraction. The pH of the slurry was adjusted to 4.0, 7.0 and 9.0 followed by sonication for 5 min.After addition of 250 mg ProtizymeTM, the slurry was incubated at 50◦C with constant stirring at 100 rpm for 1 h. After this pretreatment,ammonium sulphate (30%, w/v) andt-butanol in the ratio of 1:1 (v/v) (ratio of slurry tot-butanol) were added. The oil was recoveredfrom upper t-butanol layer as described inSection 2. (TPP, three phase partitioning; EA-TPP, TPP in presence of enzyme; sonication+ EA-TPP; sonication for 5 min followed by TPP in presence of enzymes).

The oil obtained was further analysed by FT-IRspectroscopy to detect the presence oft-butanol in theoil. The characteristics hydroxyl peak at 3368 cm−1

was absent in the IR spectra ofJatropha oil, con-firming the absence oft-butanol. The detection limitof FT-IR (as determined with standard solutions oft-butanol) was found to be 1%. Besides, no smell oft-butanol was found in the oil obtained by three phase

partitioning (and removal oft-butanol after evapora-tion at reduced pressure).

4. Conclusion

The novel approach of combining a recently de-veloped technique of TPP with enzyme pretreatment

S. Shah et al. / Industrial Crops and Products 20 (2004) 275–279 279

and sonication constitute an efficient procedure forobtaining oil from Jatropha seed kernels. It shouldbe added that unlike the soxhlet extraction, which iscarried out for 24 h, the method developed here takesonly about 2 h.

Acknowledgements

This work was partially supported by project fundsfrom Technology Mission on Oilseeds, Pulses andMaize (TMOP&M), Council for Scientific and In-dustrial Research (CSIR) and Department of Scienceand Technology (DST), Department of Biotechnology(DBT), all Government of India organisations.

References

Dennison, C., Lovrein, R., 1997. Three phase partitioning:concentration and purification of proteins. Protein Exp. Purif.11, 149–161.

Foidl, N., Foidl, G., Sanchez, M., Mittelbach, M., Hackel, S.,1996. Jatropha curcasL. as a source for the productionof biofuel in Nicaragua. Bioresour. Technol. 58, 77–82.

Gubitz, G.M., Mittelbach, M., Trabi, M., 1998. Exploitation of thetropical oil seed plantJatropha curcasL. Bioresour. Technol.67, 73–82.

Haas, W., Mittelbach, M., 2000. Detoxification experiments withthe seed oil fromJatropha curcasL. Ind. Crops Prod. 12,111–118.

Horowitz, W., 1984. Methods of Analysis of the Association ofOfficial Analytical Chemists. Washington, DC.

Makkar, H.P.S., Becker, K., Sporen, F., Wink, M., 1997. Studieson nutritive potential and toxic constituents of differentprovenances ofJatropha curcasL. J. Agric. Food Chem. 45,3152–3157.

Openshaw, K., 2000. A review ofJatropha curcas: an oil plant ofunfulfilled promise. Biomass Bioenerg. 19, 1–15.

Rosenthal, A., Pyle, D.L., Niranjan, K., 1996. Aqueous andenzymatic processes for edible oil extraction. Enzyme Microb.Technol. 19, 400–420.

Rosenthal, A., Pyle, D.L., Niranjan, K., Gilmour, S., Trinca, L.,2001. Combined effect of operational variables and enzymeactivity in aqueous enzymatic extraction of oil and proteinfrom soybean. Enzyme Microb. Technol. 28, 499–509.

Sharma, A., Gupta, M.N., 2001a. Purification of pectinases bythree phase partitioning. Biotechnol. Lett. 23, 1625–1627.

Sharma, A., Khare, S.K., Gupta, M.N., 2001b. Enzyme assistedaqueous extraction of rice bran oil. J. Am. Oil Chem. Soc. 78,949–951.

Sharma, A., Khare, S.K., Gupta, M.N., 2002a. Three phasepartitioning for extraction of oil from soybean. Bioresour.Technol. 85, 327–329.

Sharma, A., Khare, S.K., Gupta, M.N., 2002b. Enzyme assistedaqueous extraction of peanut oil. J. Am. Oil Chem. Soc. 79,215–218.

Szentmihalyi, K., Vinkler, P., Lakatos, B., Vendel, I., Then, M.,2002. Rose hip (Rosa caninaL) oil obtained from waste hipseeds by different extraction methods. Bioresour. Technol. 82,195–201.