Separation and Purification Technology 60 (2008) 272–277
Sterilization and extraction of palm oil from screw pressedpalm fruit fiber using supercritical carbon dioxide
N.A. Nik Norulaini a, Anees Ahmad b,∗, Fatehah Mohd. Omar b,Adel Ashur S. Banana b, I.S. Md. Zaidul c, Mohd. Omar Ab. Kadir b
a School of Distance Education, Universiti Sains Malaysia, 11800 Penang, Malaysiab School of Industrial Technology, Universiti Sains Malaysia, 11800 Penang, Malaysia
c National Agriculture and Bio-oriented Research Organization, National Agricultural Research Center for Hokkaido Region 1,Hitsujigaoka, Sapporo, Hokkaido 062-8555, Japan
Received 20 August 2007; received in revised form 30 August 2007; accepted 30 August 2007
bstract
The supercritical carbon dioxide (SC-CO2) was successfully used in the complete sterilization as well as simultaneous extraction of oil fromcrew pressed palm fruit fiber. The studies were conducted at different temperatures (40, 50, 70 ◦C) and pressures (13.7, 20.7 MPa) for 60 minf extraction period. The bacteria, gram negative (Bacillus), present in the sample was completely killed at 20.7 MPa and 50 ◦C. Palmitic andleic acid were found to be the major fatty acids in extracted oil. More saturated fatty acid were extracted at 50 ◦C and lower operating pressure
13.7 MPa). The unsaturated components, such as linoleic and oleic acids were extracted at higher pressures of 27.6 and 34.5 MPa, respectively.he fatty acids composition of the extracted oil analyzed using gas chromatography–mass spectrometry (GC/MS) includes caprylic, capric, lauric,yristic, palmitic, margaric, stearic, oleic, linoleic, linolenic, arachidic and gadoleic acids. 2007 Elsevier B.V. All rights reserved.
The palm oil extraction from palm fruit is traditionally doney screw press technology that is preceded by steam steriliza-ion of the fruits to ease the extraction. There is a consumptionf large quantity of water, in the processing of palm oil, whichoes to waste as palm mill oil effluent (POME) and is a big haz-rd for the environment. This process still leaves behind about% residual oil per fruit in the fruit fiber, which is wasted andhe cumulative amount of oil loss can be quite significant. Anlternative processing method is supercritical fluid extractionSFE), which inadvertently acts as sterilizer as well as extrac-or with the minimum generation of wastewater and efficiently
emoves the remaining oil from the fiber. The most commonlysed solvent for SFE is carbon dioxide (CO2), which has a criti-al temperature of 31.1 ◦C, and critical pressure 7.28 MPa. These
arameters make it an ideal solvent for extracting thermally sen-itive materials. Carbon dioxide is also the preferred fluid in foodpplications due to its non-toxic, non-flammable, inexpensivend environmentally acceptable nature [1,2].
In a previous work by Spilimbergo et al. [3] the supercriticalarbon dioxide (SC-CO2) was used for microbial inactivationf foods and the implementation of the innovative technique inhe applicability of the sterilization of thermally and pneumat-cally sensitive materials. They used temperatures of 36–75 ◦Cnd pressures from 70 to 150 bars and achieved complete bac-erial spore inactivation after SC-CO2 treatment for 2 h at5 ◦C. Thus SC-CO2 had played a key role in the inactiva-ion of microorganisms at low temperature and reduced time ofxposure.
Many researchers have investigated the use of SC-CO2 forterilization of bacteria. Ishikawa et al. [4] studied the inacti-ation of enzymes and extraction of lipophilic substances using
C-CO2. Thus the supercritical CO2 has received much attentionor its utilization in many industrial applications. The inactiva-ion of bacteria is probably due to implosion of the cell wall ofacteria at high pressure where the transfer rate of CO2 into the
ells is high. The effectiveness could be improved by optimizinghe pressure and temperature as well as time of extraction.
This study looks into the effect of various combinations ofressure and temperature of SC-CO2 extraction process on thextractability of fatty acid compositions of palm oil from screwrocess as well as on the inactivation of bacteria in the extractedil.
. Experimental
.1. Sample preparation
Palm fruit fiber was supplied by MALPOM Industrieserhad, a palm oil mill, Jalan Changkat, Nibong Tebal, Penang,alaysia. The palm fruits have been screw pressed to extract
he oil. The fruit fiber was collected at the vibrating screen oper-ting between the oil sutter and clarification tank. The oil palmxtraction process has been described by Nik Norulaini et al.5].
The experimental setup (Fig. 1) for the supercritical fluidxtraction was done as described by Nik Norulaini et al. [6].t consists of a pump (American Lewa Holigtic, MA, USA)ith a maximum capacity of 68.9 MPa, an oven (S.C.S.I.
nstrument system), chiller (Yihder, B/L-730), a 50 cm extrac-ion cell (keystone) with 13 mm diameter and 320 mm heightnd a wet gas meter (WNK-1A, Sinagawa Corp., Tokyo,apan).
Commercial liquefied carbon dioxide gas was purchasedocally from Malaysian Oxygen (MOX) Penang, in a gas cylin-er at temperature below −5 ◦C. Liquid carbon dioxide wasumped into the heated extraction cell loaded with approxi-ately 12 g of palm fruit fiber (wet basis). Temperatures of 40,
◦
0 and 70 C and pressures 13.7 and 20.7 MPa were used toxtract oil from palm fruit fiber for 60 min. Earlier studies [6,7]ndicate that one extraction is sufficient to extract almost all theatty acid components from palm fruit fiber.
Fig. 1. Experimental setup of the supercritical fluid extraction.
Hlwa1wDgta
3
ctw5smC
fication Technology 60 (2008) 272–277 273
.3. Bacteria culture
A sample of the fruit fiber (before and after supercritical fluidxtraction) was streaked across the 15% (w/v) nutrient agar andncubated for 12 h at 37 ◦C.
A small amount of bacteria was taken from the culture plateith a sterile swab and spread in a circular motion onto a micro-
cope slide. The slide was heated for approximately 30 s over aunsen burner and stained with crystal violet. Excess dye wasemoved by rinsing the slide with distilled water. Gram iodineas then applied to the slide and again rinsed with distilledater. The slide was decolorized with ethanol for 30 s and rinsedith distilled water. Finally, safranin was applied to the slide and
llowed to stain for 30 s before finally rinsing with distilled waternd dried with a blotting paper. The slide was observed underil immersion lens of microscope at 100× magnification [8].
.4. Determination of fatty acid methyl ester by gashromatography–mass spectrometry (GC–MS)
The preparation of fatty acid methyl esters was carried outccording to PORIM specifications p3.4 (PORIM Test Methods,995) as described by Hassan et al. [9].
.5. Gas chromatograph–mass spectrometer setup
The SC-CO2 extracts were dissolved in hexane (T.J. Bakernc., Philipsburg, NJ, USA); centrifuged at 4000 rpm for 10 minLabofuge 200, Heraeus Inst., Germany) and characterizedhromatographically using a GC model 5890 Series II andP5989A mass selective detector (Hewlett Packard, Palo Alto,A, USA) and Chem-station data system. Electron impact-MSf the extracted components was performed at electron energyf 70 eV with a source temperature of 200 ◦C and a scan rangef 20–550 amu at a rate of 0.81 scan per second. The columnsed was a 30 m × 0.25 mm × 0.25 �m (Quadrex Corp., Neweaven, USA) cross-linked methyl siloxane fused silica capil-
ary column maintained at 100 ◦C for 5 min. The temperatureas then raised gradually to 270 ◦C at the rate of 10 ◦C min−1
nd kept at that point for 40 min. Helium at a flow rate ofml min−1 was used as a carrier gas. The split less injectoras kept at 250 ◦C. The library used was HP 59943B WILEYatabase (PBM) format. The samples were prepared as per theeneral guide lines given in PORIM for the qualitative and quan-itative determination of fatty acids in oil samples as their fattycid methyl ester (FAME) through gas chromatography.
. Results and discussion
The SC-CO2 has a strong extraction capability on organicompounds from the solid matrix. In this study, the steriliza-ion effect of supercritical carbon dioxide on palm fruit fiberas tested at pressures 13.7 and 20.7 MPa and temperatures 40,
0, and 70 ◦C for 60 min time of extraction. The effect of pres-urized CO2 on the microorganisms has been investigated byany researchers. Cuq et al. [10] discovered that the pressurizedO2 enables inactivation of enzymes, while minimizing thermal
Fig. 2. Bacteria colonies from unsterilized oil palm fruit fiber.
egradation of sensitive and reactive substances. Enomoto et al.11] suggested that the microorganisms’ death might be highlyorrelated with gas absorption by the cells.
Thus both pressure and temperature are major operating fac-ors affecting the bacterial inactivation efficiency of pressurizedO2. Principally, bacterial inactivation by CO2 under pressure is
hought to be dependent on penetration of CO2 into cells causedy an enhanced transfer rate [12]. The inactivation rates are alsoffected by various other factors including the water content,onstituents and ages of microorganisms. Fig. 2 demonstratesacteria culture from unsterilized palm fruit fiber. The culture isense due to the presence of large number of bacterial colonies.
taining established the bacteria to be gram negative. Fig. 3hows fewer colonies of bacteria from palm fruit fiber that hasndergone SC-CO2 treatment at 40 ◦C and 13.7 MPa for 60 min.nactivation was also achieved at constant pressure, 13.7 MPa,
ig. 3. Reduced number of bacteria colonies from oil palm fruit fiber from SC-O2 treatment at temperature and pressure 40 ◦C and 13.7 MPa, respectively,
or 60 min.
ts
p
Fc
rification Technology 60 (2008) 272–277
nd increased temperature from 40 to 70 ◦C. The increase inemperature stimulates the diffusivity of CO2 into the bacterialells and this inactivates the bacteria. However, too high tem-eratures may decrease the CO2 density in suspending medium13]. Temperature has a primary role in the diffusivity of CO2nd the sterilization effect which is much more evident at 70 ◦C.t this stage, CO2 can easily penetrate through the membrane
ausing increase in fluidity and permeability and destroying itsssential domains. The results obtained at 40 ◦C clearly indicatehat CO2 has a fundamental role in bacterial inactivation.
On the other hand, when the treatment of the fiber wasone at a higher pressure of 20.7 MPa while temperature andime remained constant, complete inactivation of bacteria wasbserved as demonstrated by the absence of colonies in Fig. 4.his indicates that pressure has an influence over the inactivationf the bacteria. The presence of pressurized CO2 within the fibernhances the lethal thermal effect on bacteria [10]. Results showhat at various temperatures and at 13.7 MPa the presence of theacteria is minimal and do not vary at different temperatures.hen the temperature was increased the number of colonies was
educed. Fig. 4 shows that at 20.7 MPa and 40 ◦C the bacterialolonies were totally absent which indicates that this combi-ation proved to be the optimum in sterilizing palm oil fiber.acterial inactivation was not completely achieved at 13.7 MPand 70 ◦C. However, when the pressure was increased, there wasn enhanced sterilization effect on the fiber.
The effectiveness of SC-CO2 depends on the combined effectf temperature, pressure and time of exposure. Thus we canonclude that by coupling the action of these three parame-ers under supercritical conditions can sterilize completely the
aterial. Spilimbergo et al. [3] used supercritical carbon dioxidereatment with pressure up to 120 bar at 50 ◦C and discovered
hat, under these conditions, the viability of the bacteria was notignificantly affected.
Palm fruit fiber is the by-product from the conventional screwress oil extraction process, which contains residual oil. The
ig. 4. No bacteria colony from palm fruit fiber treated under supercriticalonditions of 20.7 MPa and 40 ◦C for 60 min.
N.A. Nik Norulaini et al. / Separation and Purification Technology 60 (2008) 272–277 275
F ng SCc
qsTbabeat
oc
narO2ttt
Ft
ig. 5. Fatty acid composition of oil extracted from palm fruit fiber residue usirude palm oil and palm oil extracted using soxhlet.
uantity of this oil extracted using SC-CO2 under various pres-ures ranging from 13.7 to 34.5 MPa is shown in Figs. 5 and 6.he oil extracted over 1 h period, shows very little differenceetween the fatty acid compositions during first and second halfn hour of total extraction period. Since most of the oil haveeen extracted using the screw press, much of the oil pores arempty which makes the cells more porous. The solvent perme-bility is enhanced within the entire porous structure throughout
he SC-CO2 extraction period.
From Figs. 5 and 6, it is evident that there is low amountf C10, C12 and C14 present in the oil extracted. It has beenonsistently shown that palmitic acid; C16 is the major compo-
ftsc
ig. 6. Fatty acid composition of oil extracted from palm fruit fiber residue using SChe crude palm oil and palm oil extracted using soxhlet.
-CO2 at 50 ◦C under different pressures for the first 30 min (t = 0–30 min), the
ent, followed by oleic acid, which is monounsaturated. Palmiticcid reflects higher yield at lower pressure of 13.7 MPa, whileemain relatively the same under the remaining higher pressures.n the other hand, a large amount of oleic acid is extracted at7.6 MPa. Too high or low pressures appear not to be favorableo oleic acid removal from the matrix. However, C18:2 seemso require a pressure of 34.5 MPa, the highest pressure tested inhis study to get the greatest yield, while lower pressures per-
ormed consistently similar. The reason for the variation withinhe extracted product of SC-CO2 extraction is the change in theolubility of different fatty acids with the change in extractiononditions. The process of extraction may be described as one in
-CO2 at 50 ◦C under different pressures for the second 30 min (t = 30–60 min),
2 nd Purification Technology 60 (2008) 272–277
wta
uaatspssotTcstleodvm
ceppsIcratCe
Fab
Fab
tC
eCSp
ccCh
76 N.A. Nik Norulaini et al. / Separation a
hich material is simultaneously desorbed and diffuses throughhe solvent in the micropores to the external stagnant fluid layernd then to the bulk fluid.
By increasing the pressure at constant temperature, the sol-bility of longer chain fatty acids (C18:0, C18:1, C18:3, C20:0nd C20:1) is increased relative to short and medium chain fattycids (C8, C10, C12, C14). Zaidul et al. [14] discovered that ashe temperature and pressure of the extraction increased, theolubility of the longer chain (C16–C18:2) fatty acids com-osition is also increased. They further determined that theolubility of shorter and medium chain (C8–C14) fatty acid con-tituents decreased simultaneously. The increase in the amountf long chain fatty acid present in the extract may be due tohe increased solubility influenced by the extraction conditions.hus, the shorter chain fatty acids have higher extractabilityapacity in any condition of SC-CO2 extraction, which can beeen in the decreasing trend of yield from fraction 1 to frac-ion 2. Rozzi et al. [15] analyzed the fatty acid composition ofycopene from tomato extracted through SC-CO2 under differ-nt conditions (32, 50, and 68 ◦C and 13.7, 27.5 MPa at flow ratef 2.5 ml min−1). The fatty acid composition of the extracts pro-uced by supercritical fluid extraction (SFE) at low pressuresaried marginally from the extracts obtained through soxhletethod.The results show that there is no difference in the percent
omposition of the fatty acids in the palm oils, which werextracted with petroleum ether when compared to that in crudealm oil (CPO). The fatty acids composition of the extractedalm oil by SC-CO2 under different extraction conditions alsohowed no selectivity differences in the extraction of fatty acids.t can be seen that the concentration of the lighter (smallerarbon number) fatty acids such as C12, C14 and C16 (all satu-ated fatty acids) present in the oil show no significant changes
s the extraction process continues. Likewise, the concentra-ions of unsaturated fatty acids collected such as C18:1, C18:2,18:3, C20:0 and C20:1 did not reflect much variation as thextraction proceeded. Similar behavior was also observed for
ig. 7. Typical chromatogram of the first 30 min extraction of oil palm fruit fibert 13.7 MPa and constant temperature of 50 ◦C, identifying fatty acids profiley comparing the elution times of standard using SC-CO2.
GaiC
4
nteAiHsim
f((afip
ig. 8. Typical chromatogram of the first 30 min extraction of oil palm fruit fibert 34.5 MPa and constant temperature of 50 ◦C, identifying fatty acids profiley comparing the elution times of standard using SC-CO2.
he extraction and fractionation of crude palm oil using SC-O2.
Figs. 5 and 6 show little variation in the percent of fatty acidsxtracted (C8, C10, C12, C14, C15, C16, C17, C 18:0, C18:1,18:2, C18:3, C20:0 and C20:1) using soxhlet technique andC-CO2 for the first and second fractions, respectively at variousressures and constant temperature of 50 ◦C.
Fig. 7 depicts the chromatogram of GC–MS for the fractionollected during first 30 min using SC-CO2 at 13.7 MPa andonstant temperature of 50 ◦C. The fatty acids methyl esters of10, C12, C14, C16, and C18:0, C18:1, C18:2 and C18:3 acidsave been detected. Fig. 8 shows the typical chromatogram ofC–MS of the first fraction extracted for 30 min using SC-CO2
t 34.5 MPa and at 50 ◦C. When the extraction pressure wasncreased to 34.5 MPa, the fatty acids (C16, C18:1, C18:2 and18:3) in the first fraction increased concurrently.
. Conclusion
Supercritical carbon dioxide extraction is a promising tech-ique that can simultaneously be used for the sterilization ofhe material. At a pressure of 13.7 MPa, bacteria are sensitivenough for the variation of temperature (from 40 to 70 ◦C).ctually higher is the temperature; the greater is the diffusiv-
ty of the CO2 into the bacterial cells that can affect the bacteria.igher bacterial inactivation was observed at 20.7 MPa pres-
ure and 40 ◦C temperature. The penetration of pressurized CO2nto the bacterial cells plays a role in the inactivation of theseicroorganisms.The palm oil fractionated by SC-CO2 contains various
C18:1), linoleic (C18:2), linolenic (C18:3), arachidic (C20:0)nd gadoleic (C20:1) acids. The major constituent of palm fruitber is palmitic acid (C16), which is best extracted at lowerressure of 13.7 MPa and extraction at higher pressures did not
Puri
ri
A
aa
R
[
[
[
[
[
N.A. Nik Norulaini et al. / Separation and
eveal much difference. Oleic acid (C18:1) can be best extractedn the pressure range of 24.1–27.6 MPa.
cknowledgements
The authors gratefully acknowledge the financial assistancend the research facilities from Ministry of Science, Technologynd Innovation (MOSTI) and Universiti Sains Malaysia.
eferences
[1] P. Ballestra, A.A. Silva, J.L. Cuq, Inactivation of Escherichia coli by carbondioxide under pressure, J. Food Sci. 61 (1996) 829–831.
[2] M. Shimoda, H. Ishikawa, T. Kawano, Y. Osajoma, Extraction of volatilecompounds from aqueous solution using microbubble, gaseous, supercrit-ical and liquid carbon dioxide, J. Food Sci. 59 (1997) 231–233.
[3] S. Spilimbergo, T. Parton, N. Elvassore, A. Bertucco, Microbial inactivationby high-pressure, J. Supercrit. Fluids 22 (2002) 55–63.
[4] H. Ishikawa, M. Shimoda, K. Tamaya, A. Yonekura, T. Kawano, Y.Osajima, Inactivation of Bacillus spores by supercritical carbon dioxidemicrobubble method, Biosci. Biotechnol. Biochem. 61 (1997) 1022–1026.
[5] N.A. Nik Norulaini, H. Nizar H, A. Omar, I.I. Muhammad Hakimi, A.K.
Mohd Omar, Dehulling and its effect on supercritical extraction of palmkernel oil, J. Chem. Eng. Jpn. 34 (2001) 407–410.
[6] N.A. Nik Norulaini, I.S. Md Zaidul, A.K. Mohd Omar, Supercritical reduc-tion of lauric acid in palm kernel oil (PKO) to produce cocoa butterequivalent (CBE) fat, J. Chem. Eng. Jpn. 37 (2004) 194–203.
[
fication Technology 60 (2008) 272–277 277
[7] N.A. Nik Norulaini, I.S. Md Zaidul, O. Anuar, A.K. Mohd Omar, Super-critical enhancement for separation of lauric acid and oleic acid in palmkernel oil (PKO), Sep. Purif. Technol. 39 (2004) 133–138.
[8] W.F. Harrigan, M.E. McCance, Department of Bacteriology, TheWest of Scotland Agricultural College Auchincruive, Ayr, Scotland,1966.
[9] M.N. Hassan, N.N. Ab. Rahman, B.O. Anuar, M.H. Ibrahim, A.K. MohdOmar, Simple fractionation through the supercritical carbon dioxide extrac-tion of palm kernel oil, Sep. Purif. Technol. 19 (2000) 113–120.
10] J.L. Cuq, H. Roussel, D. Viver, J.P. Caron, Effects of gases under pressureand influence on the thermal inactivation of microorganisms, Sci. Aliments13 (1993) 677–698.
11] A. Enomoto, K. Nakamura, M. Hakoda, N. Amaya, Lethal effect of high-pressure carbon dioxide on a bacterial spore, J. Ferment. Bioeng. 83 (1997)305–307.
12] M. Shimoda, Y. Yamamoto, J. Cocunubo-Castellanos, H. Tonoike, T.Kawano, H. Ishikawa, Y. Osajoma, Antimicrobial effects of pressured car-bon dioxide in a continuous flow system, J. Food Sci. 63 (1998) 709–712.
13] A.K. Dillow, F. Dehghani, J.S. Harkack, N.R. Foster, R. Langer, Bacterialinactivation by using near and supercritical carbon dioxide, Proc. Natl.Acad. Sci. U.S.A. 96 (1999) 10344–10348.
14] I.S.M. Zaidul, N.A. Nik Norulaini, A.K. Mohd Omar, R.L. Smith Jr.,Supercritical carbon dioxide (SC-CO2) extraction and fractionation of palm
kernel oil from palm kernel as cocoa butter replacer blend, J. Food Eng. 73(2006) 210–216.
15] N.L. Rozzi, R.K. Singh, R.A. Vierling, B.A. Watkins, Supercritical extrac-tion of lycopene from tomato processing by products, J. Agric. Food Chem.50 (2002) 2638–2643.
