Cocoa Butter Extraction from Cocoa Nibs Using Supercritical Carbon Dioxide

Cocoa beans consist mainly of cocoa butter (50-55% w/w). High quality cocoa butter used in food, cosmetic and pharmaceutical products is obtained by mechanical press, expeller, and solvent extraction using hexane. Supercritical fluid extraction (SFE) using carbon dioxide as a solvent has provided an excellent alternative to the use of chemical solvent in the extraction of cocoa butter from different plant matrices. In comparison with established methods, SFE has some important advantages, particularly in its ability to yield products that are completely free from processing residues. SFE is also an alternative from the standpoint of time-saving and for environmental reasons such as the reduction of the use of large solvent volumes. Carbon dioxide (CO2) is an ideal solvent for extraction of natural products because it is nontoxic, odorless, tasteless, non explosive, readily available, and easily removable from the products. This work is divided into several part namely; 1) the effect of flow rate with different pressure on cocoa butter extraction using supercritical carbon dioxide (SC-CO2), 2) a study on the sample matrix involving: the effect of particle size, degree of fermentation and alkalization on cocoa butter extraction using supercritical fluid; 3) a study on the effect of moisture content, roasting time and roasting temperature on cocoa butter extraction using supercritical fluid, 4) a study on the effect of cosolvents with types and concentrations of polar cosolvents and non polar cosolvents in cocoa butter extraction using SC-CO2, 5) to evaluate the mass transfer parameters in cocoa butter extraction by SFE using the Sovova Lacks plug flow model (SLM) for the effect of flow rate with different pressure, the effects of polar and non polar cosolvents, and also using single sphere model (SSM) for the effects of particle size, degree of fermentation, pH alkalization, moisture content, roasting time and roasting temperature and 6) to determine the triacylglycerols composition and fatty acid methyl esters composition of cocoa butter extracted that resulted from extraction using SC-CO2 and SC-CO2 with cosolvents. The study on the effect of flow rate and pressure found that the optimum conditions for flow rate and pressure for cocoa butter extraction using SC-CO2 were at 2 ml/min and 35 MPa, r espectively. The highest yields were obtained from cocoa nibs sample with smaller particle size (S1= 0.07 mm) with 92% yield for almost 20 h using SCCO2, unfermented nibs (F1) with 100% yield for almost 10 h using ethanol 25% as cosolvent in SC-CO2 and roasting treatment of 150 oC and 35 min (R6) with 100% yield for almost 14 h using ethanol 25% (mol %) as cosolvent in SC-CO2. Increasing roasting time and temperature have resulted in the increase of the yield. Furthermore, the highest yield was produced from high pH (7.5 -7.9) of dark alkalized cocoa liquor (A1) with 73.70% yield for nearly 18 h extraction time using ethanol 25% as cosolvent in SC-CO2 and high moisture content (M5 =17.64%) with 60.73% after nearly 20 h extraction using SC-CO2. Statistically, the yield (p<0.05) significantly was affected by all the treatment, however no significant different was observed for both light alkalized cocoa liquor (pH = 6.8-7.2) and alkalized natural cocoa liquor (pH = 5.0 – 5.9), and both moisture content of 9.79 and 17.64%, respectively. Ethanol showed the best cosolvent on the yield in SC-CO2, followed by isopropanol, acetone, hexane and cyclohexane. The yield increased with an increase in concentration of cosolvents (25%>15%>5%), except for acetone in which 15% concentration was higher than 25% and 5%. Extraction of cocoa butter using SC-CO2 with polar cosolvents obtained a yield higher than non polar cosolvents. All treatment of cosolvents studied statistically significantly different at p<0.05. The two mathematical models are based on mass transfer were used. First model, the Sovova’s lack’s plug flow model (SLM) was used to describe the extraction process of effect of flow rate with different pressure and effect of polar and non polar cosolvents. The hardly accessible solute xk and the volume mass transfer coefficients in the fluid phase (F) and solid phase S were used as fitting parameters. The maximum average deviation between measured and calculated oil yield was 6.861%. Mass transfer coefficients in the fluid phase and solid phase varied between 6.528.E 06 to 1.498E-04 s-1 and between 5.185.E-06 to 9.144E-04 s-1, respectively. Second model, the single sphere model (SSM) was used to describe the extraction process of effect of particle size, degree of fermentation, pH of alkalization, moisture content, roasting time and roasting temperature. Adjusting of effective intraparticle diffusion coefficient (De) and estimation of parameter of coefficient of mass transfer in the fluid phase (kf) and overall mass transfer coefficient kp and parameters were evaluated. The result showed that De, kf and kp varied between 8.900E-16 to 1.850E- 09 m2/s, between 1.139E-10 to 2.115E-04 m/s and between 9.581E-12 to 8.992E-05 m/s, respectively. All experimental data results of extraction curves were fitted fairly well by using both of the SLM model and SSM model with average absolute relative deviation (AARD) between measured and calculated oil yield were maximum of 6.861. The results showed that cocoa butter extracted from various treatments had three main triacylglycerol (TG) namely Glycerol-1,3-Dipalmitate-2-Oleate (POP), Glycerol-1-Palmitate-2-Oleate-3-Stearate (POS), and Glycerol-1,3-Distearate-2- Oleate (SOS). In general, POS was the major component with 42.16 to 45.78% followed by SOS and POP with 27.60 to 31.67% and 20.09 to 22.79%, respectively. Furthermore, analysis of fatty acid (FA) showed that palmitic acid (C16:0), stearic acid (C18:0) and oleic acid (C18:1) were the three main fatty acids in the cocoa butter extracted with stearic acid being the major component with 34.82 to 39.06%, followed by oleic acid and palmitic acid with 28.48 to 31.72% and 28.27 to 31.33%, respectively. Statistically, there are differences the effects of all treatment on TG and FA compositions.

Preview PDF FSTM_2009_18_abstract.pdf Download (250kB)

Actions (login required)

View Item