Journal of Scientific & Industrial Research Vol. 77, February 2018, pp. 131-135

Extraction and Characterization of Oil from Ethiopian Mangifera Indica Seed Kernels

M Kemal1, S A Jabasingh1,*, A Yimam1 and J A Kumar2 1Process Engineering Division, School of Chemical and Bio Engineering, Addis Ababa Institute of Technology, Addis Ababa University, Ethiopia 2Department of Chemical Engineering, Sathyabama Institute of Science and Technology, Chennai, Tamil Nadu, India

Received 03 October 2016; revised 17 July 2017; accepted 24 December 2017

High population growth demands alternative sources for producing valuable products from waste materials. One such waste material in Ethiopia is mango seed kernel and the oil extracted from the mango seed kernel, have immense application in the cosmetics industry. The main aim of this research is to determine the optimum operating condition for the extraction of oil from the mango seed kernel.

Keywords: Mango Seed Kernel, Oil, Cosmetic Industry, Extraction, Yield

Introduction The moisture content of the kernel was determined by Natural oils are excellent emollients derived from a measuring the weight of the sample before drying and variety of plants1,2. Mango seed kernel oil has been after drying in an oven at 105°C respectively. used as a key ingredient in the cosmetics, as it is a good Mango oil extraction 3,4 source of phenolic compounds . In the context of The experimental work was conducted using soxhlet Ethiopia, 126,800 qt of mango are produced in the extractor in triplicate with three different solvents: Arba Minch and Zuria Woreda with a total area hexane, petroleum ether and ethanol1,2. 75 g mango coverage of 634 hectares. Mango farmsteads in Asossa seed kernel at three different particle sizes (3-1.5mm, produce an average of 13,500 mangoes per farmstead. 1.5-0.5mm and 0.5-0.25mm) were extracted using Though mango seed kernels, generated by mango Soxhlet extracter with 300 mL solvent with varying consumption find some few applications, production of treatment times (2 h, 4 h and 6 h). The mango seed oil from the mango seed kernel could be resourceful5,6. kernel was found to have an oil content of 8-16%, hence, an average 12% oil content was used for Materials and Methods calculating the percentage oil yield3. The percentage oil

Chemicals and processing conditions yield and extraction yield was calculated as follows. Asossa mango varieties were purchased from (Mass of oil) Kenuma and Akea juice processing house, Addis Percentage oil yield = × 100% Ababa. Hexane, petroleum ether, ethanol, sodium (0.12 × Mass of the sample) hydroxide, potassium hydroxide, hydrochloric acid, … (1) folin ciocalteu reagent, gallic acid, sodium carbonate, (Mass of oil) Percentage extraction yield = × 100% acetone, and phenolphthalein were procured from (Mass of the sample) Merck, Germany. The mangoes were peeled, deseeded 7 … (2) and sun dried for five days . The hardcover of the seed was decorticated manually and the seed kernel was Physical and chemical characterization dried in the oven at 50°C for 18 h5,8. The dried mango The moisture content, volatile matter and seed kernel was milled in centrifugal miller with a specific gravity of the oil were determined by the 9 sieve size of 4 mm using a vibrating shaker for 15 min. standard procedures . Kinematic viscosity of the oil was measured using —————— *Author for Correspondence Vibro viscometer. The pH electrode was standardized E-mail: [email protected] with a buffer solution and immersed into 2 g of the 132 J SCI IND RES VOL 77 FEBRUARY 2018

oil sample and 13 mL of hot distilled water for the be 46.6, 46.9, 49.0, 47.6, 52.6, 51.0 and 51.4 % measurement of the pH11. The saponification value, the respectively. The mean plus the standard deviation of amount of unsaponifiable matter, acid value, total the seven samples were 49.3% ± 2.47. The oil phenolic content and percentage free fatty acid values extraction was carried out in a Soxhlet apparatus using are calculated using the standard methods10. The three different solvents. The oil yield and extraction peroxide value, iodine value and refractive index yield were calculated and summarized in Table 1. value of the oil sample was determined by the Maximum percentage oil yield was 84.81±1.7 standard procedures at JIJE Analytical testing service (equivalent to 10.18±0.2 percentage extraction yield) laboratory, Addis Ababa9. for the particle size range of 0.25-0.5 mm, extraction

Experimental design time of 6 h and hexane as the solvent. Minimum A regression model was established for the percentage oil yield was 18.89±1.11 (equivalent to dependent variables to fit the experimental data for 2.27±0.13 percentage extraction yield) for the particle the response using Design expert 7.0.0 software12,13. size range of 0.5-1.5 mm, extraction time 2 h and Factorial design is used to test the effect of each factor ethanol as solvent. Previous reports presented an (Table 1). In the factorial experiments, all the possible extraction yield of 8.46±0.1% (equivalent to 70.5±0.83 % combinations of the factor levels were tested14. oil yield) with hexane as a solvent for an extraction time of 6 h2. A reported yield of 10.2% (85% oil yield) Results and Discussion using hexane as solvent and with extraction time of 5 h 2 Outcome of process parameters on yield was reported earlier . The effect of particle size on oil The moisture content of the seed kernel with 11.6, yield for hexane as a solvent, is shown in Figure 1(a). 16.2, 10.5, 10.4, 9.5, 4.9, and 3.5 g was determined to For 2 h, 4 h and 6 h, as the particle size decreases, the

Table 1 — Yield of mango seed kernel oil using full factorial design Run Factors Oil yield (%) Extraction yield PS (mm) t (h) S R1 R2 R3 Mean±SD (%) 1 0.25-0.5 2 H 57.78 60.0 54.44 57.41±2.8 6.89±0.34 2 0.5-1.5 2 H 43.33 45.56 44.44 44.44±1.1 5.33± 0.13 3 1.5-3 2 H 42.22 40.0 44.44 42.22±2.2 5.07±0.27 4 0.25-0.5 4 H 82.22 83.33 81.11 82.22±1.1 9.87±0.13 5 0.5-1.5 4 H 58.89 57.78 58.89 58.52±0.6 7.02±0.08 6 1.5-3 4 H 45.56 43.33 5.73 45.57±2.2 5.47±0.27 7 0.25-0.5 6 H 84.44 86.67 83.33 84.81±1.7 10.18±0.2 8 0.5-1.5 6 H 68.89 70.0 71.11 70.0±1.11 8.4±0.13 9 1.5-3 6 H 64.44 64.44 63.33 64.07±0.6 7.69±0.08 10 0.25-0.5 2 E 23.33 22.22 25.56 23.70±1.7 2.84±0.20 11 0.5-1.5 2 E 21.11 20.0 20.0 20.37±0.6 2.44±0.08 12 1.5-3 2 E 17.78 20.0 18.89 18.89±1.1 2.27±0.13 13 0.25-0.5 4 E 48.89 47.78 50.0 48.89±1.1 5.87±0.13 14 0.5-1.5 4 E 40.0 42.22 41.11 41.11±1.1 4.93±0.13 15 1.5-3 4 E 30.0 30.0 31.11 30.37±0.6 3.64±0.08 16 0.25-0.5 6 E 56.67 55.57 58.89 57.04±1.6 6.84±0.20 17 0.5-1.5 6 E 46.67 46.67 47.78 47.04±0.6 5.64±0.08 18 1.5-3 6 E 43.33 44.44 47.78 45.18±2.3 5.42±0.28 19 0.25-0.5 2 PE 62.22 61.11 61.11 61.48±0.6 7.38±0.03 20 0.5-1.5 2 PE 38.89 40.0 36.67 38.52±1.7 4.62±0.20 21 1.5-3 2 PE 40.12 40.11 40.07 40.01±0.1 4.8±0.11 22 0.25-0.5 4 PE 73.33 75.56 73.33 74.07±1.2 8.89±0.15 23 0.5-1.5 4 PE 56.61 56.67 57.78 57.02±0.6 6.84±0.08 24 1.5-3 4 PE 44.44 42.22 45.57 44.08±1.7 5.29±0.20 25 0.25-0.5 6 PE 83.33 82.22 84.44 83.33±1.1 10.0±0.13 26 0.5-1.5 6 PE 68.89 71.11 67.78 69.26±1.7 8.31±0.20 27 1.5-3 6 PE 61.11 60.0 63.33 61.48±1.7 7.38±0.20 Note: *R-Replicate ;S-Solvent type; t-Time, h; PS-Particle size, mm;PE-Petroleum Ether; H-Hexane; E-Ethanol JABASINGH et al.: EXTN AND CHARACTERIZATION OF OIL FROM INDICA SEED 133

Fig. 1 — Effect of (a) Particle size, (b) Extraction time, (c) Solvent type on the Mango seed kernel oil yield (%) oil yield increases from 42.22% to 57.41%, 45.57% to minimum yield for the same operating conditions, 82.22% and 64.07% to 84.81% for the particle sizes in when petroleum ether was used as solvent was found the range of 0.25-0.50,0.50-1.50,1.50-3.0 respectively. 83.33% and 61.48%, respectively, while ethanol as However, when the particle size is too small or very solvent resulted a maximum yield of 57.04% and fine the oil yield decreases even though, the surface minimum yield of 23.70%. Hexane and petroleum area of contact is increased. The reason may be due to ether are non polar organic solvents that have a high agglomeration of fine particle which reduces the capacity to dissolve non polar compounds while contact surface area. Effect of extraction time on ethanol can extract non oil components due to the percentage oil yield is shown in Figure 1(b). The oil presence of the OH bond (polar)4. However, currently yield increased by 24.81% as the extraction time both hexane and petroleum ether have been identified increased from 2 h to 4 h and it increased by 2.59%, as as air pollutants and can react with other pollutants the time increased from 4 h to 6 h. Maximum oil yield to produce ozone and photochemical oxidants. is obtained at lower particle size. And the optimum Therefore, even if the yield obtained from ethanol was extraction time is 4 h, since 97% of the maximum yield the lowest, from the health and environmental point was obtained at this time. The effect of solvent type on of view, it is suggested for this extraction process15. percentage oil yield is shown in the Figure 1(c). The Figure 2 shows the production of mango seed kernel maximum and the minimum oil yield was 84.81% and oil. Using the process parameters that gave a 57.41% at extraction time of 6 h respectively, when maximum oil yield, the particle size range 0.25- hexane was used as a solvent. The maximum and 0.5mm, extraction time of 6h and hexane as solvent, 134 J SCI IND RES VOL 77 FEBRUARY 2018

of 85.01%, which agreed well with the experimental value 84.81%. A higher unsaponifiable matter of 3.85% gives the Ethiopian mango (Assosa variety) higher opportunity to be used in cosmetics industry for the production of liquid soap, shampoos and shaving creams.

Acknowledgment The authors are thankful to the Addis Ababa Institute of Technology, Addis Ababa University for the support. The comments and recommendations of the anonymous reviewers and the Principal Fig. 2 — Production of mango seed kernel oil from mango seeds Scientist and Editor Dr. Sanjay Sengupta are greatly the oil was extracted and characterized. The specific acknowledged. gravity of the oil was 0.905. The refractive index was obtained as 1.456. The pH and Kinematic viscosity References -5 2 were 6.07 and 5.4×10 m /s respectively. The 1 Kittiphoom S, Utilization of Mango seed, Int Food Res J, 19 saponification value was 184.66 mg KOH/g oil. The (2012)1325-1335. acid value and Iodine value was 2.39 mg KOH/g and 2 Kittiphoom S & Sutasinee S, Mango seed kernel oil and its physicochemical properties, Int Food Res J, 20 (2013) 40.44 g/100g oil respectively. The peroxide value in 1145-1149. milli equivalent oxygen per 100g oil was 2.92. The 3 Jafari J, Gharachorloo M, Hemmaci AH, The stabilizing effect phenolic content was 115.8 mg per g oil. The analysis of three varieties of crude mango seed kernel oil on tallow, of variance (ANOVA) shows a model F-value of J Food Biosci Technol, 4 (2014)31-36. 35.57 implying the significance of the model. Value 4 Fahimdanesh M & Bahrami ME, Evaluation of physicochemical properties of Iranian mango seed kernel oil, of prob>F less than 0.05 indicates the terms are 2nd Int Conf Nut Food Sci, IPCBEE, 53 (2013) 44-49. 16,17 significant . The regression equations in terms of 5 Dorta E, Gloria M L, Using drying treatments to stabilize the actual factors for the yield by using the three mango peel and seed: effect on antioxidant activity, Food Sci solvents are given below, Technol, 45 (2012) 261-268. 6 Lakshmi M, Swarnali D M, Usha R, Mango (Mangifera indica L.) Solvent type: Hexan kernel flour as a potential ingredient in the development of 푌푖푒푙푑 (%) = +52.75 − 35.37 ∗ 푃푎푟푡푖푐푙푒 푠푖푧푒 composite flour bread, Ind J Nat Prod Resour, 5 (2014)75-82. + 10.81 ∗ 푇푖푚푒 − 0.62 7 Ashoush I S, Gadallah M G E, Utilization of mango peels and seed kernels powders as sources of photochemicals in biscuit, ∗ 푃푎푟푡푖푐푙푒 푠푖푧푒 ∗ 푇푖푚푒 + 9.55 World J Dairy & Food Sci, 6 (2011) 35-42. ∗ 푃푎푟푡푖푐푙푒 푠푖푧푒 8 Ekorong F J A A, Zomegni G, Desobgo S C Z, Ndjouenkeu … (3) R, Optimization of drying parameters for mango seed kernels Solvent type: Ethanol using central composite design, Bioresour Bioprocess, 푌푖푒푙푑 (%) = +17.62 − 29.39 ∗ 푃푎푟푡푖푐푙푒 푠푖푧푒 2 (2015) 8-12. 9 Saiprabha M M & Goswami-Giri A S, Composition and + 11.77 ∗ 푇푖푚푒 − 0.63 characterization of refined oil compared with its crude oil ∗ 푃푎푟푡푖푐푙푒 푠푖푧푒 ∗ 푇푖푚푒 + 9.56 from waste obtained from mango (Mangifera indica), ∗ 푃푎푟푡푖푐푙푒 푠푖푧푒 Asian J Res Chem, 4(2011) 1415-1419. … (4) 10 AOAC, Saponification number of oil and fat, International Official Methods of Analysis of AOAC International, 17th edn, Solvent type: Petroleum ether Gaithersburg, MD, USA,1995. 푌푖푒푙푑 (%) = +50.82 − 35.41 ∗ 푃푎푟푡푖푐푙푒 푠푖푧푒 11 AOAC, pH value of oil and fat, International Official Methods + 10.75 ∗ 푇푖푚푒 − 0.61 of Analysis of AOAC International, 14th edn. Gaithersburg, ∗ 푃푎푟푖푐푙푒 푠푖푧푒 ∗ 푇푖푚푒 + 9.54 MD, USA, 2000. ∗ 푃푎푟푡푖푐푙푒 푠푖푧푒 12 Jabasingh S A & Nachiyar C V, Optimization of cellulase production by Aspergillus nidulans: application in the … (5) biosoftening of cotton fibers, World J Microbiol Biotechnol,

Conclusion 27(2010) 85-97. 13 Jabasingh S A & Nachiyar C V, Utilization of pretreated Parameter optimization shows that the 0.39 mm bagasse for the sustainable bioproduction of cellulase by particle size, 5.67 h extraction time and hexane as Aspergillus nidulans MTCC344 using Response surface solvent gives a maximum mango seed kernel oil yield methodology, Ind crops Prod, 34 (2011)1564-1571. JABASINGH et al.: EXTN AND CHARACTERIZATION OF OIL FROM INDICA SEED 135

14 Jabasingh S A & Nachiyar C V, Optimization of cellulase 16 Jabasingh S A, Optimization and Kinetics of cellulase synthesis by RSM and evaluation of ethanol production immobilization on modified chitin using Response from enzymatically hydrolyzed sugarcane bagasse using surface methodology, Ads Sci Technol, 29 (2011a) Saccharomyces cerevisiae, J Sci Ind Res, 71 (2012a) 353-359. 897-915. 15 Jabasingh S A, Lalith D, Prabhu M A, Yimam A & Zewdu T, 17 Jabasingh S A (2011) Utilization of pretreated coir Catalytic conversion of sugarcane bagasse to cellulosic pith for the optimized bioproduction of cellulase by ethanol: TiO2 coupled nanocellulose as an effective hydrolysis Aspergillus nidulans, Int Biodet Biodegrad, 65 (2011b) enhancers, Carbo Polym, 136 (2016) 700-709. 1150-1160.