Comparison of subcritical CO2 and ultrasound-assisted aqueous methods with the conventional solvent method in the extraction of avocado oil

The Journal of Supercritical Fluids 135 (2018) 45–51

Contents lists available at ScienceDirect

The Journal of Supercritical Fluids journal homepage: www.elsevier.com/locate/supflu

Comparison of subcritical CO2 and ultrasound-assisted aqueous methods with the conventional solvent method in the extraction of avocado oil Chin Xuan Tana, Gun Hean Chongb, Hazilawati Hamzahc, Hasanah Mohd Ghazalia,

T

⁎

a

Department of Food Science, Faculty of Food Science and Technology, Universiti Putra Malaysia, 43400, UPM, Serdang, Selangor, Malaysia Department of Food Technology, Faculty of Food Science and Technology, Universiti Putra Malaysia, 43400, UPM, Serdang, Selangor, Malaysia c Department of Veterinary Pathology and Microbiology, Faculty of Veterinary Medicine, Universiti Putra Malaysia, 43400, UPM, Serdang, Selangor, Malaysia b

G RA P H I C A L AB S T R A C T

A R T I C L E I N F O

A B S T R A C T

Keywords: Avocado oil Subcritical CO2 extraction Ultrasound-assisted aqueous extraction Physicochemical properties

Avocado (Persiana americana Mill) belongs to the Lauraceae family. High level of lipids in the avocado pulp can be served as an important raw material for edible oil extraction. In the present study, the physicochemical properties of avocado oil extracted using subcritical CO2 extraction (SCO2) and ultrasound-assisted aqueous extraction (UAAE) were compared with the conventional solvent extraction. In comparison to solvent extraction, the oils extracted using SCO2 and UAAE were found to have higher iodine values, but lower slip melting points, free fatty acid contents and saponification values. Regardless of the extraction methods, the major fatty acids in avocado oils were oleic (40.73-42.72%) and palmitic (28.12-34.48%) acids whereas the major triacylglycerols in avocado oils were palmitoyl-dioleoyl-glycerol (POO; 22.48-23.01%) and palmitoyl-oleoyl-linoleoyl-glycerol (POL; 17.64-18.23%). SCO2 and UAAE are effective “solvent-free” methods to extract avocado oils and potentially other edible oils.

1. Introduction Avocado (Persiana americana Mill) belongs to the Lauraceae family. The avocado tree can be grown in tropical and subtropical countries although this plant is native to Central America. An avocado fruit can be divided into three anatomical regions, in which the major portion is the pulp (65%), followed by the seed (20%) and peel (15%) [1]. Unlike oil extracted from other fruits, the oil from avocado fruit is extracted from the pulp rather than the seed as the seed contains a low amount of

⁎

oil ( < 2%) and hepatoxic agents [2]. The latter may alter fat metabolism in the liver by enhancing hepatic lipids secretion, liver lipogenesis and the level of lipid biosynthesis enzymes [2]. In contrast, avocado pulp contains high amount of lipids (10–30%) and minerals [3]. In many countries, the avocado pulp is utilized to produce salads, milk shakes, ice-cream and guacamole [4]. Compared to animal fats (e.g. beef tallow and lard), plant oils contain lower levels of saturated fat, but higher amounts of bioactive components such as vitamin E (tocopherols and tocotrienols) and

Corresponding author. E-mail address: [email protected] (H.M. Ghazali).

https://doi.org/10.1016/j.supflu.2017.12.036 Received 28 August 2017; Received in revised form 24 November 2017; Accepted 24 December 2017 Available online 27 December 2017 0896-8446/ © 2017 Elsevier B.V. All rights reserved.

The Journal of Supercritical Fluids 135 (2018) 45–51

C.X. Tan et al.

phytosterols [5]. Plants with high amounts of lipids in either the pulp nor the seed can serve as important raw material for edible oil extraction. Several researchers reported on the similarities between olive oil and avocado oil. For example, both oils contain high levels of oleic acid and are highly digestible [6]. Besides, these oils are extracted from the fruit pulp and the oils contain substantial amounts of bioactive components such as phenolic compounds, carotenoids, phytosterols and antioxidant vitamins [3,7]. Thus, avocado oil can potentially be a substitute for olive oil. Inside the avocado pulp, the evenly scattered thick-walled idioblast cells are surrounded by numerous thin-walled parenchyma cells [2]. These idioblast and parenchyma cells, also known as oil cells, are the oil-containing cells of the avocado pulp [3]. The idioblast cells contain a large oil sac whereas the parenchyma cells contain finely dispersed oil emulsion [2]. Extractions of avocado oil using the organic solvent [4], hot water [8], enzyme [9], centrifugation force [10] and supercritical fluid [11] have been investigated previously. These methods promote the recovery of avocado oil by breaking down the cell walls of the oil cells and the structure of the oil emulsion. However, the use of high temperatures in conjunction with an organic solvent or hot water may cause the oxidative deterioration of polyunsaturated fatty acids in the oil, thereby producing rancid-off flavors [12]. Also, extraction of oil at high temperatures can render some of the bioactive compounds in the plant materials to become inactivated [13]. Subcritical CO2 extraction (SCO2) operates in a similar manner to supercritical CO2 extraction, except that it operates below the critical temperature and pressure (31.10 °C and 72.9 bar, respectively) of CO2. Unlike supercritical CO2 extraction, an oil extracted via SCO2 is usually lighter in color and contains fewer waxes and resins [14]. The mild temperature and pressure used in SCO2 are able to retain most of the thermally sensitive bioactive components in the plant materials. As evidenced by Chia et al. [15], the concentrations of tocols and oryzanol in rice bran oil extracted by SCO2 were 10 folds greater than solvent extraction. The use of ultrasound extraction techniques, particularly ultrasound-assisted aqueous extraction (UAAE) in extracting oils from plant materials is becoming the interest of food industry. Unlike conventional extractions, UAAE does not require the use of organic solvent and the operating procedure is relatively simple. UAAE can be carried out using an ultrasonic bath or an ultrasonic horn transducer. Both types of equipment utilize the cavitation forces produced by acoustic waves to break down the cell walls of the oil cells and the structure of the oil emulsion. UAAE has been well documented for its effectiveness in recovering oil from plant materials [16,17]. For instance, the yield of rice bran oil extracted using UAAE and solvent extraction were almost similar [17]. Previous studies pointed out the commercialization potential of SCO2 and UAAE in extracting oil from plant materials. However, there is little or no study on the extraction of avocado oil by SCO2 and UAAE. Thus, the purpose of this study was to compare the extraction efficiencies and physicochemical properties of avocado oils extracted by SCO2 and UAAE. Conventional solvent extraction served as the control method of the study.

Fig. 1. Schematic diagram of SCO2 extraction system. A1: reboiler unit; A2: condenser unit; A3: extractor unit; V1, V2 and V3: valves.

through a 360 μm stainless steel sieve. The powdered samples were then stored in an air-tight plastic container at −20 °C until use. 2.2. Solvent extraction A semi-continuous solvent extraction method as described by AOAC 920.39 [18] was used. Ten grams of avocado powder was extracted with 200 mL of hexane in a Soxhlet apparatus for 8 h at 70 °C. The hexane was removed by evaporation using a rotary evaporator at 70 °C. To remove residual hexane, the oil was dried in an oven at 70 °C for 15 min. Oil yield was expressed as the percent based on the weight of avocado pulp powder used. 2.3. Subcritical carbon dioxide extraction (SCO2) Fig. 1 shows the schematic diagram of SCO2 instrument (FeyeCon Development, Weesp, Netherland). Three hundred grams of avocado powder was initially loaded into the extractor unit. Liquid CO2 from the liquid CO2 tank was supplied to the reboiler unit via the V1 valve and was converted into CO2 gas. The CO2 gas was channeled to the condenser unit and condensed into liquid CO2 again. Liquid CO2 was continuously supplied to the extractor unit until it was detected by the level sensor and a signal was then sent to the V2 valve. This allows the oil containing-liquid CO2 (extracted from the avocado powder) to flow into the reboiler unit. Inside the reboiler unit, the liquid CO2 evaporated whereas the avocado oil was sedimented at the bottom of the reboiler unit. A complete cycle (about 3 mins) of extraction was achieved when the CO2 gas from the reboiler flowed back into the condenser unit and condensed into liquid CO2. The extraction was carried out at 27 °C and 68 bar. Oil yield was expressed as the percent of oil obtained based on the weight of sample used. 2.4. Ultrasound-assisted aqueous extraction (UAAE)

2. Materials and methods

Ultrasound-assisted aqueous extraction (UAAE) was carried out using the method described by Tan et al. [19]. Ten grams of avocado powder was mixed with 60 mL of distilled water and sonicated in an ultrasonic water bath (Thermo-10D, with an internal dimension of 500 × 300 × 150 mm) operated at an ultrasonic output power of 240 W and a frequency of 40 kHz. The sonication time and temperature were 30 min and 35 °C, respectively. A laboratory scale screw press was used to press the mixture to obtain an aqueous-oil mixture. The aqueous-oil mixture was centrifuged at room temperature for 8000 rpm and 20 min to separate the oil from the water layer. A Pasteur pipette was used to remove the top oil layer. Oil yield was expressed as the percent of oil obtained based on the weight of sample used.

2.1. Sample preparation Avocado fruits were collected from Universiti Putra Malaysia campus, Malaysia in February 2016. Ripened avocado fruits were manually cut into halves and the pulp was separated from the peel and the seed. The pulp was wrapped with aluminum foil and a tray type dryer (Memmert UFB 400, Schwabach, Germany) operating at 35 °C was used to dry the pulp for three consecutive days. The dried avocado pulp (moisture content: 2–3%) was ground into powder using a home blender (Panasonic MS801S, Petaling Jaya, Malaysia) and sieved 46

The Journal of Supercritical Fluids 135 (2018) 45–51

C.X. Tan et al.

2.5. Determination of physicochemical properties

Table 1 Extracting conditions and oil yield.

The free fatty acid, slip melting point, iodine value, saponification value and unsaponifiable matter of samples were determined using AOCS official methods [20]. The color of samples was determined using a calibrated HunterLab UltraScan Pro spectrophotometer (Hunter Associate Laboratory Inc., Rseton, USA) along with the EasyMatch QC software. A 50 mm path length transmission cell was used and all samples were prepared without any dilution. The results were expressed as L*, a* and b* values.

Conditions

Extracting solvent Time (min) Temperature (°C) Pressure (bar) Oil yield (%)

Extraction methods Solvent

SCO2

UAAE

hexane 480 70 atmospheric 20.79 ± 0.27a

CO2 450 27 68 16.97 ± 0.66b

water 30 35 atmospheric 15.13 ± 0.23c

Mean values in the same row with different letters are significantly different at p < 0.05.

2.6. Determination of fatty acid composition

melting profile was divided into various temperatures (0 °C, 10 °C, 20 °C, 30 °C, 40 °C and 50 °C) and the total crystallization energy (J/g) was transformed into percentage at each temperature for SFI.

The fatty acid composition was analyzed after conversion of samples to fatty acid methyl ester (FAME) [21]. The FAME was prepared by mixing 0.1 g of oil with 5 mL of hexane and 250 μL of sodium methoxide reagent in a test tube. The test tube was capped, vortex for 1 min and added with 5 mL of saturated sodium chloride solution. The test tube was vigorously shaken for 15 s before left standing for 10 min. The top layer (1 μL) was injected into a gas-chromatography (Perkin-Elmer Clarus 500, Shelton, Connecticut, USA) equipped with a flame-ionization detector (FID). A polar capillary column (30 m × 0.25 mm, with a film thickness of 0.25 μm; SGE incorporated, Austin, Texas, USA) was used. The column temperature was initially 100 °C, held for 2 mins and then increased at the rate of 5 °C/min to 230 °C and held for 10 mins. Throughout the analysis, both the injector and detector temperatures were fixed at 250 °C. The peaks of the samples were identified by comparison of the retention times with FAME standards. The percentage of fatty acid was calculated as the ratio of the partial area to the total peak area.

2.10. Statistical analysis All the analyses were performed in triplicate. The data were analyzed by one-way analysis of variance (ANOVA) accompanied with Turkey’s post hoc using SPSS version 20 software (IBM Corp., Armonk, New York). The level of significant was set at p < .05. 3. Results and discussion 3.1. Comparison of oil yield The extracting conditions and oil yields of solvent, SCO2 and UAAE are shown in Table 1. Generally, solvent extraction used the longest extracting time and highest extracting temperature. Results indicated solvent extraction (20.79%) produced the highest oil yield, followed by SCO2 (16.97%) and UAAE (15.13%). There was a significant difference (p < .05) between the oil yield and extraction methods. This is in good agreement with Hamzah [8], who stated that organic solvents have the greater capability to extract most of the lipid components including waxes, phospholipids and pigments from the avocado pulp. According to Woolf et al. [3], virgin avocado oil is obtained by mechanical or natural means at low temperatures ( < 50 °C) and without undergoing chemical refining. Thus, the avocado oils extracted following the methods of SCO2 and UAAE are considered as virgin avocado oils. Although solvent extraction produces the highest oil yield, but the extraction time is long yet the extraction temperature is high, leading to high energy consumption and hence, uneconomical. The requirement of high pressure (72.9-500 bar) in supercritical CO2 extraction hindering the wide adoption of this method in the food industry as its high operating cost [26]. Unlike supercritical CO2 extraction, the pressure used in SCO2 is much lower (68 bar) and hence, more economical. Meanwhile, edible oils extracted using SCO2 and UAAE contain no traces of chemical solvents, a well-embraced food property by the community.

2.7. Determination of triacylglycerol profile The triacylglycerol (TAG) profile of samples was analyzed using a high-performance liquid chromatography (Shimadzu LC-10AD, Kyoto, Japan) equipped with a refractive index detector (Shimadzu RID-6A, Kyoto, Japan) on a LiChrospher RP-18 column (125 mm × 4 mm, with a particle size of 5 μm; Merck, Darnstadt, Germany) according to the method of Ghazali et al. [22]. The mobile phase was an acetone-acetonitrile mixture (63.5:36.5) held at a column temperature of 30 °C and a flow rate of and 1.0 mL/min. The injector volume was 10 μL of 5% (w/w) oil in acetone. The chromatogram was processed using a Shimadzu CR4AX-integrator. The TAG of samples was quantified based on the retention time of TAG standards and compared with the TAG composition of avocado oil reported by Lísa and Holčapek [23]. Peak areas generated from the data integrator were used to quantify the components according to the relative percentage. 2.8. Determination of thermal behavior The crystallization and melting behaviors of samples were determined using a differential scanning colorimeter (Perkin–Elmer Diamond DSC, Shelton, Connecticut, USA) following the method of Man and Swe [24]. Zinc and indium were used to calibrate the instrument. The purge gas used was 99.99% nitrogen with a pressure of 20 psi and a flow rate of 100 mL/min. About 5–7 mg of oil was measured into the aluminum pan and hermetically sealed. The reference used was an empty hermetically sealed aluminum pan. The sample was subjected to the temperature programs: the sample was held isotherm at 60 °C for 2 min, then cooled at 5 °C/min to −60 °C and held for 2 min before heated from −60 °C to 60 °C at 5 °C/min.

3.2. Physicochemical properties The functionality of oils is affected by its physicochemical properties. The physicochemical properties of avocado oils extracted by SCO2 and UAAE are compared with the solvent extraction as shown in Table 2. At room temperature, SCO2- and UAAE-extracted oils appeared as liquid, while solvent-extracted oil was semi-solid. This observation is supported by their respective slip melting point (SMP) values, in which the oils extracted with SCO2 (24.10 °C) and UAAE (24.20 °C) were significantly lower (p < .05) than solvent (31.90 °C). The difference of the SMP among the examined oils is also reflected in their iodine values. Iodine value measures the degree of unsaturation in oils. According to Emil et al. [27], oils with more unsaturated fatty acid bonds have higher iodine values and tend to stay liquid at room temperature. The iodine value of SCO2- and UAAE-extracted oils (81.70 g I2/100 g

2.9. Determination of solid fat index The solid fat index (SFI) of samples was calculated from the DSC melting profile following the method of Adhikari et al. [25]. The DSC 47

The Journal of Supercritical Fluids 135 (2018) 45–51

C.X. Tan et al.

a* and b* parameters of avocado oil extracted using various methods were compared. The SCO2-extracted oil displayed higher L* and b* values than UAAE- and solvent-extracted oils. This indicates SCO2-extracted oil was lighter-colored and contains more yellow pigments. This is in accordance with the results obtained through naked eyes observation, in which the SCO2-extracted oil appeared light yellow. Negative a* value in the solvent-extracted oil (-0.71) suggests the presence of greater green pigment components (e.g. chlorophylls) in the oil. Meanwhile, the readings of a* value in the SCO2-extracted oil (11.25) was higher than the UAAE-extracted oil (0.15), implying the extraction conditions used in SCO2 favors the extraction of red pigment components (e.g. carotenoids) from the avocado pulp. Compared with darker (lower L* value) plant oils, lighter-colored plant oils are more suitable for edible and industrial purposes [12]. Hence, avocado oils extracted by SCO2 and UAAE are more suitable for edible and industrial purposes, as compared with the oil extracted using solvent.

Table 2 Physicochemical properties of avocado oils extracted by different extraction methods. Parameters

Slip melting point (°C) Iodine value (g I2/100 g) Saponification value (mg KOH/g) Unsaponifiable matter (%) Free fatty acid (% oleic acid) Color L* a* b*

Extraction methods Solvent

SCO2

UAAE

31.90 ± 0.10a 75.61 ± 0.91a 186.16 ± 0.88a

24.10 ± 0.10b 81.70 ± 0.46b 179.77 ± 1.05b

24.20 ± 0.10b 83.21 ± 0.38b 177.80 ± 0.74b

1.58 ± 0.13a

1.64 ± 0.03a

1.41 ± 0.07b

0.38 ± 0.03a

0.32 ± 0.00b

0.29 ± 0.03b

1.29 ± 0.32a -0.71 ± 0.57a 1.09 ± 0.26a

43.99 ± 0.20b 11.25 ± 0.08b 73.45 ± 0.98b

19.18 ± 0.23c 0.15 ± 0.04c 18.75 ± 0.31c

Mean values in the same row with different letters are significantly different at p < 0.05.

3.3. Fatty acid profile and 83.21 g I2/100 g, respectively) were significantly higher (p < .05) than solvent-extracted oil (75.61 g I2/100 g). Literature comparison indicates the avocado oils exhibit higher iodine values than papaya seed oils (66–69 g I2/100 g), but lower than sunflower (120–127 g I2/100 g) and Kalahari melon (125–141 g I2/100 g) seed oils [12,28,29]. The range of saponification value of avocado oils (177.80186.16 mg KOH/g) was comparable to the oils extracted from the Kalahari melon (173–185 mg KOH/g) and sunflower (185–190 mg KOH/g) seeds [12,29]. From Table 2, the saponification value of solvent-extracted oil was significantly higher (p < 0.05) than SCO2- and UAAE-extracted oils. Plant oils with a greater saponification value implies the existence of higher molecular weight triacylglycerols, and this suggests the suitability in the production of body care products [27,30]. Meanwhile, the unsaponifiable matter of the oils extracted by SCO2 (1.64%) and solvent (1.58%) was significantly greater (p < .05) than the oil extracted by UAAE (1.41%). This could be due to the interaction between unsaponifiable matter such as lipid-soluble vitamins, sterols, triterpene alcohols and hydrocarbons with non-polar solvents (e.g. hexane and CO2) are stronger than polar solvent (e.g. water). Free fatty acid is used to monitor the quality of edible oils. The degree of acceptability and edibility of an oil are deemed to have a negative relationship with the amount of free fatty acid [30]. The maximum allowable of free fatty acid content in avocado oil is 1% [3]. All the extraction methods displayed low free fatty acid values (< 0.50%), indicating the oils extracted are of good quality. The natural pigments of oils affect its observable color. The primary pigments imparting the color of oils are chlorophylls (green) and carotenoids (red, yellow and orange) [31]. Fig. 2 shows the avocado oils obtained by each extraction method. Through naked eyes observation, the solvent-extracted oil was dark golden yellow whereas SCO2- and UAAEextracted oils were light yellow and dark yellow, respectively. The L*,

Fatty acid composition affects the physical states, nutritional value and stability of oils. The fatty acid composition of avocado oils extracted using SCO2 and UAAE are compared with the solvent extraction as shown in Table 3. Regardless of the extraction methods, the major fatty acids in the avocado oil were oleic (40.73-42.72%), palmitic (28.12-34.48%), linoleic (15.52-18.88%) and palmitoleic (6.64-8.50%) acids. This trend is in accordance with the results reported by Yanty et al. [32]. There was significant difference (p < 0.05) in the amounts of the major fatty acids in the solvent-, SCO2- and UAAE-extracted oils. High oleic and linoleic acids plant oils could be used as salad oil [30]. Besides, plant oils with high oleic acid contents were reported to have enough oxidative stability in domestic culinary applications such as frying [30,33]. The presence of high amounts of oleic (> 40%) and linoleic (> 15%) acids in the avocado oil suggest its suitability as domestic culinary oil besides being used as salad oil. Saturated fatty acids (SFA) such as palmitic and stearic acids have been positively related to the increment of blood cholesterol levels and the risk of cardiovascular diseases [34]. Results indicated the SFA of solvent-extracted oil (35.55%) was significantly higher (p < .05) than SCO2- and UAAE-extracted oils (31.11% and 29.21%, respectively). A similar observation was reported by Werman and Neeman [10], whereby solvent-extracted avocado oil contained a higher level of SFA than centrifuge-extracted avocado oil. The variation of SFA in the present study could be attributed to the selectivity of specific extracting solvent for certain fatty acids [35]. 3.4. Triacylglycerol composition One of the main components in plant oil is triacylglycerol (TAG) as it represents 98% of the whole oil composition [36]. The physical properties of oils such as melting and crystallization are influenced by Fig. 2. Avocado oils obtained by different extraction methods. A: solvent; B: SCO2; C: UAAE.

48

The Journal of Supercritical Fluids 135 (2018) 45–51

C.X. Tan et al.

The data in the present study demonstrated the amount of tri-unsaturated TAGs (LLL, LLO, LOO and OOO) of solvent-extracted oil (22.47%) was lower than SCO2- and UAAE-extracted oils (25.25% and 25.30%, respectively). This observation disagrees with Samaram et al. [37], who reported the amount of tri-unsaturated TAG contents of ultrasound-assisted solvent-extracted papaya seed oil (41.90%) was slightly lower than solvent-extracted papaya seed oil (45.25-46.65%). The difference may have arisen from the extracting condition used in the ultrasound-assisted extraction (UAE), as the sonication temperature used by Samaram et al. [37] was higher. As reported by Corbin et al. [38], thermal damages on the unsaturated compounds can occur if a higher extracting temperature is used during the UAE. Besides, the high melting tri-saturated TAG (PPP) of solvent-extracted oil (1.88%) was significantly higher (p < .05) than SCO2- and UAAEextracted oils (1.30% and 0.62%, respectively). According to Tanaka et al. [39], oil with a higher amount of PPP produces larger granular crystals when long storage. The presence of large granular crystals in oil gives an unpleasant sandy sensation during consumption [40].

Table 3 Fatty acid composition of avocado oils extracted by different extraction methods. Fatty acids

C16:0 C16:1 C18:0 C18:1 C18:2 C18:3 SFA MUFA PUFA

Determined values

Reported range of values [32]

Solvent

SCO2

UAAE

34.48 ± 0.24a 6.64 ± 0.12a 1.07 ± 0.03a 40.73 ± 0.43a 15.52 ± 0.04a 1.55 ± 0.10a 35.55 ± 0.23a 47.37 ± 0.32a 17.07 ± 0.11a

30.88 ± 0.34b 6.80 ± 0.29b 0.23 ± 0.02b 42.72 ± 0.32b 17.76 ± 0.10b 1.51 ± 0.04a 31.11 ± 0.34b 49.62 ± 0.26b 19.27 ± 0.07b

28.12 ± 0.12c 8.50 ± 0.30c 0.63 ± 0.18c 41.74 ± 0.12c 18.88 ± 0.32c 2.14 ± 0.13b 29.21 ± 0.15c 50.07 ± 0.13b 21.02 ± 0.43c

14.80−30.37 4.40−7.44 0.27−1.56 43.65−63.73 12.75−17.45 1.09−2.03 15.07−31.66 48.87−68.59 13.95−19.48

Mean values in the same row with different letters are significantly different at p < .05. SFA: total saturated fatty acids; MUFA: total monounsaturated fatty acids; PUFA: total polyunsaturated fatty acids. Table 4 Triacylglycerol composition of avocado oils extracted by different extraction methods. TAG

LLL PLLn LLO PLL PLPo PPLn LOO POL PLP PPPo OOO POO POP PPP SOO POS Others

3.5. Solid fat index

Extraction methods (%) Solvent

SCO2

UAAE

0.78 ± 0.06a 0.89 ± 0.15a 2.64 ± 0.17a 3.74 ± 0.02a 3.84 ± 0.19a 0.39 ± 0.12a 8.05 ± 0.06a 18.02 ± 0.24a 5.11 ± 0.03a 1.57 ± 0.13a 11.00 ± 0.01a 23.01 ± 0.20a 11.60 ± 0.51a 1.88 ± 0.11a 0.71 ± 0.18a 0.62 ± 0.29a 6.29 ± 0.96

1.67 ± 0.06b 0.92 ± 0.18a 3.57 ± 0.40a 3.76 ± 0.35a 3.58 ± 0.16a 0.36 ± 0.21a 8.50 ± 0.04a,b 18.23 ± 0.30a 4.91 ± 0.17a 1.17 ± 0.41a 11.51 ± 0.13b 22.48 ± 0.03a 10.72 ± 0.62a 1.30 ± 0.13b 0.67 ± 0.05a 0.44 ± 0.18a 6.22 ± 0.52

1.77 ± 0.07b 1.09 ± 0.05a 3.58 ± 0.24a 3.65 ± 0.14a 3.65 ± 0.01a 0.37 ± 0.00a 8.59 ± 0.21b 17.64 ± 0.13a 5.01 ± 0.13a 1.23 ± 0.08a 11.36 ± 0.04b 22.66 ± 0.95a 11.18 ± 0.27a 0.62 ± 0.13c 0.93 ± 0.30a 0.53 ± 0.15a 6.47 ± 0.37

Solid fat index (SFI) examines the solid-liquid ratio in the lipids and is essential for the purposes of quality control and monitoring processes such as hydrogenation, tempering and fractionation. The SFI of avocado oils extracted using SCO2 and UAAE are compared with the solvent extraction as shown in Fig. 3. When the temperatures increase, the SFI values of oils decrease. At 0 °C, the SFI of oils extracted by solvent, SCO2 and UAAE were 42.33%, 22.28%, and 13.61%, respectively. In the temperatures range of 10–20 °C, the SFI values of UAAE-extracted oil decreased faster than SCO2- and solvent-extracted oils. The SFI values of SCO2- and UAAE-extracted oils became 0% at the temperatures range of 25–30 °C while the SFI value of solvent-extracted oil became 0% at the temperatures range of 40–45 °C. This observation in agreement with Yanty et al. [32], who reported the SFI value of solvent-extracted avocado oil became 0% at the temperature range of 40–50 °C. The melting point of lipids is affected by the degree of saturation and chain length [41]. In the present study, the levels of unsaturated fatty acids (Table 3) and tri-unsaturated TAGs (Table 4) of avocado oils extracted using SCO2 and UAAE were slightly greater than solvent extraction, hence the oils melted completely at much lower temperatures.

Mean values in the same row with different letters are significantly different at p < .05. L: linoleic; O: oleic; S: stearic; P: palmitic; Po: palmitoleic; Ln: linolenic.

the TAG composition [37]. The TAG composition of avocado oils extracted using SCO2 and UAAE are compared with the solvent extraction as shown in Table 4. Regardless of the extraction methods, POO (22.4823.01%) was the major TAG, followed by POL (17.64-18.23%), OOO (11.00-11.51%) and POP (10.72–11.60%).

3.6. Thermal behavior Differential scanning calorimetry (DSC) is a thermoanalytical tool to characterize the phase transitions of oils. It provides an alternative way to study the chemical compositions of oils. The crystallization and melting Fig. 3. Solid fat index of avocado oils obtained by different extraction methods.

49

The Journal of Supercritical Fluids 135 (2018) 45–51

C.X. Tan et al.

Fig. 4. DSC curves show (i) crystallization [a1-a3, b1-b3 and c1-c3 are the exothermic peaks of A: solvent, B: SCO2 and C: UAAE, respectively] and (ii) melting [a1-a5, b1-b3 and c1-c3 are the endothermic peaks of A: solvent, B: SCO2 and C: UAAE, respectively] of avocado oils extracted by different extraction methods.

hands, solvent-extracted oil exhibited an exothermic peak above 20 °C, but SCO2- and UAAE-extracted oils did not display any peak in these temperatures regions. The differences in the crystallization peaks may be attributed to the differences in the extraction methods [42]. As shown in Tables 3 and 4, different extraction conditions influenced the degree of saturation in the avocado oils. Generally, oils with high levels of saturated fatty acids and tri-saturated TAGs crystallize at higher temperatures regions [43]. In Fig. 4 (ii), the melting curves of avocado oils extracted by solvent, SCO2 and UAAE are represented by the curves (A), (B) and (C), respectively. The melting curve of solvent-extracted oil exhibited five endothermic peaks, with two shoulder peaks (a3 and a4) located within three endothermic peaks (a1, a2 and a5). Meanwhile, the melting curves

curves of avocado oils extracted using SCO2 and UAAE are compared with the solvent extraction as shown in Fig. 4 (i) and (ii), respectively. In Fig. 4 (i), the crystallization curves of avocado oils extracted by solvent, SCO2, and UAAE are represented by the curves (A), (B) and (C), respectively. All the crystallization curves displayed three exothermic peaks, with a major peak (a2, b2 and c2) at the center, along with two small peaks at the initial (a1, b1 and c1) and end (a3, b3 and c3) points of the crystallization. The Fig. 4 (i) clearly shows the points of crystallization peaks of each extraction method were distinctly different. For example, the initial crystallization peaks of solvent-, SCO2- and UAAEextracted oils were a1 (22.27 °C), b1 (13.62 °C) and c1 (8.07 °C), respectively. Compared with the solvent- and SCO2-extracted oils, the crystallization peak of UAAE-extracted oil was the lowest. On the other 50

The Journal of Supercritical Fluids 135 (2018) 45–51

C.X. Tan et al.

of SCO2- and UAAE-extracted oils exhibited three endothermic peaks, with a broader peak (b1 and c1) appeared at the initial point of melting, followed by two peaks (b2, b3, c2 and c4). Interestingly, solvent-extracted oil displayed two peaks (a1 and a2) at low-melting region (< 0 °C) and three peaks (a3, a4 and a5) at high-melting region (> 0 °C). However, this phenomenon was not observed on the melting curves of SCO2- and UAAE-extracted oils. All the melting peaks of SCO2- and UAAE-extracted oils were located within the low-melting region. This clearly shows the melting behavior of avocado oils was affected by extraction methods. The possible reasons for these differences can be due to the variations in the proportional distributions of the identified fatty acid and TAG components. As discussed, the thermal transition of lipids is affected by the compositional changes such as the degree of saturation and fatty acid chain length [43]. The SFA and tri-saturated TAG of solvent-extracted oil were slightly higher than SCO2- and UAAEextracted oils, and highly saturated compounds have higher melting points. Thus, higher temperatures are needed to melt completely.

[14] Vitalis, Subcritical Fuid Extraction, (2016) (accessed October 13, 2016). http:// vitaliset.com/subcritical-extraction/. [15] S.L. Chia, H.C. Boo, K. Muhamad, R. Sulaiman, F. Umanan, G.H. Chong, Effect of subcritical carbon dioxide extraction and bran stabilization methods on rice bran oil, J. Am. Oil Chem. Soc. 92 (2015) 393–402. [16] T.C. Kha, M.H. Nguyen, P.D. Roach, C.E. Stathopoulos, Ultrasound-Assisted aqueous extraction of oil and carotenoids from microwave-Dried gac (Momordica cochinchinensis spreng) aril, Int. J. Food Eng. 11 (2015) 479–492. [17] M. Khoei, F. Chekin, The ultrasound-assisted aqueous extraction of rice bran oil, Food Chem. 194 (2016) 503–507. [18] AOAC, Official Methods of Analysis of AOAC International, 18th ed., Association of Official Analytical Chemists, Washington, 2007. [19] C.X. Tan, G.H. Chong, H. Hamzah, H.M. Ghazali, Optimization of ultrasound-assisted aqueous extraction to produce virgin avocado oil with low free fatty acids, J. Food Process Eng. (2017) e12656, http://dx.doi.org/10.1111/jfpe.12656. [20] AOCS, Official and Tentative Methods of the American Oil Chemists’ Society, 5th ed., AOCS, Champaign, USA, 1999. [21] L. Cocks, C. Rede, Laboratory Handbook for Oil and Fat Analysts, Academic Press, London, 1966. [22] H.M. Ghazali, A. Maisarah, S. Yusof, M.S.A.M. Yusoff, Triglyceride profiles and melting properties of lipase-catalysed transesterified palm stearin and coconut oil, Asia-Pac. J. Mol. Biol. Biotechnol. 3 (1995) 280–289. [23] M. Lísa, M. Holčapek, Triacylglycerols profiling in plant oils important in food industry, dietetics and cosmetics using high-performance liquid chromatography-atmospheric pressure chemical ionization mass spectrometry, J. Chromatogr. A 1198–1199 (2008) 115–130. [24] Y.B.C. Man, P.Z. Swe, Thermal analysis of failed-batch palm oil by differential scanning calorimetry, J. Am. Oil Chem. Soc. 72 (1995) 1529–1532. [25] P. Adhikari, X. Zhu, A. Gautam, J. Shin, J. Hu, J. Lee, K. Lee, Scaled-up production of zero-trans margarine fat using pine nut oil and palm stearin, Food Chem. 119 (2010) 1332–1338. [26] A.N. Mustapa, Z.A. Manan, C. Azizi, Subcritical and supercritical fluid extraction: a critical review of its analytical usefulness, J.Chem. Nat. Resour. Eng. (2008) 164–180. [27] A. Emil, Z. Yaakob, M.N.S. Kumar, J.M. Jahim, J. Salimon, Comparative evaluation of physicochemical properties of jatropha seed oil from Malaysia, Indonesia and Thailand, J. Am. Oil Chem. Soc. 87 (2010) 689–695. [28] T. Puangsri, S.M. Abdulkarim, H.M. Ghazali, Properties of Carica papaya L. (papaya) seed oil following extractions using solvent and aqueous enzymatic methods, J. Food Lipids 12 (2005) 62–76. [29] K.L. Nyam, C.P. Tan, Y.B. Che Man, O.M. Lai, K. Long, Physicochemical properties of Kalahari melon seed oil following extractions using solvent and aqueous enzymatic methods, Int. J. Food Sci. Technol. 44 (2009) 694–701. [30] K.L. Nyam, C.P. Tan, O.M. Lai, K. Long, Y.B. Che Man, Physicochemical properties and bioactive compounds of selected seed oils, LWT − Food Sci. Technol. 42 (2009) 1396–1403. [31] D.M. Barrett, J.C. Beaulieu, R. Shewfelt, Flavor Color, Texture, and nutritional quality of fresh-cut fruits and vegetables − desirable levels, instrumental and sensory measurement, and the effects of processing, Crit. Rev. Food Sci. Nutr. 50 (2010) 369–389. [32] N.A.M. Yanty, J.M.N. Marikkar, K. Long, Effect of varietal differences on composition and thermal characteristics of avocado oil, J. Am. Oil Chem. Soc. 88 (2011) 1997–2003. [33] K. Warner, S. Knowlton, Frying quality and oxidative stability of high-oleic corn oils, J. Am. Oil Chem. Soc. 74 (1997) 1317–1322. [34] E. Fattore, R. Fanelli, Palm oil and palmitic acid: a review on cardiovascular effects and carcinogenicity, Int. J. Food Sci. Nutr. 64 (2013) 648–659. [35] R. Carvalho, E. Galvao, J. Barros, Extraction, fatty acid profile and antioxidant activity of sesame extract (Sesamum Indicum L.), J. Chem. Eng. 29 (2012) 409–420. [36] M.J. Lerma-García, R. Lusardi, E. Chiavaro, L. Cerretani, A. Bendini, G. RamisRamos, E.F. Simó-Alfonso, Use of triacylglycerol profiles established by high performance liquid chromatography with ultraviolet-visible detection to predict the botanical origin of vegetable oils, J. Chromatogr. A 1218 (2011) 7521–7527. [37] S. Samaram, H. Mirhosseini, C.P. Tan, H.M. Ghazali, Ultrasound-assisted extraction (UAE) and solvent extraction of papaya seed oil: yield, fatty acid composition and triacylglycerol profile, Molecules 18 (2013) 12474–12487. [38] C. Corbin, T. Fidel, E.A.E. Leclerc, E. Barakzoy, N. Sagot, A. Falguiéres, S. Renouard, J.P. Blondeau, C. Ferroud, J. Doussot, E. Lainé, C. Hano, Development and validation of an efficient ultrasound assisted extraction of phenolic compounds from flax (Linum usitatissimum L.) seeds, Ultrason. Sonochem. 26 (2015) 176–185. [39] L. Tanaka, T. Isogai, S. Miura, Effect of triacylglycerol species on the crystallizing behavior of a model water/oil emulsion, Eur. J. Lipid Sci. Technol. 112 (2010) 304–309. [40] A. Watanabe, I. Tashima, N. Matsuzaki, On the formation of granular crystals in fat blends containing palm oil, J. Am. Oil Chem. Soc. 69 (1992) 1088–1080. [41] J. Berg, J. Tymoczko, L. Stryer, Fatty Acid Metabolism, 5th ed., W. H Freeman and Company, New York, 2002. [42] S. Samaram, H. Mirhosseini, C.P. Tan, H.M. Ghazali, Ultrasound-assisted extraction and solvent extraction of papaya seed oil: crystallization and thermal behavior, saturation degree, color and oxidative stability, Ind. Crops Prod. 52 (2014) 702–708. [43] J.M.N. Marikkar, H.M. Ghazali, Y.B. Che Man, O.M. Lai, The use of cooling and heating thermograms for monitoring of tallow, lard and chicken fat adulterations in canola oil, Food Res. Int. 35 (2002) 1007–1014.

4. Conclusion Avocado oil has the potential to be functional oil and its full potential could be further explored. The utilization of “solvent-free” extractions such as SCO2 and UAAE in the food industry are highly recommended in the wake of potential emission of organic solvent into the atmosphere during oil manufacturing processes. Besides, edible oil extracted using organic solvent may contain chemical residue which could be toxic to human health. In the present study, avocado oils extracted using SCO2 and UAAE were compared with the conventional solvent extraction. Although the oil yield obtained by SCO2 and UAAE (16.97% and 15.13%, respectively) were lower than solvent extraction (20.79%), but SCO2- and UAAE-extracted oils were found to be lighter in color and contained higher levels of unsaturated fatty acids than solvent-extracted oil. It can be concluded that SCO2 and UAAE are effective methods for avocado oil extractions and there is a commercialization potential using these methods. Acknowledgement This work was financially supported by Universiti Putra Malaysia under grant number GP-IPS 9535600. References [1] G. Costagli, M. Betti, Avocado oil extraction processes: method for cold-pressed high-quality edible oil production versus traditional production, J. Agric. Eng. 46 (2015) 115. [2] X. Qin, J. Zhong, A review of extraction techniques for avocado oil, J. Oleo Sci. 65 (2016) 1–8. [3] A. Woolf, M. Wong, L. Eyres, T. McGhie, Avocado Oil, AOCS Press, USA, 2008. [4] E.D. Mooz, N.M. Gaiano, M.Y.H. Shimano, R.D. Amancio, M.H.F. Spoto, Physical and chemical characterization of the pulp of different varieties of avocado targeting oil extraction potential, Food Sci. Technol. 32 (2012) 274–280. [5] R. Foster, C.S. Williamson, J. Lunn, Culinary oils and their health effects, Nutr. Bull. 34 (2009) 4–47. [6] M. Reddy, R. Moodley, S.B. Jonnalagadda, Fatty acid profile and elemental content of avocado (Persea americana Mill.) oil–effect of extraction methods, J. Environ. Sci. Health B 47 (2012) 529–537. [7] I. Berasategi, B. Barriuso, D. Ansorena, I. Astiasarán, Stability of avocado oil during heating: comparative study to olive oil, Food Chem. 132 (2012) 439–446. [8] B. Hamzah, The effect of homogenization pressures on extraction of avocado oil by wet method, Adv. J. Food Sci. Technol. 5 (2013) 1666–1667. [9] M. Buenrostro, A. López-Munguia, Enzymatic extraction of avocado oil, Biotechnol. Lett. 8 (1986) 505–506. [10] M. Werman, I. Neeman, Avocado oil production and chemical characteristics, J. Am. Oil Chem. Soc. 64 (1987) 229–232. [11] B. Botha, R. McCrindle, Supercritical fluid extraction of avocado oil: south african avocado grow, Assoc. Yearbook 27 (2004) 24–27. [12] S. Latif, F. Anwar, Effect of aqueous enzymatic processes on sunflower oil quality, J. Am. Oil Chem. Soc. 86 (2009) 393–400. [13] G. Spigno, L. Tramelli, D. De Faveri, Effects of extraction time, temperature and solvent on concentration and antioxidant activity of grape marc phenolics, J. Food Eng. (2007).

51