Microwave and Ultrasound Pretreatment of Moringa oleifera Lam. Seeds: Effects on Oil Expression, Oil Quality, and Bioactive Component

Extraction of oil from fruit oils, such as those from Moringa oleifera , is either performed through mechanical Abstract: This study investigates the application of green technologies (microwave and ultrasound pretreatment) in the extraction of Moringa oleifera Lam. seed oil and its effects on oil expression, oil quality, and bioactive component. Moringa seeds were pretreated with microwave (90 W, 60 s) or ultrasound (50 W, 1 h) before mechanical expression. A separate group received no pretreatment before oil extraction. Oils from these groups were then compared. Results show that oil yield increased with ultrasound pretreatment (1.24%) and significantly increased with microwave pretreatment (3.11%). For oil flow rate, the microwave and ultrasound pretreatment resulted in faster extraction (7.67 and 6.93 kg/h respectively) as compared with the control (6.51 kg/h). For physicochemical parameters, the microwave and ultrasound group had significantly less free fatty acids and significantly greater unsaponifiable matter as compared with the control. For fatty acid composition, results show that moringa seeds procured from Davao Oriental had greater oleic acid content (~77%) as compared with those reported by other literature. For phytosterol content, the predominant phytosterols found were β-sitosterol, stigmasterol, and campesterol. Microwave and ultrasound pretreatment significantly increased total phytosterol (680.58 and 369.32 mg/kg respectively) as compared with the control (72.69 mg/kg) due to the mass transfer of the phytosterols. Microwave and ultrasound pretreatment also led to stigmastanol formation. For antioxidant activity, a comparison of both DPPH and FRAP assays depicts that the microwave group exhibited the best overall antioxidant activity. Lastly, for oil stability, a lower peroxide value was found in the microwave and ultrasound groups across time intervals, which may be attributed to their antioxidant activity. In summary, ultrasound and microwave pretreatment can improve oil expression, oil quality, and bioactive content of the mechanically expressed moringa oils.

or solvent extraction 2,11,13 . Oil extraction aims to achieve high oil yield without adversely affecting oil quality 13 . Solvent extraction is advantageous in providing a higher oil yield 2 . However, solvent extraction poses risk as chemicals may remain in extracted samples 14 . Moreover, the use of solvents has ecological impacts, high energy consumption, and longer processing time 2,15,16 . The use of heat in solvent extraction may also destroy phytochemicals and alter the fatty acid composition of the oil 2 .
Thus, there is a trend in employing green technologies in food processing. Contemporary literature has examined using these green technologies to increase oil extraction efficiency while retaining the quality of the oil 13,16 . Examples of these green technologies include microwaves and ultrasound. Microwaves involve the use of low-frequency electromagnetic waves 17 . Microwaves are advantageous in that they require minimal energy, time, and operational expenses while retaining bioactive components of food 17 19 . On the other hand, ultrasound incorporates the use of sound waves that induces cavitation and agitates the food matrix, consequently building up pressure and heat 20 . The advantages of ultrasound include higher extraction yields, microbe decapacitation, affordability, and minimal energy utilization 20,21 . The extent of use of microwaves and ultrasound should be monitored as it may adversely affect oil quality. These include the formation of free radicals and harmful compounds, changes in physicochemical properties e.g. color , and destruction of bioactive compounds 13,16,20,22 . This implies that there is a need to further investigate the effects of green technologies such as microwave and ultrasound in the pretreatment extraction of Moringa oleifera. Employing mechanical extraction would extract oil without the use of harmful solvents. Coupled with ultrasound or microwave, mechanical extraction of the oil may result in improved oil yield. Hence, this study investigated the effects of microwave and ultrasound pretreatment of Moringa oleifera seeds on oil extraction, oil quality, and bioactive component of mechanically expressed moringa oil.

Chemicals, reagents and materials
Mature and dry moringa pods were sourced from Davao Oriental, Mati City, which has an average temperature of 28 . Mature seeds were collected and handpicked from several moringa trees from the same farm plot. The collected seeds were treated as a composite sample. The seeds were then cleaned to remove foreign materials, such as rocks.

Moringa seeds preparation and pretreatment
The moringa seeds were inspected to eliminate damaged seeds and other unnecessary materials. The seeds were airdried down to a moisture content of about 11.00 based on the recommendations of Fakayode and Ajav on oil extraction of Moringa oleifera 23 . A hundred moringa seeds were sampled for the average mass of the seeds. The seed coats were then removed to determine the average mass of the kernels. The seeds were then pretreated with microwave, ultrasound, or neither control group . For the microwave pretreatment, each batch approx. 100 g was placed in a glass tray inside the microwave oven. The seeds were exposed to microwave irradiation 90 W, 60 s with frequency of 2.45 GHz. For the ultrasound pretreatment, two 250-ml beakers were placed in the ultrasonic bath capacity of 600 mL containing 200 mL distilled water. Each batch approx. 150 g were then exposed to ultrasound 50 W, 60 min. with a frequency of 40 kHz.
After pretreatment, the seeds were stored in moistureresistant plastic bags doubly wrapped with each bag containing 2.000 kg of moringa seeds. A bag of seeds represented one trial for a pretreatment method. Three bags of seeds were prepared for each pretreatment method for a total of three trials before oil extraction.

Moringa oil extraction
For the extraction of the oil, a 60-kg oil expeller was used. Each batch of seeds 2.000 kg was fed to the hopper for oil extraction. Oils were collected in a cylindrical plastic container. The extracted oils were then filtered through gravity filtration pore size: 11 μm to remove impurities, which could affect oil composition and quality. The filtered moringa oil were stored in 500-mL amber bottles inside an incubator at 20.0 with minimal exposure to light. The oil yield and flow rate as a percentage of the kernel weight were then computed using Formulas 1 and 2.

Determination of fatty acid composition of Moringa oil
The fatty acid composition of the moringa oil was determined by esterification of fatty acids using AOAC 969.33 through gas chromatography 24 . For gas chromatography, Rt ® -2560 fused silica column 0.25 mm ID, 100 m with nonbonded bis-cyanopropyl polysiloxane phase column was used. Final temperature was run at 260 with nitrogen as the carrier gas. The GC-FID was then turned on and allowed to stand for 20 minutes to stabilize. A standard of fatty acid methyl ester FAME was run. For the moringa oil, 1 μL of the sample was injected. The FAME standard was then integrated as basis of the fatty acid methyl ester peaks.

Determination of phytosterols in Moringa oil
Phytosterols in moringa oils were determined using derivatization to silyl ester derivatives of the unsaponifiable matter and through gas chromatography-mass spectrometry GC-MS based on a revised method of Wretensjö and Karlberg 25 . For sample preparation, moringa oil 10 g was mixed with 100 μL of 100 ppm α-cholestanol. The sample was saponified using 2 M ethanolic potassium hydroxide. Sample was then boiled for 1.5 h. The sample was cooled and then extracted thrice using 50 mL hexane. The organic layer was collected and then dried using rotary evaporation. The remaining organic layer was further dried using nitrogen gas. Acetone 5 mL was added, and the sample was dried again, leaving the unsaponifiable matter. For derivatization to silyl ester derivatives, hexamethyldisilazane 100 μL and trimethylchlorosilane 100 μL were added to the unsaponifiable matter. Mixture was subjected to water bath 60 for 1 h. Finally, mixture was cooled down and diluted using hexane. The resulting solution was used for GC-MS.
For GC-MS, Shimadzu GCMS-QP2010 Ultra gas chromatograph was used. The gas chromatograph was fitted with an RTX-5 0.25 mm ID, 30 m . The samples were then injected with an injector temperature of 200 . The column temperature was initially held at 70 and further increased to 290 . For GC-MS, an ion source temperature of 220 was used. Phytosterols were identified by mass spectra analysis 35-750 m/z range . The quantification of sterols was done by calculating the respective areas relative to the area of α-cholestanol as the internal standard.

Physicochemical properties of Moringa oil
The specific gravity of moringa oil was obtained through AOAC 985.19 24 . The refractive index of moringa oil was obtained through AOAC 921.08 24 . Iodine value of moringa oil was measured using AOAC standard method 993.20 24 . Saponification value of moringa oil was measured using AOAC standard method 920.160 24 . Free fatty acids were measured using AOAC Official Method 940.28 24 . Lastly, color measurement was based on Phan et al. method 26 . A Minolta Chroma Meter was used to measure color of moringa oil using the CIELAB system.

Determination of antioxidant properties of oil 2.7.1 DPPH assay
In this case, 2,2-Diphenyl-1-picrylhydrazyl DPPH assay was performed based on the procedure of Pérez-Jimenez et al. with minor modifications 27 . The moringa oil was extracted using dichloromethane as the solvent. Trolox 0.01 g and ethyl alcohol 25 mL were mixed to prepare a standard solution. Aliquots of Trolox with increasing concentration 1, 2, 3, 4, and 5 mL were diluted with 25 mL of ethyl alcohol to produce Trolox solutions 90-455 μM . Afterwards, a calibration standard was prepared by dissolving 0.02 g of DPPH in 100 mL of ethyl alcohol. The Trolox solution 1 mL was then mixed with the DPPH solution 2 mL and 1 mL of the moringa oil for 10 minutes. Using ultraviolet UV -visible spectroscopy, absorbance was then measured at 515 nm. The results were then expressed as mg of Trolox per 100 g of sample.

FRAP assay
Ferric reducing ability of plasma FRAP assay was performed based on a modified procedure of Pérez-Jimenez et al. 27 . The moringa oil was extracted using dichloromethane as the solvent. A standard solution was prepared by mixing 0.01 g of Trolox and 25 mL of ethyl alcohol. Aliquots of Trolox with increasing concentration 1, 2, 3, 4, and 5 mL were diluted with 25 mL of ethyl alcohol to produce Trolox solutions 90-455 μM . Meanwhile, FRAP reagent was prepared by combining 5 mL of 2,4,6-Tri 2-pyridyl -striazine TPTZ with 5 mL of 20 mM of iron III chloride and 50 mL of acetate buffer. The prepared FRAP reagent 3 mL was then combined with Trolox solution 1 mL and with the moringa oil 1 μL at a water bath of 37 for 10 minutes. Using ultraviolet UV -visible spectroscopy, absorbance was then measured at 593 nm. The results were then expressed as mg of Trolox per 100 g of sample.

Oil stability assessment of Moringa oil
To test the oil stability of the moringa oils, the Schaal Oven Test AOCS Cg 5-97 was conducted to simulate the accelerated aging of oils 28 . A mechanical convection oven was operated and set at 62.8 for 18 days. Afterwards, a total of nine 150-mL beakers surface area 29.2 cm 2 were placed inside the oven three beakers per treatment method . Each beaker was filled with 100 grams of moringa oil and this was left exposed and uncovered. Sampling was done on the initial, 6 th , 12 th , and 18 th day and was tested in triplicate from the same sample. Peroxide value was calculated using AOAC 965.33 24 .

Data analysis
Analysis of data characterization, physicochemical, oil stability, antioxidant properties was done in triplicate. Statistical analysis of data was performed using the analysis of variance ANOVA and Duncan s multiple range test to determine significant differences between treatments p 0.05 . For oil stability, repeated measures analysis of variance was performed.

Moringa seeds characteristics
The characteristics of the obtained moringa seeds are summarized in Table 1. The kernel/seed ratio was found to be 0.735. This means that the kernel source of moringa oil approximates to about 73.50 of the entire seed in 4 terms of mass. This is consistent with the seeds from Casal et al. study 7 . The average mass of the seeds was found to be similar but slightly lighter as compared with those from Ayerza et al. study 1 . Despite the lighter mass of the seeds from Davao Oriental, the kernel/seed ratio was greater than that of Ayerza et al. study due to the lighter seed coating of the procured moringa seeds. The greater kernel/seed ratio suggests the potential of the seeds for oil extraction.
For moisture content, air-drying of seeds resulted in a moisture content of 10.500 , which is approximately near the target moisture content of 11.00 based on Fakayode and Ajav study involving mechanical extraction of moringa oil from the seeds 23 . An ideal moisture content is critical before oil expression.

Oil yield and oil ow rate
Oil extraction was performed using a screw-press oil expeller. Oil yield and flow rate from the pretreated seeds are presented in Figs. 1 and 2. Oil yield ranged from 19.92 -23.03 . Oil yield was found to be less compared with solvent extraction, which was estimated to provide an oil yield of about 35 -37 in Moringa oleifera seeds 7 . This is to be expected as solvent extraction is regarded for its efficiency in oil extraction 29 . Moreover, when comparing the results to other studies involving mechanical extraction, the yield falls on a reasonable range. For instance, Fakayode and Ajav study performed screw-press extraction of Moringa oleifera and the oil yield ranged from 11.42 -28.85 depending on the applied parameters moisture, temperature, pressure, etc. 23 . The results depict that oil yield increased with ultrasound pretreatment and significantly with microwave pretreatment 3.11 . For oil flow rate, a significant increase was observed through microwave pretreatment. The results imply that green technologies such as microwave and ultrasound are capable of increasing oil yield as well as extraction efficiency from seeds. This is likely due to the alteration of cellular structure such as the formation of pores in cell membranes that caused leaching of cellular contents for better oil expression 17 . For ultrasound pretreatment, the parameters used 50 W, 60 min were based on a pretreatment study involving mechanical expression of oilseeds. The study implied that greater power of up to 90 W and duration resulted in better yield 30 . Given that the ultrasound bath used had a maximum power of 50 W, it is suggested to further explore higher ultrasound power prior to extraction to increase oil yield.

Fatty acid pro le of Moringa oil
Fatty acid profile of the moringa oils is presented in Table 2. Based on the findings, it can be confirmed that moringa oil is a good source of monounsaturated fatty acid mostly in the form of oleic acid. Due to the greater fraction of saturated and monounsaturated fatty acids, it can be inferred that moringa oil is relatively stable against oxidation. Moreover, the percentage of oleic acid along with total monounsaturated fatty acids of the moringa oils obtained from Davao Oriental was found to be greater as compared with those from other studies 1, 7, 10, 31 . High monounsaturated fatty acid content is attributed for health benefits in reducing the risk against heart disease by reducing low-density lipoprotein, which transports cholesterol from the liver to other parts of the body 31 . Besides oleic acid, palmitoleic and eicosenoic acid were found as sources of monounsaturated fatty acid. The presence of other fatty acids, along with their compositions, was expected in comparison with Table 1 Characteristics of obtained Moringa oleifera seeds.

Physical quantity Value
Average mass of seeds (g) 0.210±0.043 Average mass of kernels (g) 0.154±0.034 Average mass of seed coat (g) 0.056±0.012 Kernel/seed ratio 0.735 Moisture content (%) 10.500±0.350  other studies 1, 7, 10, 31 . When comparing the fatty acid profile of the moringa oils from different pretreatment methods, there were some statistical differences. Nonetheless, these variations were minute and were under a percent difference. Some notable differences were the greater oleic acid percentage in the ultrasound group while less in the microwave group. The monounsaturated fatty acid eicosenoic acid compensated for the decrease in oleic acid in the moringa oil from the microwave pretreatment. Table 3 summarizes the phytosterol content and composition of the moringa oils. In terms of composition, moringa oil has a phytosterol content similar to that of Leone et al. and Özcan wherein β-sitosterol comprised the majority 44-49 of phytosterols 9, 10 . β-sitosterol is a phystosterol known to decrease systemic blood cholesterol, to exhibit antitumorigenic activity, and to encourage immunomodulation 32 34 . Besides β-sitosterol, other predominant phytosterols that were consistent with other literature were stigmasterol and campesterol 2,9,10 . Stigmasterol is known to help reduce low-density lipoprotein from the body and it is also found to have antitumorigenic effects 33,35 . Other notable differences in composition include the absence of avenasterol reported to comprise about 8 of phytosterol content in moringa oil along with seven other phytosterols, which were expected to be present in minute amounts 1 9,10 . Besides their hypocholesterolemic function, the presence of various phytosterols in moringa oil may provide antioxidant activity to the oil as certain phytosterols act as antioxidants such as campesterol, β-sitosterol, and stigmasterol 32,34,35 .

Phytosterol content of Moringa oil
In terms of total amount of phytosterols, the microwave group had the greatest amount of total phytosterols followed by the ultrasound group. The increase in phytosterols with microwave and ultrasound pretreatment suggests the disruption of cell membrane integrity of the moringa seeds prior to extraction. Phytosterols are major components of cell membranes 32 . It is likely that microwave radiation and cavitation of the sample enabled these components to break free and transfer to the oil. Based on the results, β-sitosterol and stigmasterol accounted for the greatest increase in these phytosterols. Also, it is worth looking into the absence or undetectability of stigmastanol in the control group. This is consistent with other literature that found low to nil stigmastanol in moringa oil 2,9,10 . The groups that utilized microwave and ultrasound pretreatment had significant amounts of stigmastanol. In this instance, it is likely that the green technologies have induced chemical reactions such as the hydrogenation of stigmasterol to stigmastanol 32 .

Physicochemical properties of Moringa oil
The physicochemical properties of the oils were analyzed as presented in Table 4. All physicochemical parameters fell under normal range when compared with various literature 7, 10, 36 . In terms of free fatty acids, the control group had the highest free fatty acids followed by the microwave and ultrasound group. For the control group, the longer retention time of the seeds in the oil expeller may have contributed to the greater formation of free fatty acids due to the hydrolysis of triglycerides. Table 5 depicts the analysis of color from the moringa oils. Based on the results, the microwave group had a statistical difference on all three Lab color parameters. In particular, the L* value of the microwave group was significantly greater compared with the other groups, signifying a lighter color. On the other hand, the a* value of the microwave group was slightly greater, indicative of a slightly redder color. Lastly, the b* value of the microwave group was lower as compared with the other groups, indicating a less yellow color. This is contrary to Zhong et al. study where microwave treatment increased the yellowness of the oil 22 . The greater yellow color in ultrasound and control groups may be indicative of Maillard reaction attributed to a longer retention time of processing the seeds in the oil expeller.

Antioxidant property of Moringa oil
The antioxidant properties of the moringa oils were analyzed using DPPH and FRAP assays as presented in Table  6. Based on the results, the microwave and ultrasound   37 . The DPPH assay is considered to be accurate, valid, responsive to both hydrophilic and lipophilic antioxidants, and sensitive even to weak antioxidants 38 . The microwave and ultrasound treatment may have significantly greater DPPH antioxidant activity due to the greater amounts of phytosterols. For instance, β-sitosterol acts as an antioxidant through its ability to scavenge radicals 39 . For ferric reducing antioxidant power FRAP , the control group had the greatest value and was statistically greater compared with the ultrasound group. The FRAP assay measures the ability of antioxidants to transfer electrons and reduce iron 40 . Considering both DPPH and FRAP assay, the microwave group may be considered to have the best overall antioxidant activity.

Oil stability of Moringa oil
The stability of the moringa oils were analyzed using the Schaal Oven Test, which was conducted at 62.8 for 18 days. Figure 3 presents the peroxide value of the moringa oils per treatment. Results show that as time progresses, the peroxide value significantly increased. This is to be expected as the elevated heat and oxidation induce chemical reactions that lead to the formation of primary oxidation products such as peroxides 30,41 . Analysis has shown that the treatment method significantly accounts for the differences in peroxide value. In particular, the control group had statistically greater peroxide value as compared with both the microwave and ultrasound treatment. This was attributed to less antioxidant activity of the control group as compared with the microwave and ultrasound groups as antioxidants help in protecting oil against oxidation 37 .
It was also notable that the changes in peroxide value per sampling time for each treatment were significant except for days 12 to 18. This was expected based on the peroxide values on these intervals. The stagnancy and decrease in peroxide values on these intervals may be indicative of other reactions occurring such as the formation of secondary oxidation products 42 .

Conclusion
The study elucidated on the use of green technologies microwave and ultrasound in extracting moringa oil using an oil expeller to improve oil yield while evaluating the consequence on oil extraction, oil quality, and bioactive content. Enhanced permeability of the cell membrane led to greater cellular leaching and an increase in oil yield in microwave group 3.11 and ultrasound group 1.24 . All oils regardless of treatment were comprised of mostly monounsaturated fatty acids mainly from oleic acid. The green technologies induced thinning of cell membrane components that enabled mass transfer of phytosterols in microwave and ultrasound treatments. The high monounsaturated fatty acid and greater amounts of phytosterols depict its potential as a functional food in reducing risk against cardiovascular disease. The presence of certain phytosterols may contribute to its antioxidant activity. Greater antioxidant activity found in the microwave treatment resulted in better oil stability. Mechanical extraction of moringa seeds, coupled with microwave and ultrasound pretreatment, may be exploited for their use in improving yield and quality of oils while preventing dependency on solvents for mechanical extraction.

Acknowledgements
The corresponding author acknowledges the scholarship CC BY-SA 4.0 Attribution-ShareAlike 4.0 International . This license allows users to share and adapt an article, even commercially, as long as appropriate credit is given and the distribution of derivative works is under the same license as the original. That is, this license lets others copy, distribute, modify and reproduce the Article, provided the original source and Authors are credited under the same license as the original.