Seeds dimension. Large, medium, and small diameters of seeds were measured with a caliper with an accuracy of 0.01 mm. For this purpose, 75 seeds were randomly selected, and their three main dimensions were measured. Then, the measured data were reported as the length, width, and thickness of the samples, for which the arithmetic mean of the diameter or dimensions of the object and the geometric mean of the dimensions were calculated (Razavi et al., 2007).

Bulk density. A measuring cylinder (100 mL) was used to measure the bulk density. The seeds were dropped into the cylinder until a certain volume of the cylinder was filled by the MP seeds. The seeds were weighed and the bulk density of the samples was determined according to Eq. 1:

Where ρ b is the bulk density; M is the mass of seeds (in grams), and V b is the volume (mL) of the graduated cylinder.

Absolute density. The principle of liquid displacement was used to measure the absolute density of the seeds, which used Archimedes’ law of buoyancy and the absolute density of the seeds was calculated according to the Eq. 2 and Eq. 3 (Mohsenin, 2020).

Where MT: toluene weight, MP: weight of the blank graduated cylinder, MPTS: weight of graduated cylinder with MP seeds and toluene, MPS: weight of the graduated cylinder with MP seeds, PT is toluene density at laboratory temperature, VK is MP seed volume and Kρ is seed absolute density.

Crude protein content was calculated as a mass percentage in dry matter. To calculate the content of crude protein in the dry matter, the measured nitrogen content was multiplied by the conversion factor of 6.25 (Coates, 2011).

This experiment was performed according to the ISO 5498:1981 standard for the determination of crude fiber content.

This experiment was performed according to the ISO 11085:2015 standard. The experimental method is the hydrolysis of the test sample by hydrochloric acid in the presence of ethyl alcohol and formic acid.

Five grams of the test material were burned in a furnace at 550 °C until complete combustion of the organic matter, and the rest of the material was weighed. This experiment was performed according to ISO 2171:2007 standard.

Oil extraction was performed by Soxhlet using n-hexane as the solvent. The solvent was continuously circulated between the extraction chamber and the solvent reservoir until the oil was completely extracted (Özcan et al., 2019). The obtained oil was stored in airtight, dark containers at 4 °C until analysis.

The seed oil was extracted by a mechanical press machine. Ambient temperature was controlled during oil extraction of 27 °C, helix speed was 15 rpm and oil extraction temperature was controlled at 50 °C.

The ISO 660: 2009 standard was used to measure free acidity. The sample was titrated with potassium hydroxide (0.01 N) and in the presence of phenolphthalein until the appearance of a purple color.

The ISO 3960:2017 standard was used to measure the peroxide index. The basis of oil titration with sodium thiosulfate (0.01 N) was in the presence of potassium iodide and starch adhesive reagent. Finally, the result was expressed as milliequivalents of reactive oxygen per kilogram of oil sample.

The method established in ISO 6886:2016 was used to measure oxidative stability using 743 Rancimat apparatus from Metrohm, Herisau, Switzerland (Maszewska et al., 2018). Rancimat or OSI is an oxidative stability-measuring device. A sample of 2.50 g ± 0.01 g was used for each run. All the samples were assessed at the temperatures of 120 °C under a constant airflow (20 L/h).

The iodine index of oil samples was calculated from the percentage of fatty acids obtained by gas chromatography according to AOCS standard cd 1c-85 (Maszewska et al., 2018).

This index indicates the oxidation rate of oils based on the percentage of unsaturated fatty acid composition and is calculated according to Eq. 5 (Fatemi and Yang, 1998).

The AOCS standard number cc 9a-48 was used to determine the oil smoke point using YD-1 fully automatic Oil Smoke Point instrument. A sample of each oil was heated in a cup until a continuous bluish smoke appeared.

This experiment was performed according to the AOCS standard number CC-Loa-25 (Wang et al., 2018). The weight of a certain volume of sample was measured at 25 °C relative to the water of its entire volume at 25 °C. For this purpose, a 50 mL glass pycnometer with a sealed lid was employed.

The ISO 6320: 2000 standard was used in this experiment, and a digital refractometer was used, after first cleaning and then calibrating the lens using distilled water. Then, a sample of oil was poured onto the desired part of the device and the refractive index was recorded.

To determine fatty acids, the sample was prepared as a methyl-ester derivative. The oil samples were esterified with 2N methanolic potassium hydroxide for 30 minutes at 50 °C and analyzed using gas chromatography based on ISO 24363: 2023. The gas chromatograph was equipped with a flame ionization detector (FID) and a hydrogen flow rate of 40 mL/min and an airflow rate of 450 mL/min. An HP-88 fused silica capillary column (length:100 m; internal diameter: 0.25 mm, film thickness: 0.2 µm) was employed. Nitrogen was used as the carrier gas (flow rate: 2 mL/min). The oven temperature was initially maintained at 120 °C for one minute. It was then increased to 175 °C at a rate of 10 °C per minute and held at this temperature for 10 minutes. Subsequently, the temperature was raised to 210 °C at a rate of 5 °C per minute and maintained for 5 minutes. Finally, the temperature was increased to 230 °C at a rate of 5 °C per minute and held constant for 5 minutes.

Sterol compounds were determined by the IOC standard (COT/T.20/DOC NO.10/REV.1.2001). Unsaponifiable compounds were obtained through saponification with an ethanolic potassium hydroxide solution, followed by extraction with diethyl ether. These compounds were then spotted onto a thin-layer chromatography (TLC) plate, and the different components were separated and identified using a rhodamine detector at a concentration of 0.01 % in ethanol. The sterol-related fraction was isolated from the TLC plate, and sterols were extracted using chloroform. The analysis was performed using an Agilent model 7890A gas chromatograph equipped with FID. The chromatograph operated with a hydrogen flow rate of 40 mL/min, an airflow rate of 450 mL/min, and an HP-88 fused silica capillary column (length: 100 m, internal diameter: 0.25 mm, film thickness: 0.2 µm). The split ratio was set at 1:50. Nitrogen was used as the carrier gas (flow rate: 2 mL/min). The oven temperature was initially maintained at 120 °C for one minute. It was then increased to 175 °C at a rate of 10 °C per minute and held at this temperature for 10 minutes. Subsequently, the temperature was raised to 210 °C at a rate of 5 °C per minute and maintained for 5 minutes. Finally, the temperature was increased to 230 °C at a rate of 5 °C per minute and held constant for 5 minutes.

Tocopherol analysis was performed using HPLC according to the ISO 9936:2016 standard. The mobile phase used methanol–acetonitrile (30:70 v/v) solvent. Detection was achieved using a UV-Visible detector. Single-tocopherols were isolated, and their concentrations were determined using calibration curves.

The antioxidant activity was determined using the ABTS free radical scavenging method. ABTS radical was obtained by exposing potassium persulfate for 16 h. The absorbance at a wavelength of 734 nm in the presence of oil for 5 min was determined by ELISA-Reader (Re et al., 1999).

The Chan et al. (2010) method was used to measure the total phenol content by the Folin-Ciocalteu reagent. The results were analyzed using linear regression from the standard gallic acid curve and the equivalent concentration of gallic acid was determined (Chan et al., 2010).

The mean of the data obtained from the measurement of the physical characteristics of 75 MP seeds used in the present study was calculated by a caliper. Arithmetic mean diameter and geometric mean diameter were calculated using the results of measuring the large, medium, and small diameters of samples (in three dimensions). The results of these measurements are shown in Table 1.

The values of 100-seed weight, 33-seed weight, bulk density, absolute density, and porosity were calculated with 7 replications and are shown in Table 1.

The gravitational characteristics of MP seeds are important in the processes of transfer, washing, processing, oil extraction, and storage of this product. Designing equipment for harvesting, transferring, sorting, grading, processing, and other industrial processes requires information about the physical and geometric characteristics of the seeds.

The porosity of the crops used in this study was less than the porosity rate reported (41.85% vs. 45.4%). This difference can be due to differences in the growing area of the two plants, crop moisture content, and test error.

The dry matter percentage, protein, fiber, lipid, and ash content obtained from the analysis of MP seeds are presented in Table 1. The lipid content in MP seeds of 41.12 % was the highest composition in MP seeds based on dry matter. MP seeds harvested from the Kohat forests showed protein, fiber, moisture, and ash content of 31.65, 7.54, 8.90, and 6.53 %, respectively. The high percentage of protein in MP seeds makes them a valuable nutrient in the diet. Recommended Dietary Allowance (RDA) for a person weighing 70 kg should contain 56 g of protein. For adults weighing 50 kg, the value is 46 g/day. Most of the fiber in food is in the form of cellulose and lignin. Water-insoluble carbohydrates, such as fiber, are separated by various methods (Whitney and Rolfes, 2018). Raw fiber weight has no significant nutritional value, it plays a major role in facilitating bowel movements. The percentage of fiber in the seeds was higher than in the leaves of MP (p < 0.05).

Another study reported MP seed compounds native to Sudan with a moisture content of 2.9 % and the content of protein, ash, lipid, and carbohydrate equal to 30.16, 2.60, 77.40, and 57.23 %, respectively. The lipid and ash contents measured in the present study were consistent with the values reported by (Sulaiman et al., 2017). Scientists reported that the dry matter content in the leaves and seeds of MP was 98.4 and 97.8 %, respectively, which was approximately equal to the contents measured in the present study. The protein content in the leaves and seeds of MP was 8 and 12 %, respectively, which was less than that of the present study. The ash content in the leaves and seeds of MP measured by these researchers was 10.3 and 6.1 %, respectively, higher than the content measured in the present study. Differences in the measured content of compounds depend on many factors, including differences in a plant growth area, accuracy of measurement methods, and test error.

Ash usually indicates the number of minerals in the sample. The ash content in natural foods is about 5%, which can be more than 10% for processed foods. Minerals are usually required in very small amounts for the human body. The high ash content in food indicates very low food quality, while the low ash content indicates high food quality (Pomeranz, 2013). This study revealed that the seeds and leaves of MP as natural food had acceptable levels of ash. The ash content in MP leaves was higher than in its seeds (p < 0.05). The percentage of lipids or oil in MP seeds was about 10 times higher than the oil in its leaves, which showed a significant difference (p < 0.05). Somali et al., (1984) expressed the values of compounds in MP seeds as 54.3 % lipid, 22.1 % protein, 3.6 % fiber, and 2.5 % ash (Somali et al., 1984). These values were measured with seed moisture content of 1.8 %. The ash content measured in the present study was lower than the value reported by these researchers. Lipid, protein, and fiber contents were much higher than the reported data. Factors such as moisture percentage, sampling time, etc. can be the reason for the differences in the values measured in the two experiments.

Some characteristics of MPSO and MPCO are presented in Table 2; such as acidity, iodine index, and density of MP oil were 0.7 (mg KOH/g), 67.73 (gI₂/100g), and 896 (kg/m³), respectively (Salaheldeen et al., 2014). The results reported by Salaheldeen were consistent with our results.

The percentage of free fatty acids was expressed as the acid with the highest content in the sample, such as oleic acid, where the acidity of the sample was determined by dividing the acid number by 1.99. The permissible acidity limits for refined oils and virgin oils (oils extracted by cold press) were 0.3 and 2%, respectively. The peroxide number was obtained by measuring the iodine content released from potassium iodide and was expressed as meq O₂/ kg oil. The permissible numbers for peroxide in refined and virgin oils (oils extracted by cold press) are 10 and 15 meq O₂/kg, respectively (Homapour et al., 2014). MP seed oil had the standard acidity and peroxide in the permitted range. The acidity of the MPSO was higher than the MPCO. Also, as expected, the peroxide content in MPCO was higher than MPSO (p < 0.05).

Oxidative stability is the induction time, which is determined using the Rancimat device. Temperatures of 100-120 °C are commonly used to measure oxidative stability. The optimal induction time was between 6-24 h. The oxidative stability of MPSO and MPCO was 5 h and 4 h, respectively. The longer the induction period, the higher the oil quality. Low-quality oils usually have a shorter induction period. Oxidation leads to unpleasant odors and reduced food quality.

The iodine number indicates the degree of unsaturation of the lipid or oil, expressed as grams of iodine absorbed by 100 g of lipid. The iodine number is often accurate for unpaired double bonds. The lower the number of double bonds, the lower the iodine number and therefore lower oxidation. The Cox index indicates the oxidizability of oils based on the composition of unsaturated fatty acids. The Cox index was 1.15 for MPSO and 1.16 for MPCO. The lower the Cox index, the greater the resistance of the oil to oxidation. The Cox index for olive oil was reported to be 1.47 (Homapour et al., 2014). The smoke point is the temperature at which the oil begins to smoke and indicates the beginning of the breakdown of triglycerides in the presence of air. The smoke point for oils is normally in the range of 200-230 °C, which is slightly higher in real frying (outside of laboratory devices). The measured smoke point for MPSO was higher than the measured smoke point for MPCO. As the unsaturation of fatty acids and oils increases, the oil density increases, i.e., as the number of carbons in the fatty acid chain increases, the density decreases. The relative density of MPCO was higher than the relative density of MPSO, which meant a significant difference between the two extraction methods (p < 0.05).

The fatty acid profiles of MPCO and MPSO are presented in Table 3.

The highest amount of fatty acid in the sample of MPSO was related to oleic acid at 78.1 % and palmitic acid at 11.47 %. Saturated fatty acids in this oil were 15.74 % including butyric, caproic, lauric, myristic, palmitic, stearic, and arachidic acids. In addition. 84.20 % of the fatty acids in this oil were unsaturated fatty acids including palmitoleic, elaidic, oleic, linoelaidic, linoleic, α- and γ-linolenic acids.

Oleic acid is one of the ω9 fatty acids with a content of 78.01% in this oil. The ω6 fatty acid content in this oil was 1.52 %, including linoleic acid and γ-linolenic acid. α-linolenic acid was one of the ω3 fatty acids with a content of 1.60 % in this oil. The ratio of ω6 fatty acid to ω3 fatty acid was calculated to be 0.95 %.

The highest value for fatty acid in the sample of MPCO was related to oleic acid at 77.73 % and palmitic acid at 11.49 %. The saturated fatty acids in this oil were equal to 15.9% including butyric, caproic, lauric, myristic, palmitic, stearic, and arachidic acids. Moreover, 84.08% of the fatty acids in this oil were unsaturated fatty acids including palmitoleic, elaidic, oleic, linoelaidic, linoleic, α- and γ-linoleic acids. Oleic acid content was equal to 77.73 %, while the ω6 content was 1.65%, including linoleic acid and γ-linolenic acid. α-linolenic acid as ω3 fatty acids was equal to1.60 %. The ratio of ω6 fatty acid to ω3 fatty acid was calculated to be 1.03 %.

MP seed oil belongs to the oleic-linoleic acid oils among vegetable oils. The saturated fatty acid contents in oils such as corn, olive, sesame, and canola oils was reported to be less than 20 % (Pantazaki et al., 2010; Kurt, 2018).

In one study, the fatty acid content in MPSO was reported to be oleic acid at 72.19 %, palmitic acid at 9.9 %, saturated fatty acids at 18.78 %, and unsaturated fatty acids at 81.22 %. The results reported in previous studies were consistent with the results obtained in this study (Salaheldeen et al., 2014). According to the results reported by Anwar and Rashid (2007), the content and type of fatty acids in MP native to Iran measured in this study were close to the content and type of fatty acids in M. oleifera oil native to Pakistan (76 % oleic acid, 6.5 % palmitic acid) (Anwar and Rashid, 2007).

Sterols exist in all oils and are the most important components of substances which are non-saponifiable, neutral crystalline alcohols with a high melting point. The most abundant plant sterol (phytosterol) is β-sitosterol, which makes up the majority of sterols along with campesterol, stigmasterol, and Δ⁵-onasterol. Generally, phytosterols reduce the absorption of total cholesterol, and low-density lipoprotein (LDL) and thus reduce the risk of heart disease. The sterol compounds in MPSO and MPCO are presented in Table 4.

The most abundant sterol compound in MPSO was stigmasterol with 31.75%. Δ⁵-avenasterol and campesterol are the other two major constituents of this oil with 26.30 and 17.20 %, respectively. The most abundant sterol compound in MPCO was stigmasterol with 30.73%. The content of Δ⁵-avenasterol and campesterol were equal to 26.22 and 17.08 %, respectively. The major sterols in MP oil were stigmasterol, Δ⁵-avenasterol, campesterol, and β-sitosterol, while the major sterol composition of olive oil was β-sitosterol with a concentration of 68% (Tsaknis, 1998). The sterol compounds of Pakistan M. oleifera seed oil mainly included β-sitosterol (46.16 %), stigmasterol (18.80 %), campesterol (17.95 %), and Δ⁵-avenasterol (9.26 %) (Anwar and Rashid, 2007).

Figure 1 shows the type and content of tocopherol compounds in both MPCO and MPSO.

According to the results obtained in this experiment, the highest content of tocopherols in MP oil was related to γ-tocopherol, α-tocopherol, and δ-tocopherol, respectively. The content of β-tocopherol was not detectable in the experiments. Anwar and Rashid (2007) studied the physical and chemical characteristics of seeds and seed oil of M. oleifera native to Pakistan and measured α-, γ- and δ-tocopherols in oil at 140.5, 18.68 and 61.70 mg/kg, respectively, which decreased after gumming (Anwar and Rashid, 2007). Differences in Moringa species were considered to be due to the differences in the values of other tocopherols.

Figure 2 shows the antioxidant activity of MPCO and MPSO based on the percentage of ABTS radical scavenging with 3 replicates.

The results showed that MPSO had higher ABTS radical scavenging activity (antioxidant activity) than MPCO (19.36 % vs. 12.88 %). According to the statistical analysis, the antioxidant activity of MPSO was significantly different from that of MPCO (p < 0.05).

According to the standard gallic acid curve, y=0.0012x+0.0254 (R²=0.98), equivalent concentrations of gallic acid were obtained for each sample (Table 5).

Phenolic compounds play an important role in the oxidative stability of oils. According to the results obtained from the study of phenolic content based on the equivalent concentration of gallic acid, it was found that MPSO at the concentration of 0.1 g/mL had the highest concentration of equivalent gallic acid (phenolic content) while MPCO at a concentration of 0.003125 g/mL depicted the lowest concentration equivalent to gallic acid (phenolic content). There was no significant difference between MPSO (0.1 g/mL) and MPCO (0.025 g/mL) (p < 0.05). The results for other concentrations were significantly different from each other (p < 0.05). There was a significant difference between the phenolic contents in MPSO and MPCO (p < 0.05). Phenolic compounds are sensitive to heat, so they are predicted to be destroyed during prolonged exposure to heat, such as extraction by Soxhlet; therefore, the content of total phenol in MPCO was higher than MPSO.