Coconut oil and flaxseed oil blends were prepared in different ratios on weight basis (95:5, 90:10, 85:15, 80:20, and 75:25). Oil samples were weighed and thoroughly mixed using a magnetic stirrer (witeg MSH-30A) for 30 minutes, flushed with nitrogen and stored at -20 °C until analysis within a week.

The storage study was performed according to the AOCS Recommended Practice Cg 5-97 (AOCS, 2009a). The same amount of sample (50 g) was taken into sample bottles (100 mL, clear glass bottle, O.D. 51.7 mm × H. 94.5 mm) loosely capped and stored at 65 °C for 14 days in an incubator (Memmert IN55plus) provided with ventilation. Samples were taken on the 1^st, 3^rd, 7^th, and 14^th days of storage and analyzed.

Fresh potatoes were cut into uniformly-sized pieces (3cm ×1cm ×1cm) using a mechanical slicer and fried in the oils and their blends continuously for four hours, and the oil samples were collected for analysis in every one-hour interval. Each hour is considered as one frying cycle. Samples were collected up to the 4^th frying cycle. The temperature was maintained at 180 ± 5 °C throughout the frying and the ratio of oil: potato was kept constant.

The AOCS method was used to determine the acid value of the samples (AOCS, 2009b). About 4 g of the sample was weighed precisely into a clean dry Erlenmeyer flask and 25 mL of ethanol were added. The flask was then shaken to dissolve the oil. Then the contents were boiled in a water bath for 30 min (Witeg-digital control system). The hot mixture was titrated with a 0.01 N ethanolic KOH solution with phenolphthalein as an indicator until the pink color persisted for 30 seconds.

The iodine value was determined according to the AOAC official method 993.2 (AOAC, 2000a). About 0.2 g sample were taken into a glass conical flask with a stopper and dissolved completely in chloroform. Then Wijs reagent (25 mL) was added and kept in the dark for 30 min followed by the addition of 10 mL of 15% potassium iodide solution. Immediately, the mixture was titrated with a 0.01N sodium thiosulfate solution with starch indicator while shaking vigorously until the color disappeared. A blank was also carried out in parallel to calculate the iodine value.

The peroxide value was determined according to the AOCS SURPLUS Method Cd 8-53 (AOCS, 2017). About 5 g of sample was taken into a 250-mL Erlenmeyer flask and it was closed immediately with a glass stopper. Then, 30 mL of a mixture of glacial acetic acid and chloroform (3:2, v/v) were added. After that, 0.5 mL of potassium iodide solution (saturated) was added and shaken well. The flask was then stored in the dark for 1 minute. 30 mL of distilled water were added and shaken well and titrated with 0.01 N sodium thiosulfate solution using a starch indicator. A blank test also was carried out in parallel.

The saponification value of the samples was determined according to the AOAC method (AOAC, 2000b). About 3 g oil sample was dissolved in a 0.5 N ethanolic KOH solution and the mixture was refluxed for complete saponification for 2 h. Then, the solution was titrated with 1 N sulfuric acid solution using phenolphthalein as an indicator.

Fatty Acid Methyl Esters (FAMEs) of the oil samples were prepared by transesterification (WHO, 2020). FAMEs were analyzed by Gas Chromatography (GC). About 50 mg oil was dissolved in 2 mL toluene in a screw-capped glass test tube and 1 mL of 7% BF₃ in methanol was added. Then the tube was heated at 100 °C for 45 minutes in an oven with occasional shaking. After 45 minutes, the tube was removed from the oven and allowed to cool to room temperature. Then 5 mL of distilled water, 1 mL of hexane, and 1 g of sodium sulfate were added and shaken well and left for the hexane to separate. The separated hexane layer was carefully transferred into the small vial. Finally, the vial with FAMEs was sealed with parafilm and stored at -20 °C until analysis by GC.

FAMEs were analyzed by injecting 1µl sample into the GC (Agilent technologies 7890B GC system), equipped with a flame ionization detector (FID) and a fused silica capillary column (100 m length, 0.250 mm diameter, 0.20 µm film) (SP®-2560 Capillary GC Column) using nitrogen as carrier gas (flow rate - 20 mL/min). The split ratio was 30:1 and the injector and detector temperatures were 220 and 260 °C, respectively. The initial column oven temperature was maintained at 140 °C for 5 minutes and increased to 240 °C at the rate of 4 °C/min, then maintained at that temperature for 10 minutes. Fatty acids were identified using authentic standards (FAME 37 Component Mix) by comparing their retention times (Sivakanthan et al., 2019).

The acid value is a measure of rancidity as free fatty acids are generated during lipid oxidation and hydrolysis. The higher the acid value, the higher the amount of free fatty acids, which is interpreted as decreased oil quality (Choudhary, 2013). Changes in acid value during the storage period are presented in Figure 1a. The acid value for oils and their blends increased with storage. There was a significant difference between coconut oil and flaxseed oil (p < 0.05). However, there was no significant difference between blend 1 and 2, nor was blend 5 significantly different (p > 0.05). The results obtained were in accordance with Ngassapa et al. (2017), who reported that there was an increment in the acid value for oils and their blends stored at air-free ambient temperature compared to oil samples stored under air-tight dark conditions for 60 days. The initial acid value for flaxseed oil was lower than coconut oil. However, throughout the storage period, the acid value for flaxseed oil was higher than coconut oil. The blends also presented increased acid values but were lower than flaxseed oil. This observation could be attributed to the higher degree of unsaturation in flaxseed oil compared to coconut oil. A study evaluated the stability of oil blends, and found that the blend of rice bran oil with sesame oil had lower free fatty acid content than pure sesame oil throughout the twelve-month storage period (Gulla and Waghray, 2011).

The iodine value is the indicator of the degree of unsaturation. Unsaturated fatty acids are more susceptible to oxidation. Therefore, there will be a decrease in the degree of unsaturation with the progress of oxidation. The iodine value for flaxseed oil was higher than coconut oil because of the higher unsaturation of flaxseed oil than coconut oil. The iodine values for oils and their blends decreased with storage time, which was attributed to the oxidation of unsaturated fatty acids. The initial iodine values for coconut oil and flaxseed oil were, 7.08 ± 0.23 and 190.03 ± 0.45, respectively. The initial iodine value for oil blends increased with an increasing proportion of flaxseed oil because of the higher unsaturation of flaxseed oil. However, throughout the storage period, oil blends exhibited a downward trend (Figure 1b). The iodine values for coconut oil, blend 1, and blend 2 were not significantly different (p > 0.05) throughout storage, whereas other blends and flaxseed oil showed significantly different (p < 0.05) iodine values. Anwar et al. (2007) reported that a blend of 0-80% Moringaoleifera oil with sunflower oil and soybean oil resulted in decline in the iodine value as a function of storage time. The slow decline in the iodine value of oil blends could be attributed to the induction period where fat is oxidized, slowly demonstrating the initiation of the oxidation reaction. The rapid alterations in the iodine value for oil blends may be due to the propagation of the oxidation process where hydroperoxides are formed from free radicals in fatty acids generated in the initiation stage of the oxidation reaction (Gulla and Waghray, 2011).

Peroxide value is denoted as an important indicator of oil stability (Bhardwaj et al., 2015). Peroxides are the primary oxidation products generated in the early stages of the oxidation of lipids, and they contribute to the successive oxidation reactions to form a range of nonvolatile and volatile secondary oxidative compounds (Shahidi et al., 2017). The peroxide values for both oils and their blends increased with an increasing storage period (Figure 1c). There was no significant difference between the peroxide values for blends 1, 2, and 3, whereas blend 4 and blend 5 differed significantly (p < 0.05). The results obtained were in accordance with Siddique et al. (2010), who reported that peroxide value increased with storage time. Further, after 14 days of accelerated storage, flaxseed oil showed a significantly (p < 0.05) higher peroxide value (3.79 ± 0.01 meq O₂/kg) compared to coconut oil (2.19 ± 0.01 meq O₂/kg), and blends obtained lower peroxide values than pure flaxseed oil. The higher peroxide values for flaxseed oil and the blends with a high proportion of flaxseed oil are due to the higher degree of unsaturation of flaxseed oil than coconut oil. A study by Bhardwaj et al. (2015) on enhancing the oxidative stability of flaxseed oil by blending it with palm oil revealed that blending flaxseed oil with 60% palm oil was significantly stable throughout the storage period compared to blending flaxseed oil with 40% palm oil because of the oxidative stability of palm oil. A study by Hamed and Abo-Elwafa (2012) reported that blending flaxseed oil with Nigella seed oil inhibited hydroperoxide formation by 31.71% (relative to flaxseed oil) compared to that of a blend with sesame oil which showed only 9.75% hydroperoxide formation inhibition. This means blending Nigella seed oil with flaxseed oil reduced the peroxide value of flaxseed oil compared to blending with sesame oil.

The saponification value is a measure of the average molecular weight of all the fatty acids present. The higher the saponification value, the shorter the fatty acids on the glycerol backbone (Marina et al., 2009). The initial saponification value for coconut oil was 249.33 ± 0.58 mg KOH/g and flaxseed oil was 188.24 ± 0.01 mg KOH/g. This indicates that coconut oil has a higher amount of short chain fatty acids than flaxseed oil. The initial saponification value of oil blends decreased with an increased proportion of flaxseed oil. The saponification value for oils and their blends showed an increasing trend with storage time (Figure 1d). Blends and oil exhibited significant differences (p < 0.05), although no significant differences were found between blends 2 and 3 (p > 0.05). An increase in the saponification value of the samples with storage indicates a decrease in the chain length of fatty acids, possibly due to the breakdown of fatty acids during oxidation. The results obtained were in accordance with Siddique et al. (2010), who reported that the saponification value for oil samples increased with storage time. This trend explains that with the extended storage of these oils, free fatty acids are formed, which increases the saponification value.

The acid values for oils and their blends were significantly increased throughout frying (Figure 2a). After the 4^th frying cycle, except for blends 1 and 3, all other samples showed a significant increase in acid value (p < 0.05). During frying, both oxidation and hydrolysis could lead to the formation of free fatty acids. Che and Wan (1998) also found that the acid value increased throughout frying for palm oil and coconut oil blends.

The iodine value for oils and their blends showed a downward trend (Figure 2b). Other than coconut oil and blend 1, the iodine values for all samples differed significantly (p < 0.05). This indicates the progress of oxidation and coconut oil and blend 1 showed higher stability than the other blends and flaxseed oil. This is evident from the non-significant changes in the fatty acid profile before and after frying for these two samples. Alireza et al. (2010) found a similar decreasing trend in the iodine value for oils and their blends during deep frying and reported that the decrease in the iodine value is due to the decrease in double bonds as oil becomes oxidized. Che and Wan (1998) also reported the same downward trend in the iodine value for palm oil and coconut oil blends.

The peroxide value was increased with frying cycles (Figure 2c). The values for blends 1 and 4 were not significantly different (p > 0.05). However, other samples were significantly differed (p < 0.05). After the 4^th cycle of frying, flaxseed oil exhibited the highest peroxide value (3.90 ± 0.15 meq O₂ /kg), while coconut oil had a peroxide value of 1.99 ± 0.001 meq O₂/kg. Blend 1 exhibited a higher peroxide value (2.80 ± 0.00 meq O₂/kg) than that of blend 2 (2.22 ± 0.03 meq O₂ /kg). However, blend 5 obtained the highest peroxide value (2.94 ± 0.07 meq O₂/kg) among all the blends studied. Blends 3 and 4 manifested similar peroxide values after the 4^th frying cycle. This observation also can be explained based on the degree of unsaturation. Flaxseed oil and blend 5, which consisted of a high amount of flaxseed oil, presented more extensive oxidation than the other samples. Farhoosh et al. (2009) observed a similar increasing trend in peroxide value for canola, olive, palm olein, and corn oils, and the blending of these four oils showed a lower peroxide value compared to pure oils. Che and Wan (1998) also found the same scenario in the peroxide value pattern and stated the same fact.

The saponification value also increased throughout frying (Figure 2d). However, the values for blends 2 and 3 were not significantly different (p > 0.05). Flaxseed oil showed a higher increment compared to coconut oil (from 188.71 ± 0.66 mg KOH/g to 200.22± 0.19 mg KOH/g) and showed a significant increase in saponification value (p < 0.05). An increase in the saponification value indicates the formation of short-chain fatty acids due to the oxidative degradation of fatty acids. Goburdhun and Jhurree. (1995) observed an increasing trend in the saponification value of soybean oil blended with palm kernel olein.