The plant material consists of palm, soy, peanut and sunflower oil. The choice of

frying oils is based on their physicochemical characteristics and their common use. Figures 1-5 show the samples of peanut, soybean, palm, sunflower oils and frozen potato, respectively.

Frying without the incorporation of fresh oil was applied. This process causes the evaporation of water present on the surface of the food, thus promoting its dehydration and the formation of a crust [11] .

A series of eight fryings at 180˚C/10min, with an initial volume of oil used of 2 liters and 400 g of frozen potato.

The different oil samples analyzed are collected after each 10-minute series of frying.

The determination of the acid number is based on the official method of the Association of Official Analytical Chemists (AOAC) and ISO 660:2020, which describes methods for titration of free fatty acids in oils and fats [12] .

The acid index (AI) is the number of milligrams of potassium hydroxide required to neutralize the free fatty acids present in one gram of fat [13] . This index measures the quantity of free fatty acids resulting from the hydrolysis and oxidation reactions of triglycerides [14] . The acid index is determined by calculation, using the formula:

$\text{AI}=\frac{56.1\times N\left(\text{NaOH}\right)\times V\left(\text{NaOH}\right)}{m\left(\text{prise}\text{d'essai}\right)}$ (1)

The reference peroxide index was determined by the standardized method NF T60-220 [15] . The principle of this method is based on the titration by a sodium thiosulfate solution of the iodine molecules released by oxidation of the iodides by the hydroperoxides of the sample solubilized in an acetic acid/chloroform mixture.

The results are expressed in Meq·O₂·kg⁻¹.

$\text{Ip}=\frac{10\times N\left(V_1-V_0\right)}{m}$ (2)

The color of nectar samples was measured using a colorimeter (type: KONICA MINOLTA, Japan) based on the CIELAB color system (L^*, a^*, b^* and L^*, C^*, h, YI). The color parameters (L^*, a^*, b^* and L^*, C^*, h, YI) were measured 3 times for each sample. L^*, a^*, b^* describe black-white, Green-Red and Blue-Yellow colors respectively (−b = Blue, +b = Yellow) ( Figure 6 ) [16] . The brownness index (BI) is calculated using the following formula: a positive value indicates a tendency to yellow, while a negative value indicates a tendency to blue the sample. The BI is given by the following formula:

$\text{BI}=\frac{100\left(X-0.31\right)}{0.172}$ $X=\frac{\left(a^{*}+1.75L^{*}\right)}{(5.645L^{*}+a-3.012b^{*}}$ (3)

Statistical analyses were performed using one-way ANOVA with R software version 3.2.4 Revised (2018-03-16, R-70336). Analyses were repeated three times per sample for each frying series. The X value of each sample is assigned a superscript letter (X(i) where i = a, b, c, …). Samples with the same letter are not statistically different at the 5% level.

The steps of frying are illustrated in Figure 7(a)-(c) .

Figure 8 and Figure 9 show the oil and fried potato samples respectively after each 10-minute frying at 180˚C.

The acidity index (AI) is an indicator of the degradation of oil during cooking. Table 1 shows the acidity index values of each oil sample before the series of fryings at 180˚C for 10 minutes.

After the first frying, the results show an increase in the acidity index in each type of oil due to the breakdown of triglycerides ( Figure 10 ).

The evolution of the acid number of oils during frying cycles highlights significant variations depending on the type of oil. From the first frying, an increase is observed for all the oils analyzed. Crude peanut oil, for example, sees its acid number increase from 0.973 ± 0.003 mg·KOH·g⁻¹ to 1.119 ± 0.019 mg·KOH·g⁻¹, an increase of 15% compared to the control. Palm oil records an even more marked increase, going from 0.251 ± 0.003 mg·KOH·g⁻¹ to 0.446 ± 0.015 mg·KOH·g⁻¹, an increase of 77%. For soybean oil, the acid number increases from 0.111 ± 0.003 mg·KOH·g⁻¹ to 0.139 ± 0.009 mg·KOH·g⁻¹ (25%), while sunflower oil increases from 0.388 ± 0.003 mg·KOH·g⁻¹ to 0.474 ± 0.009 mg·KOH·g⁻¹, a 22% increase.

After the first round of frying, only crude peanut oil exceeded the standard of 0.60 mg·KOH·g⁻¹ set for refined oils. However, after eight frying cycles, all the oils studied saw their acid index exceed this reference value. Sunflower oil reached 1.194 mg·KOH·g⁻¹, while soybean oil recorded 1.032 mg·KOH·g⁻¹. For refined palm oil, the value climbed to 1.179 mg·KOH·g⁻¹, an increase that is consistent with the results obtained by Ndjouenkeu et al. (2002), who measured 1.50 mg·KOH·g⁻¹ after 120 minutes of frying at 180˚C [17] . Peanut oil shows the highest increase, reaching 1985 mg·KOH·g⁻¹ after eight fryings.

These values are significantly higher than those reported by Yilmaz et al. (2023), who measured 0.063 mg·KOH·g⁻¹ on five brands of oils under similar conditions [18] . Similarly, the indices recorded in this study exceed those obtained by Khelifa (2017) on the mixed oil “Elio” (80% soybean, 20% sunflower), which had an index of 0.735 mg·KOH·g⁻¹ after 8 hours of heating [19] .

According to Dobarganes and Márquez-Ruiz (2015), free fatty acids can, under the influence of high temperatures, be transformed into glycerols, which in turn lead to the formation of a toxic aldehyde called acrolein [20] . The acidification of a frying oil depends on many factors, including storage conditions before refining, where crude oils are subjected to prolonged temperature variations in sheds or ship holds [21] . This phenomenon accelerates autoxidation, the cause of oxidative rancidity responsible for the loss of quality of oils during storage.

Peroxide value is a parameter used to assess the oxidative stability of oils. Oxidation of oils during high-temperature frying is an unavoidable process that can affect oil quality and food safety. Table 2 shows the peroxide value of each oil sample before the frying series at 180˚C for 10 minutes.

After the first frying, the results reveal a very rapid increase in peroxide indices in all oil samples, as shown in Figure 11 .

The increase in the peroxide index (PI) during frying cycles is mainly due to the appearance of peroxidized compounds (ROO˚ peroxides) formed from unstable free radicals (R˚) of unsaturated fatty acids, as highlighted by [3] . This chemical transformation significantly affects the stability of oils used in frying.

In the case of peanut oil, the peroxide value after the first frying is 4.8 times higher than its initial value, and after eight fryings, it reaches 6.452 Meq·O₂·kg⁻¹, which is 8.8 times the control value of 0.730 Meq·O₂·kg⁻¹. However, this value remains below the maximum standard of 10 Meq·O₂·kg⁻¹.

For palm and soybean oils, the effect of frying is even more pronounced at the eighth frying, their peroxide values are 15 times higher than those of the control values. As for sunflower oil, its peroxide value increases from 0.495 to 4.489 Meq·O₂·kg⁻¹, an increase of 9 times compared to the initial value.

Sunflower oil is renowned for its superior food quality, offering a lemon-yellow color, a clear texture, and a mild, pleasant flavor [22] . It is particularly rich in linoleic acid (C18:2) and virtually free of linolenic acid (C18:3) [23] . However, due to its high level of unsaturation, it is very sensitive to heat and should not be heated above 180˚C or for a prolonged time [24] . Under the influence of heat, oxygen, and water, the oil undergoes complex chemical reactions that alter its quality and nutritional properties [25] . During frying, the degradation of fatty acids leads to the formation of products that are harmful to health, such as free radicals and aldehydes [26] . Additionally, polar compounds from these altered oils can cause liver and kidney enlargement, due to their involvement in detoxification mechanisms [27] .

The brown color index is a measure to assess the brown tint of edible oils after several frying cycles. It reflects the presence of oxidized and polymerization compounds that form under the effect of heat. The evolution of the brown index over frying reveals distinct behaviors depending on the type of oil ( Figure 12 ).

The brown index decreases sharply for peanut oil (from 61.55 to 11.28), reflecting a progressive degradation of the compounds responsible for browning, probably due to oxidation or loss of initial pigments, while palm oil shows a notable reduction (from 24.74 to 10.32), indicating relative stability despite a progressive degradation of the colored constituents; on the other hand, soybean oil shows a significant increase (from 10.00 to 28.45), revealing an accumulation of thermal and oxidative degradation products, while sunflower oil, although subject to progressive oxidation, experiences a more moderate increase (from 2.98 to 10.86), less pronounced than that of soybean oil. According to Kpata-Konan et al. (2020), the evolution of the physicochemical indices of frying oils shows that oils rich in unsaturated fatty acids (such as soybean and sunflower oil) tend to brown more after several frying cycles [28] .

When frying at high temperature (180˚C/10 min), the yellow color given by parameter (b) evolved for all oils ( Figure 13 ).

Analysis of the results shows significant variations in the yellow color index depending on the oils and the number of frying at 180˚C.

The yellowness index drops significantly from 50 to 12.34 after eight fryings in peanut oil. This suggests a significant degradation of yellow pigments under the effect of heat and oxidation. For palm oil, a decrease is also observed, from 30 to 11, which indicates a loss of yellow color, probably due to the degradation of carotenoids naturally present in palm oil. In sunflower oil, unlike the previous oils, the yellowness index increases, going from 4.29 to 12.13. This change could be due to the formation of new colored compounds under the effect of heat, resulting from the degradation of fatty acids or Maillard reactions. The increase is even more marked with soybean oil, going from 12.85 to 27.49. This increase suggests that frying causes a chemical change leading to the formation of additional yellow pigments. Several studies converge to demonstrate that repeated exposure of oils to high temperatures, particularly in frying processes, induces notable physicochemical alterations, manifesting mainly through changes in color. Renard highlights that chromatic variations constitute a reliable indicator of the thermal degradation of oils [29] . In a complementary manner, Schlienger emphasizes that the reuse of oils and the prolonged action of heat lead to a degradation of lipids, perceptible through organoleptic changes, particularly in color and odor [30] . Fatima et al. confirm the relevance of color as a key parameter for assessing the state of degradation of oils [31] . Jurid et al. (2020) report that refined palm oil, after several frying cycles, shows a significant increase in viscosity, marked darkening, and an accumulation of oxidized compounds [32] . Finally, Dodoo et al. (2022), through a colorimetric analysis of the oils used for frying sweet potato fries, observed a progressive blackening after five to six cycles of use [33] . Darkening and the formation of oxidized compounds indicate degradation, which makes monitoring the visual characteristics of oils crucial.