1 Introduction

Deep fat frying, which involves food immersed in hot oil in the temperature between 140 and 200 °C, is a widely used cooking method due to its ease and distinctive effects on food flavor and texture [1, 2]. A few amount of oil is absorbed by the food during this process, which results in the certain amount of breakdown products from the oil are obtained by the food being fried [3]. During deep frying, a number of changes like oxidation, hydrolysis, isomerization, and polymerizations occur which alter the flavor and spoil the compounds present in cooking oil, affecting both the quality of oil and fried food. Food with different proximate composition, such as vegetables, chicken, potatoes, fish, etc., tends to transfer various substances like fat, water, salt, sugar, antioxidants, etc.) into frying oil, changing the rate and type of reactions that occur in the frying oil [4, 5]. As per Codex Alimentarius Commission, FAO and WHO Food Standards, repetitive deep fat frying lowers the amount of unsaturated fatty acids in cooking oil while increasing the number of free fatty acids, foaming, color darkening, viscosity, density, and specific gravity [6]. Due to harmful substances present in the oil as well as in fried food, the degraded frying oil might lead to negative impact on human health [7]. According to Good [8], using vegetable oil for deep frying repeatedly reduces the smoke point causing oil to smoke when heated at low temperature. Further Kaleem et al. [9] reported that the poor quality cooking oil indicator is the change in color, lower smoke point and iodine level, increased free fatty acid, peroxide value, viscosity, total polar material and high foam qualities. The effect of repetitive deep fat frying on the different brands of cooking oil’s physicochemical properties was also examined by Omara et al. [10]. Finding shows that the repeated use of oil for frying potatoes can be done just for 7 times on an average for hard oils (solid at room temperature) and 6 times for soft oils (liquid at room temperature), at 140 °C for 10–20 min. In another study, frying capacity of palm olein oil and its blend containing soybean oil, canola oil and sunflower oil was investigated by Siddique et al. [11]. Involving the frying of fish, whole process took place for > 60 min in 3 stages lasting 20 min for each frying at 200 °C. Among the frying oils tested in this study, palm olein oil was shown to be better against decomposition and oxidation. Mariana et al. [12] explained that frying at high temperatures (over 180 °C) and exposure to air and water caused a variety of degradation reactions in oil that gave rise to a number of reaction chemicals.

Meanwhile, a number of researches revealed that, consuming vegetable oil that has been heated repeatedly while being used for frying has been associated with atherosclerosis risk, low density lipoprotein (LDL) levels, and total blood lipid [13,14,15,16,17]. Repeatedly used frying oil causes lipid peroxidation, resulting in malondialdehyde (a genotoxic harmful degradative byproduct of lipid peroxidation) which shows both mutagenic and carcinogenic effects in body by changing cell membrane properties [18,19,20,21]. In this concern, some in vivo findings reported by Srivastava et al. [22] investigated that after 12 weeks of feeding Wistar mice a repeatedly heated cooking oil diet regularly, there were gene alterations and pre-cancerous lesions found in their liver cells. This result is strongly supported by the research of Sutejo et al. [23] stating that feeding repeatedly heated frying oil to experimental mice for 14 days and overall 27 times caused increased cholesterol levels and liver cell damage. Furthermore Farag et al. [24] demonstrated that organ damage was inversely correlated with the length of oil heating. The longer and more frequently the frying oil was heated, the damage to the kidneys, liver and heart of the subjects also got worsted.

In India, various food items that are cooked by deep frying include Kachori, Poori, Wada, Samosa, Bread pakoda, french fries, patties, etc. The aim of this study is to ascertain how the physicochemical and antioxidant qualities of cooking oil used by local fried food sellers, changed with repeated heating, whose consumption increases the risk of various diseases such as atherosclerosis, cardiovascular disease, diabetes, cancer, etc. For this purpose, fresh, after 5th batch, and 10th batch of repeated frying, total 15 samples (200 mL each) were collected from five randomly selected street fried food vendors located at different areas of Lucknow city. Table 1 shows the maximum allowable limit for Refined Soybean Oil for all physicochemical characteristic that has been analyzed for this study [25].

Table 1 Physicochemical characteristic and maximum allowable limit for refined soybean oil
Full size table

2 Materials and methods

2.1 Sample collection

A total number of fifteen oil samples (200 mL each) were collected from five different areas namely Indira Nagar, Gomtinagar, Hazratganj, Alambagh and Telibagh, all situated in Lucknow, Uttar Pradesh. Oil samples were collected from these areas at different stages of frying (after 5th batch and the 10th batch), as well as fresh oil sample before heating (0 batch). Almost all shops were chosen as same in terms of oil type [mixture of refined soyabean oil (RSO) of Fortune brand and hydrogenated vegetable oil commonly known as dalda vanaspati ghee (10% of RSO)], food items being fried, the type of fryers, and frying conditions. The amount of break time between two successive frying cycles was 5–10 min. The duration of sample collection took 5 days. The collected oils were transferred to dark colored glass bottles. The oil samples were filtered through a filter paper and stored at 4 °C to prevent further degradation. All the analysis was performed within 5 days. This research work was conducted in the laboratory of the Department of Food & Nutrition and Department of Chemistry of Babasaheb Bhimrao Ambedkar University, Lucknow, India. All the solvents and chemicals used in this investigation were of analytical grade purchased from SRL Pvt. Ltd. Mumbai (India).

2.2 Physicochemical analysis

2.2.1 Specific gravity measurement

Using a pycnometer with a 50 mL capacity, the densities of oil samples before and after frying were determined. Oil sample was added (W1) after the empty bottle had been weighed (W0). The contents were removed, filled with water, and reweighed as W2 again. In order to calculate the specific gravity, following equation was used [1].

$$ {\text{S}}.{\text{G}}. = \frac{{{\text{w}}_{1} - {\text{w}}_{0} }}{{{\text{w}}_{2} - {\text{w}}_{0} }}. $$

2.2.2 Viscosity measurement

An Ostwald Viscometer with a technical constant of 0.05 Cs/c, ASTMAD 445 England, was used to evaluate the viscosity of oil samples both before and after frying. Oil sample flow time (measured in seconds) was noted using a stopwatch. Following equation was used to compute viscosity [1].

$$ {\text{viscosity}} = \frac{4.39 \times t}{8}, $$

where t is the flow time in seconds.

2.2.3 Saponification value measurement

The saponification value was obtained according to ASTMD464 [26]. 2 g of oil sample weighed in a conical flask, 25 mL of 0.5 N ethanolic KOH was added into the flask, and vigorously mixed. The mixture was heated at 70 °C for an hour with a condenser attached to the conical flask. Two drops of 1% phenolphthalein indicator were used to titrate the heated mixture with 0.5 N HCl, until a colorless solution was achieved. The same method was applied to determine a blank. Following equation was used to obtain the SV [1].

$$ {\text{S}}.{\text{V}}. = \frac{{{\text{A}} - {\text{B}} \times {\text{N}} \times 56.1}}{{\text{W}}}, $$

where A is the HCl (mL) for blank, B is the HCl (mL) for sample, W is the weight of sample (g), N is the normality of HCl.

2.2.4 Iodine value measurement

Hubl method was used to determine the iodine value (IV) in oil samples. The Hubl’s iodine was filled in a 50 mL burette and the initial reading was noted. 5 mL of chloroform and 2–3 drops of Hubl’s iodine were mixed thoroughly. This serves as a control for color comparison. Then 5 mL of chloroform was taken in another beaker. 0.5 mL of oil sample was mixed and gently swirled to dissolve. Hubl’s iodine was added slowly from the burette until the color of iodine (matching with control) appears in the solution. The burette reading was noted. Following equation was used to calculate the iodine value [1].

$$ {\text{IV}} = \frac{{260{\text{x}}}}{45.150}, $$

where x is the volume of Hubl’s iodine in mL.

2.2.5 Peroxide value measurement

Oil samples were examined for peroxide value using AOCS Cd 8b-90 method [2]. In 250 mL conical flask 5 g oil sample was weighed firstly. The flask was then filled with 30 mL solution of glacial acetic acid: chloroform in a ratio of 3:2 and rapidly shaken. Then 0.5 mL of potassium iodide was added into the flask and shaken for 1 min. After adding 30 mL of distilled water, the solution was titrated with 0.1 M sodium thiosulphate. The flask was shaken vigorously until the yellow color of the solution got disappeared. The color changed to blue after a few drops of 1% starch solution were added. The mixture was again shaken until the blue color got disappeared. Blank titration was done twice using the same method. The PV was calculated using the formula below:

$$ {\text{PV}} = {{\left[ {\left( {{\text{V}}_{{{\text{sample}}}} - {\text{V}}_{{{\text{blank}}}} } \right) \times {\text{N}} \times 1000} \right]} \mathord{\left/ {\vphantom {{\left[ {\left( {{\text{V}}_{{{\text{sample}}}} - {\text{V}}_{{{\text{blank}}}} } \right) \times {\text{N}} \times 1000} \right]} {\text{W}}}} \right. \kern-0pt} {\text{W}}}, $$

where Vsample is the titration volume of sample (mL); Vblank is the titration volume of blank (mL); N is the normality of potassium thiosulphate; W is the weight of the sample (g).

2.2.6 Free fatty acid measurement

Oil samples were analyzed for free fatty acids using AOCS Ca 5a-40 [2]. 10 g of oil sample was weighed in a 250 mL beaker. 60 mL of neutralized ethanol was added into it and shaken vigorously. The solution was titrated with standardized 0.025 M Sodium hydroxide using phenolphthalein indicator until the solution just turned pink. FFA was expressed as a percentage of oleic acid in the sample.

$$ \% {\text{FFA}}\;{\text{as}}\;{\text{oleic}}\;{\text{acid}} = \frac{{{\text{alkali}}\;{\text{volume}}\left( {{\text{mL}}} \right) \times {\text{alkali}}\;{\text{normality}} \times 28.2}}{{{\text{sample}}\;{\text{weight}}\left( {\text{g}} \right)}}. $$

2.3 Antioxidant analysis

2.3.1 Total phenolic content

Total phenolic content of the oil samples was measured using the Folin–Ciocalteu reagent (FCR). In a conical flask 0.3 mL of oil sample was weighed firstly. Then 3 mL of Folin–Ciocalteu reagent was added into it. 3 mL of 6% sodium carbonate was added after 5 min, and the mixture was left to stand at room temperature for 90 min. A UV/VIS Spectrophotometer was used to measure the mixture’s absorbance at 725 nm. According to the calibration curve, the total phenolic content was determined, and the results were depicted in Fig. 3a as mg of gallic acid equivalent per gram of oil (mgGAE/g oil) [17, 27].

2.3.2 Total flavonoid content

To determine total flavonoid content, the oil sample (0.5 mL) was weighed into a conical flask. 20 mL of 80% aqueous methanol was added into it. The mixture was shaken for 3 h using an orbital shaker. Then it was filtered using Whatman filter paper. After that, the filterate was evaporated for 1 h at 105 °C, followed by cooling and weighing again [17]. Results are shown in Fig. 3b as Total Flavonoid Content (%).

2.4 Statistical analysis

All physicochemical analysis was performed in three replications from each batch and the obtained data were subjected to Mean and Standard Deviation (SD) values using Microsoft Excel 2007.

3 Results and discussion

3.1 Physicochemical properties of the oils

One of the physical characteristics that govern how quickly oil or fat breaks down is color [28, 29]. In this study, as the amount of time for frying grew the color of the oils continued to become darker as shown in Fig. 1.

Fig. 1
figure 1

Oil samples of different areas; A Indira Nagar, B Gomtinagar, C Hazratganj, D Alambagh, E Telibagh. a Is indicating fresh oil sample, b is for oil sample after 5th batch fry, and c indicates oil sample after 10th batch fry

Full size image

One important parameter is specific gravity that affects the heat transmission through natural convection. Bhavsar et al. [30] suggested that specific gravity measurements are extremely helpful for determining oil contamination or adulteration, as the SG of oil is influenced by chemical composition present in the oil. An increase in the amount of aromatic compounds present in oil results in an increase in the specific gravity. Table 2, Fig. 2 show the result of the specific gravity of fresh oils, oils after fifth and tenth batches of fry that were collected and examined. Oils obtained after the tenth batch of deep frying have the highest specific gravity values, ranging from 0.916 to 0.923. While fresh oils had values in the range of 0.911 to 0.920, and values after fifth round of frying ranged from 0.914 to 0.922. Here we can observe how the SG of each oil sample changed as the frying time increased.

Table 2 Changes in physicochemical properties of oil after frying at different cycle
Full size table
Fig. 2
figure 2

Mean; a specific gravity (g/mL), b viscosity (millipoise), c free fatty acid (%), d peroxide value (g/mL), e iodine value (g/mL), f saponification value (mg KOH/g)

Full size image

Another parameter used to evaluate physical changes in oil sample is viscosity. It depends on the specific gravity, molecular weight, temperature and the degree of unsaturation [1]. The measured value of all samples is presented in Table 2; Fig. 2. Before heating (in fresh oil) the viscosity was lower than after repeated heating in all oil samples. After ten batches of repeated frying the oil’s viscosity peaked between 690.32 and 740.32. On the other hand, after five batches the viscosity was calculated between 670.23 and 720.12. The least viscous oil was fresh oil with values ranging from 638.26 to 648.32 in all oil samples. The increase in viscosity is probably because of saturation of the triglyceride chain by the process of hydrogenation while the decrease may be likely due to unsaturation of oil. The more oil is in unsaturated form (liquid at room temp.) the less its viscosity would be [31].

An indicator of the molecular weights of the triglycerides in the oil is saponification value (S.V.). It is inversely correlated with the average molecular weight or chain length of the fatty acids [32]. A low saponification value denotes longer fatty acid chain and a higher molecular weight in a sample; while a high SV indicates triacylglycerols with shorter fatty acid chains [33]. Table 2, Fig. 2 depict the S.V measured for all oil samples. It shows the fresh oil values ranging from 60.78 to 64.32 mg/KOH/g, the oil collected after the fifth batch of frying ranges from 60.82 to 64.43 mg/KOH/g, and the highest values from tenth batch of frying oils ranging from 61.92 to 65.24 mg/KOH/g. This conclusion is consistent with that of Alajtal et al. [34] who found that after frying, the saponification value of sunflower, olive oil and corn oil increased from 21.09 to 22.44 mg KOH/g, 21.54 to 22.27 mg KOH/g, and 19.41 to 21.15 mg KOH/g respectively. The increase in the SV shows that the oil has degraded and should be dumped or used to make soap instead.

Another crucial indicator of the degree of unsaturation in fatty acids is the iodine value. The potential of oil to oxidise while heating at higher temperature as in deep frying, increases with the decrease in iodine value. According to Cheplogoi et al. [35] a fall in iodine value is another sign that vegetable oils are degrading. Table 2, Fig. 2 depict the IV measured for all oil samples. It shows the values for fresh oil samples ranging from 1.199 to 1.207 g, the values for oils collected after fifth batch of frying ranging from 1.072 to 1.189 g, and the highest values for the frying oils collected after tenth batch ranging from 0.787 to 0.839 g. The observed reduction in iodine value is consistent with the reduction in double bonds that has been reported by Cheplogoi et al. and is attributed to oxidation and heat degradation. When considering the oxidative stability of cooking oils after frying food items, the iodine value is of particular importance.

Another frequent measure used to find out the oxidative stability of oils is the peroxide value (PV) [36]. The oxidation of carbon atoms adjacent to the double bonds in the triacylglyceride structure resulting in the formation of hydroperoxides, is what causes the increase in PV in oil samples after repeated frying. Therefore, the oil oxidation happens more quickly the higher the PV. The reason of decrease can be due to the peroxides produced during heating becoming unstable substances at high temperatures as well as during frying, where they turn into carbonyl and aldehyde molecules. A study done by Obeng et al. [37] suggested that continuous contact to light, oxygen, and high temperatures reacts with the oil to generate peroxides. According to Ramadan et al. [38] a good quality vegetable oil for frying should have a peroxide value of < 2 meq/kg. In our investigation, all fresh oil samples before frying have the PV below 2 meq/kg whereas all samples collected during frying operations after 5th batch and 10th batch showed the PV range from 3.6 to 17.3 meq/kg depicts in Table 2; Fig. 2. Decker et al. [39] suggests that adding antioxidants to oils may result in reduced peroxide value. Antioxidants are included to extend shelf life and stop rancidity.

Increases in free fatty acid (FFA) concentration and decreases in unsaturation of oil are indicators of the oxidation and chemical changes in oils during repeated heating [40]. A decrease in FFA may result from the loss of low molecular weight acids from the oil through volatilization process [41]. According to Gunstone [42] oils should have an FFA concentration of less than 0.1%. Table 2, Fig. 2 show that all fresh oil samples had the lowest FFA values between 0.05 and 0.08%, while oil samples from tenth batch of fry had the highest values ranging between 1.20 and 2.10%. Melton et al. [43] reported that the typical FFA concentration of used frying oil is around 4.95%, while Gupta [44] reported that the FFA concentration of absorbed frying oil ranged from 0.25 to 0.4% in the majority of snack foods. On the other hand, FFA levels in commercial fryers can go up to 0.5%, while the ideal range for fresh frying oil is between 0.03 and 0.5%. Tseng et al. [45] also suggests that if the FFA levels are greater than 1%, frying oil should be discarded. After 5 days of frying, a study done by Obeng et al. [37] likewise revealed comparable findings, with% FFA ranging from 0.62 to 2.41%. The FFA values of the frying oils differ as a result of new oil being added throughout the frying process, which reduces these values. According to Mcwilliams [46] the release of free fatty acid causes oil to smoke and develop acreolin. The acreolin can be seen as bluish and acrid smoke and its vapour can cause eye, nasal and respiratory tract irritations in low level exposure. Further Thomas et al. [47] explained that free fatty acid has a significant impact on the smoke point of oils. The faster the oil breaks down and begins to smoke, the more FFA it contains.

3.2 Antioxidant value

3.2.1 Total phenolic content (TPC)

Phenolics are known for their role in the oxidative stability of oils. Due to the frying oil's prolonged exposure to high heat during deep frying, the total phenolic content of the oil decreases during frying operations. Through quenching the radical processes that cause rancidity, their presence may prevent the deterioration of oils [48, 49]. Figure 3a displays the result of TPC of all collected oils. For fresh oil, the total phenolic content was 0.37, 0.38, 0.37, 0.35 and 0.32 mg GAE/g respectively. For fifth batch of repeated frying oil, values were 0.14, 0.15, 0.16, 0.11, and 0.09 mg GAE/g respectively. While for the tenth batch, values reported were 0.09, 0.09, 0.10, 0.08 and 0.07 mg GAE/g respectively. Result shows that all five fresh oils contained a good quantity of TPC. On the other hand, the TPC of the frying oils decreased after fifth and tenth batch of frying significantly. The findings here are consistent with Alonso et al. [4] study, where the number of deep-frying cycles and the decline in olive oil’s antioxidant activity were positively correlated.

Fig. 3
figure 3

a Total phenolic content (mgGAE/g), and b total flavonoid content (%) of oil samples collected as fresh and after frying at different batches

Full size image

3.2.2 Total flavonoid content (TFC)

Antioxidant activity is the most well-known property of flavonoids (flavonols and flavanols). Therefore, consumers and food manufacturers have developed an interest as the anti-inflammatory, anti-atherosclerotic, antioxidant, and anti-carcinogenic effects of flavonoids are well known. They are among the antioxidant defence mechanisms defending vegetable oils against oxidative deterioration [50, 51]. Figure 3b illustrates the finding of the flavonoid content of measured frying oils. Flavonoid content of all fresh oils obtained were 2.58, 2.81, 2.35, 2.79 and 2.17 respectively. After multiple batches of frying, the flavonoid content reported were 1.42, 1.96, 1.51, 1.86, 1.32 and 0.64, 0.63, 0.59, 0.79, 0.60 for 5th batch and 10th batch respectively. This finding is supported by Mala and Kurian [50] where the total flavonoid content of the oil samples used for general frying was significantly reduced, while the concentration of flavonoid increased in oil samples used for frying meat and fish. This could be due to flavonoids leaching into the frying oils from other food materials, particularly onions. Onion is a good source of quercetin, and flavonoids like quercetin are known for their in vitro stability even at high temperatures.

4 Conclusion

The result of this investigation demonstrated that repeated frying of vegetable cooking oil degraded its physicochemical and antioxidant qualities. All the oil samples that were taken when they first gathered from various streets were below acceptable limits, but as the heating time went on, they were typically beyond the allowable limit, specifically after 10th batch of frying. Additionally, in comparison to fresh oil samples, the oils' total phenolic and flavonoid contents were considerably low after repeated frying. This study thus demonstrates the degree to which the quality of the oils used by street vendors to fry their food has degraded. Therefore, safety precautions should be put in place because these declines in oil quality could have a negative impact on health due to the harmful compounds they create. Although degraded oils should not be consumed by humans, they can be used in alternative processes to make biodiesel and soap. We suggest that restaurants and food processing facilities perform oil quality checks and create regulatory requirements for frying oil quality.