Toward Rapid Analysis of Unsaturated Fatty Acid Oxidation in Edible Oils via Square-Wave Voltammetry

Toward Rapid Analysis of Unsaturated Fatty Acid Oxidation in Edible Oils via Square-Wave Voltammetry | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL Journal of the American Oil Chemists' Society This is a preprint and has not been peer reviewed. Data may be preliminary. 9 April 2025 V1 Latest version Share on Toward Rapid Analysis of Unsaturated Fatty Acid Oxidation in Edible Oils via Square-Wave Voltammetry Authors : Matthew M. Thelen 0009-0005-4736-3233 , Sebastian A. Flores , Jack A. Kelley , Brianna R. Swank , Jill Winkler-Moser 0000-0002-7210-1092 , and Matthew J. Fhaner 0000-0003-3008-3639 [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.174421805.53368145/v1 Published Journal of the American Oil Chemists' Society Version of record Peer review timeline 360 views 245 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Square-wave voltammetry (SWV) was investigated as a potential analytical tool for quantitation of unsaturated fatty acid oxidation in edible oils exposed to high heat. A traditional linear regression calibration curve and a novel single-point calibration method were used to quantitate linoleic acid in oxidized oil samples. These methods were compared to a standard method for fatty acid analysis via gas chromatography flame ionization detection (GC-FID). Both GC-FID and SWV methods were able to detect a decrease in linoleic acid concentration for heated oil samples. SWV methods routinely calculated higher linoleic acid concentrations compared to GC-FID analysis. A novel finding is that SWV oxidation peak intensities, which are associated with electroactive compounds, decrease in intensity over time for peaks below 1.0 V and increase for peaks above this voltage. These findings represent a step forward in electrochemical characterization of heated oils in addition to further advancing opportunities for electrochemical oil quality screening. Title: Toward Rapid Analysis of Unsaturated Fatty Acid Oxidation in Edible Oils via Square-Wave Voltammetry Matthew M. Thelen 1 , Sebastian A. Flores 1 , Jack A. Kelley 1 , Brianna R. Swank 1 , Jill K. Winkler-Moser 2 , Matthew J. Fhaner 3 1 University of Michigan-Flint undergraduate student 2 United States Department of Agriculture, Agricultural Research Service, 1815 N University Street, Peoria, IL 61604 3 University of Michigan-Flint Department of Chemistry and Biochemistry and Corresponding Author Running Title: Toward Rapid Voltammetric Analysis of Edible Oils ii. Abstract and keywords : Please provide an abstract of 150 to 250 words. The abstract should not contain any undefined abbreviations or unspecified references. Please provide 4 to 6 keywords which can be used for indexing purposes. Abstract and Keywords: Square-wave voltammetry (SWV) was investigated as a potential analytical tool for quantitation of unsaturated fatty acid oxidation in edible oils exposed to high heat. A traditional linear regression calibration curve and a novel single-point calibration method were used to quantitate linoleic acid in oxidized oil samples. These methods were compared to a standard method for fatty acid analysis via gas chromatography flame ionization detection (GC-FID). Both GC-FID and SWV methods were able to detect a decrease in linoleic acid concentration for heated oil samples. SWV methods routinely calculated higher linoleic acid concentrations compared to GC-FID analysis. A novel finding is that SWV oxidation peak intensities, which are associated with electroactive compounds, decrease in intensity over time for peaks below 1.0 V and increase for peaks above this voltage. These findings represent a step forward in electrochemical characterization of heated oils in addition to further advancing opportunities for electrochemical oil quality screening. Electrochemistry, PUFA, Linoleic Acid, Edible Oils, Lipid Oxidation Introduction: Rapid analysis of edible oils continues to be an area of research interest as highlighted by a recent review article comparing various methodologies for analyzing oil blends by Bian et al (Bian et al., 2022) . In that review, the term “rapid” or “fast” was included nearly forty times. The interest in advancing analytical strategies for edible oil analysis has led to a variety of research and practical applications including the authentication of edible oils via low-field NMR (Zhu et al., 2017), monitoring quality indices during thermal oxidation using infrared spectroscopy (Liu et al., 2020), and assessing fatty acid degradation using voltammetry (Ramirez-Montes et al., 2021). Still, there remains an opportunity for improving rapid analysis methodologies to assess fatty acid oxidation in frying oils. In part, continued improvements in assessing lipid quality is a response to the potential link between fatty acid oxidation and negative health outcomes. In a 2021 commentary by Martin Grootveld et al. , the authors present a compelling case that lipid oxidation products from polyunsaturated fatty acids (PUFAs) produced during thermal oxidation pose a public health concern deserving of focused attention (Grootveld et al., 2021). Additionally, recent studies observed analytical markers of oil quality (Peroxide Value, p-Anisidine Value, and Total Polar Compounds) to be significantly higher than accepted values in both laboratory and restaurant settings (Ghobadi et al., 2018; Multari et al., 2019). In fact, after just 20 minutes of deep-frying, the relative content of aldehydes such as acrolein, (E)-2-heptenal, and 2,4-heptadienal can significantly increase. This is alarming because acrolein is identified as being connected to numerous thermal oxidation reaction mechanisms that result in the production of Group 2A (probably carcinogenic) and Group 2B (possibly carcinogenic) species (Jiang et al., 2022). A variety of analytical approaches have been explored to assess edible oil quality under frying conditions. Most rapid analysis methods rely on colorimetric assays, changes in substrate electrical properties, or viscosity to assess various molecule classes including polymers, total polar compounds, and free fatty acids (Bansal et al., 2010). Commercially, there are currently a handful of pre-made testing kits available that rely on secondary assays such as total polar compounds, acid value, peroxide value, and p-anisidine value. However, these test methods rely on correlative analysis via spectroscopy instead of direct quantitative measurement of fatty acids. Research in our group has focused on using green electroanalytical methods to directly monitor oxidation of omega-3/6/9 fatty acids and antioxidants under thermal stress (Keene et al., 2019; Montney et al., 2023). In the work presented in this manuscript, a green electroanalytical method utilizing square-wave voltammetry (SWV) was used to determine the assess PUFA oxidation in edible oils during simulated frying temperatures. SWV methods were able to quantitate linoleic acid concentrations within an order of magnitude of GC-FID values. Additionally, both SWV and GC-FID calibration methods identified a decrease in linoleic acid concentration the longer an oil was exposed to elevated temperatures. However, SWV methods were not as precise or sensitive as GC-FID analysis and further optimization of electrode choice and experimental parameters would improve upon the results obtained in this study. Experimental Procedures: Materials: Commercially available canola oil, peanut oil, safflower oil, and vegetable oil were purchased from a local retailer. Chemical reagents included methanol (HPLC Grade, Sigma-Aldrich, Burlington, MA, USA), hexane (HPLC grade, Fisher Scientific, Waltham, MA, USA), ethanol (200 proof; Fisher Scientific: 04-355-451), ethyl acetate (HPLC Grade; Fisher Scientific: E195-4), sulfuric acid (96.5% w/w; Fischer Scientific: LC255501), potassium hydroxide (ACS grade, Fisher Scientific: ), linoleic acid standard (>99%; Sigma-Aldrich: L1376), linolenic acid standard (>98.5%; Sigma-Aldrich: 62160), oleic acid standard (>99%; Sigma-Aldrich: W281506), Supelco 37-component fatty acid methyl ester (FAME) standard (certified reference material; Sigma-Aldrich: CRM47885), methyl heneicosanoate (>99%, Nu-Check Prep, Elysian, MN, USA: N-21-M). A potentiostat and software for recording square wave voltammograms (CHI660E) were purchased from CH Instruments (Bee Cave, TX, USA). Additionally, a glassy-carbon working electrode (CHI104), silver/silver-chloride reference electrode (CHI111), platinum wire counter electrode (CHI115) electrode polishing kit (CHI120), and Picoamp Booster faraday cage (CHI200) were also purchased from CH Instruments. Electrode Pretreatment and Electrochemical Parameters: Square wave voltammograms were collected using a CH Instruments potentiostat with three-electrode system consisting of a glassy carbon working electrode, silver/silver-chloride reference electrode, and a platinum wire counter electrode. Prior to analysis, the glassy carbon working electrode was polished for 30 seconds using a 1.0-micron MicroPolish Powder (CH Instruments) slurry prepared with deionized ultrapure water on a polishing pad. The working electrode was polished for 60-90 seconds using a circular motion in a clockwise and counterclockwise pattern, changing direction every 15 second. The glassy carbon working electrode was rinsed with deionized ultrapure water and all electrodes were soaked in the solvent system (1:1 v/v ethanol-ethyl acetate with 0.24 M sulfuric acid) for 5 minutes. Background voltammograms were collected from 0.0 V to 1.5 V using a 0.004 V step increment, 0.005 V step amplitude, 2 Hz frequency, and a quiet time of 2 s. A crucial step in generating dependable SWV traces is establishing a stable baseline. Sample analysis proceeded only after a stable background was observed. Characterization of Edible Oils: Commercially available canola, peanut, safflower, and vegetable oils were purchased within 14 days of analysis. For SWV characterization, 750 µL of oil was mixed with 7.5 mL of the electrochemical solvent system and vortexed for 15 seconds. Following electrode pretreatment and obtaining stable background scans, the sample was placed into a fresh 10 mL glass electrochemical cell and voltammograms were collected from 0.0 V to 1.5 V using the same experimental parameters utilized for background collection. Determination of Fatty Acid Standards Figures of Merit and Correlation of Standard Peaks to Edible Oil Voltammograms: The fatty acids linoleic, linolenic, and oleic acid were selected for quantitative studies due to their significant contribution to the total fatty acid profile for the selected oils (Kostik et al., 2013). Standards of oleic and linolenic acids were prepared to 10,000 mg/kg, 20,000 mg/kg, 30,000 mg/kg, and 40,000 mg/kg for oleic and linolenic acid. Due to limited amounts of standard, the concentrations 8,089 mg/kg, 15,328 mg/kg, 22,992 mg/kg, 30,656 mg/kg, and 38,320 mg/kg were used for linoleic acid. Each standard was prepared in the ethanol-ethyl acetate solvent system. Following electrode pretreatment and obtaining stable background scans, voltammograms for each standard solution were collected in quadruplicate with a fresh aliquot of standard solution representing a single replicate. Oxidation peaks observed in each fatty acid standard were qualitatively correlated to oxidation peaks observed in the edible oils. Additionally, linear regression calibration curves were used to assess the analytical figures of merit for each fatty acid standard. To this end, background subtracted SWV traces were manually analyzed to extract oxidation peak potentials, peak areas, and peak currents for the quadruplicate data set. The analytical figures of merit for each calibration plot were assessed similar to previous studies via linear regression(Keene et al., 2019; Montney et al., 2023). Limits of detection (LOD) and lower limits of quantitation (LLOQ) were calculated using the standard deviation (“s”) for the lowest concentration standard solution and the slope of the linear regression analysis (“m”) via the following equations: Eq. 1 \(Limit\ of\ Detection=\frac{3s}{m}\) Eq. 2 \(Lower\ Limit\ of\ Quantitation=\ \frac{10s}{m}\) Single-Calibration Validation of Fatty Acid Standards: Individual fatty acid standards of oleic, linoleic, and linolenic acids were prepared to 10,000 mg/kg in the ethanol-ethyl acetate solvent system at a volume of 10 mL. Following electrode pretreatment and obtaining stable background scans, the full volume of the 10,000 mg/kg standard was placed into a clean 10 mL glass electrochemical cell and ten voltammograms were collected in five minutes intervals using the same experimental parameters as the background scans. Each fatty acid standard was prepared and evaluated as described at least twice. Comparison of Electrochemical and Gas Chromatography Flame Ionization Detector Analysis of Polyunsaturated Fatty Acid Oxidation via Simulated Frying Temperatures: Edible oil samples were prepared by adding 10 grams of each oil to a 20 mL scintillation vial. Fifteen individual samples were prepared for each oil. All samples were placed uncapped into a Fisher Scientific IsoTemp laboratory oven held at a constant temperature of 180° C. At 0-, 2-, 4-, 6-, and 8-hrs, three vials of each oil were removed from the oven. Each sample was split into two aliquots and stored in a -80° C deep freezer until analysis. SWV analysis was completed at UM-Flint and GC-FID analysis was completed at the USDA-ARS laboratory in Peoria, IL. For electrochemical analysis, all samples of a single oil type were removed from the -80° C deep freezer and allowed to come to room temperature. Following electrode pretreatment and obtaining stable background scans, 750 µL of a 0-hr sample was mixed with 7.5 mL of the electrochemical solvent system and vortexed for 15 seconds. The sample was placed into a clean electrochemical glass cell and a voltammogram was collected from -0.5 V to 1.5 V. All other experimental parameters were the same as previously described. This process was repeated for a second trial such that two SWV traces were collected for every individual oil sample. Before analyzing the next 0-hr sample, all electrodes were placed back into the solvent and background voltammograms were collected to ensure no adsorption had occurred at the electrode surface. Collecting these backgrounds between samples was also critically important for ensuring reproducible results. This process was repeated until all three 0-hr samples were analyzed in duplicate. The same procedure was followed for the remining time points. After completing sample analysis, the electrodes were once again placed in the solvent system to ensure no adsorption had occurred and then a voltammogram of a 10,000 mg/kg linoleic acid standard was collected for quantitative analysis. This procedure was completed for all four oil samples assessed. Fatty acid methyl ester (FAME) preparation and gas chromatography analysis Oil samples (20 mg) were weighed into tared 13 x 100 mm glass test tubes. Internal standard (2.0 mg) of methyl heneicosanoate (5.0 mg/ml in hexane) was added to each tube, followed by 1.6 ml hexane and 0.2 ml 2 N KOH in methanol (Ichihara et al., 1996). Tubes were mixed by vortex for 2 min. After allowing the layers to separate, 1.0 ml of the top (hexane) layer containing FAME was transferred to an autosampler vial. FAMEs (1.0 µl) were analyzed on a Shimadzu 2010 GC (Columbia, MD, USA) with FID detector. The column was a ZB-FAME (30 m x 25 mm i.d. x 20 µm film thickness, Phenomenex, Torrance, CA) with hydrogen as carrier gas (1.2 ml/min). The injector temperature was 275 °C, the detector temperature was 285 °C, oven program was 100 °C held for 2 min, followed by 4 C/min ramp to 240 C where it was held for 2 min. Each sample was injected in duplicate. Peaks were identified by retention times of the Supelco 37-component FAME mix (SigmaAldrich). Peak areas were calculated, and quantitation was based on the internal standard method. Data were collected and analyzed by LabSolutions software (Shimadzu). Statistical Analysis: Where applicable, data are presented as a mean and associated error is presented as standard deviation. For graphical presentation, error bars are presented as standard deviations. Statistical significance between data points within a treatment were calculated using the Student’s t-test with p < 0.05 used to determine statistical significance. Where presented, the units of mg/kg represent mg solute per kg solvent. Results & Discussion: Characterization of Edible Oils: Square wave voltammograms (SWV) of 30,000 mg/kg linoleic, linolenic, and oleic acid standards are presented in Figure 1. Linoleic acid produces an oxidation peak near 0.67 V that is not observed in linolenic or oleic acid. Similarly, linolenic acid produces an oxidation peak near 1.0 V that is not observed in linoleic or oleic acid. Oleic acid produces oxidation peaks near 1.3 and 1.45 V, respectively. These peaks overlap with oxidation currents observed in linoleic and linolenic acid. For the purposes of this study, the 0.67 V oxidation peak was used to quantitate linoleic acid due to its lack of interfering peaks from linoleic and oleic acid. The 1.0 V peak was used for calibration analysis of linolenic acid and the 1.45 V peak was used for quantitation of oleic acid, respectively. Each fatty acid produces a unique SWV trace due to the differing molecular structures and oxidation chemistry of each fatty acid. For example, autoxidation of the methyl ester of oleic acid has been shown to produce four allylic hydroperoxides at the 8, 11 and 9, 10 carbons, respectively. The methyl ester of linoleic acid produces a mixture of cis-trans and trans-trans conjugated diene hydroperoxides at the 9 and 13 position carbons. Finally, the methyl ester of linolenic acid produces a wide variety of hydroperoxides including cis-trans, trans-trans-, and cis conjugated diene hydroperoxides at the 9, 12, 13, and 16, position carbons (Madhujith & Sivakanthan, 2019). These differences in the number of allylic oxidation points on each fatty acid and energy required to abstract hydrogens impactd the unique SWV profiles observed in Figure 1. Figure 2 presents representative SWV traces for canola, peanut, safflower, and vegetable oils. As would be expected, based on their differing fatty acid profiles, each oil produces a unique SWV trace that reflects the contents of the oleic, linoleic, and linolenic acids as well as other fatty acids and oxidizing and reducing species. Canola oil is high in oleic acid (61%) and lower in linoleic acid (21%) and linolenic acid (11%), which correlates to the prominent peaks at 1.5 V for oleic acid, 0.67 and 1.3 V for linoleic and oleic acid, and 1.1 V for linoleic acid (Lewinska et al., 2015). Peanut oil is roughly 45-68% oleic acid and 15-35% linoleic acid with minimal linolenic acid present (Carrín & Carelli, 2010), which explains the lack of a prominent peak near 1.1 V due to the minimal linolenic acid content. Safflower oil is roughly 17% oleic acid and 76% linoleic acid, as where vegetable oil (soybean) is roughly 28% oleic acid, 50 % linoleic acid, and 8% linolenic acid (Kostik et al., 2013). Like peanut oil, both safflower and vegetable both lack a distinct peak near 1.1 V due to the low amounts of linolenic acid present in these oils. Additional oxidation peaks are observed in the edible oils presented in Figure 2 that were not present in the fatty acid standard as observed in Figure 1. Peaks seen near 0.48 V in canola oil, 0.78 V in peanut oil, safflower oil and vegetable oil, and peaks near 0.85 in vegetable oil do not correlate to the SWV traces obtained for the fatty acid standards in Figure 1. These peaks could be related to natural antioxidants such as tocopherols which are present in the oils evaluated (Matthaus et al., 2015; Sikorska et al., 2005) and have been shown to have oxidation potentials near the unidentified peaks under similar experimental conditions (Lubeckyj et al., 2017). Another explanation for these unidentified oxidation peaks could be due to other fatty acids within the oil and/or blending of the commercial oils evaluated; however, these hypotheses are outside the scope of this particular study and would need to be pursued with more rigor if their presence is deemed to be relevant. Determination of Fatty Acid Standards Figures of Merit and Correlation of Standard Peaks to Edible Oil Voltammograms: A primary goal of this study was to assess the feasibility of using SWV as a quick and robust analysis method for quantifying polyunsaturated fatty acids in common oils subjected to elevated temperatures. To that end, calibration curves were developed for linoleic, linolenic, and oleic acid standards. Figure 3 details linear regression plots for each of the fatty acid standards. Figures of merit, including limit of detection (LOD), lower limit of quantitation (LLOQ), and correlation coefficient for each standard are outline in Table 1. Linoleic acid was analyzed using the oxidation peak near 0.67 V and produced a LOD of 1152 mg/kg, LLOQ of 3840 mg/kg, and a correlation coefficient of 0.9766. Linolenic acid was analyzed using the oxidation peak near 1.00 V and produced a LOD of 1646 mg/kg, a LLOQ of 5485 mg/kg, and a correlation coefficient of 0.9957. Oleic acid was analyzed using the oxidation peak near 1.45 V and produced a LOD of 3972 mg/kg, a LLOQ of 13240 mg/kg, and a correlation coefficient of 0.9828. On their own, the figures of merit do not consider experimental conditions, which rely on oil samples as the source of fatty acids. These values can be further extrapolated to experimental conditions using a solvent density of 0.8455 g/mL and an oil density of 0.95. Thus, the limits of detection, when expressed as mg fatty acid per gram of oil are 11.3 mg/g for linoleic acid, 16.1 mg/g for linolenic acid, and 38.9 mg/g for oleic acid. Lower limits of detection are 37.6 mg/g for linoleic acid, 53.7 for linolenic acid, and 129.6 for oleic acid. The LOD and LLOQ values are high when compared to other quantitative analytical methods such as GC-FID, LC-MS methods (Kalogiouri et al., 2022; Slobodianiuk et al., 2022), however the values obtained in this study do compare with results obtained for IR methods which are more commonly used for seed oil adulteration assessment than quantitative analysis (Laouni et al., 2023; Rodrigues de Aguiar et al., 2024). Still, given the concentration of linoleic, linolenic, and oleic acid range from 8-76 wt%, or 80,000-760,000 mg/kg, the figures of merit outlined in Table 1 provide an acceptable starting point for the objectives of this study. Single-Calibration Validation of Fatty Acid Standards: The primary disadvantage of SWV linear regression calibration analysis is that collecting the data points for a calibration curve can take upwards of 30 minutes. In an effort to decrease the time spent on calibration data collection, it was hypothesized that a single-point calibration method may provide a quick and reliable option for quantifying polyunsaturated fatty acid oxidation in the presence of high temperatures. As detailed in Figure 4, 10,000 mg/kg standard solutions of linoleic, linolenic, and oleic acid were subjected to repetitive SWV every five minutes for a total of ten trials. This process was completed at least twice for each fatty acid standard and oxidation currents normalized to the first scan of a series were plotted to determine if analytical signal deviated over multiple SWV scans. Oxidation currents for the 0.67 V peak in linoleic acid deviated between an average analytical signal of 96-101% over the ten trials. The 1.00 V oxidation peak average signal for linolenic acid deviated between 89% and 111% with scans 2 and 4-7 yielding statistically different values from the initial scan. The 1.45 V oxidation peak average signal for oleic acid deviated between 100-114% with scan 4 and 6-7 yielding statistically different values from the initial scan. From this data, it was decided to focus on linoleic acid as the primary analyte of interest for comparing single-point and linear regression quantitation. Comparison of Electrochemical and Gas Chromatography Flame Ionization Detector Analysis of Polyunsaturated Fatty Acid Oxidation via Simulated Frying Temperatures: Canola, peanut, safflower, and vegetable oils were left open to 180 °C and SWV traces were collected at 0-, 2-, 4-, 6-, and 8-hr. Overlays of the averaged SWV traces for each time point are provided in Figure 5. Peanut, safflower, and vegetable oils yielded reproducible SWV traces with peaks displaying minimal shifts in oxidation potential from 0-hr to 8-hr time points. Minor shifts in oxidation peak potentials are to be expected as the high heat treatment can cause changes in sample resistance requiring an increase in overpotential at the electrode surface to induce oxidation (Amatore et al., 2009). Canola oil produced a high variability in SWV traces for different time points compared to other oil samples. This could be due to local variations in oven temperature where the samples were heated or possibly an experimental error, however the 6- and 8-hr time points yield the largest deviations in oxidation peak potentials. All oils exhibited a decrease in oxidation current for peaks at or below 1.0V versus an Ag/AgCl refence electrode. Peanut and safflower oils accentuate the decreases, with peanut oil oxidation peaks between 0.50 and 1.00 V nearing the baseline by 4-hr and the 0.67 V safflower oil oxidation peak converging with the baseline between the 6- and 8-hr marks. Conversely, a general trend was observed in which oxidation currents for peaks above 1.0 V increased over time. An interesting finding from this study is that as a peak’s oxidation potential moves more negative from 1.0 V, the analytical signal decreases at a faster rate over time. In contrast, as a peak’s oxidation potential moves more positive from 1.0 V, the analytical signal increases at a faster rate over time. Figure 6 provides an overview of the change in analytical signal from 0-Hr for each oxidation peak observed for all oils assessed between 0- and 8-Hr. Linear regression was used to generate best-fit-lines for all data presented in Figure 6. The slope of the best-fit-line along with the correlation coefficient are outlined in Table 2 along with the raw data from the Figure 6 plots. Using Canola Oil as an example, Figure 6 shows that the two peaks below 1.0 V (0.48 V and 0.60 V) both yielded a decrease in analytical signal from 0-hr to 8-hr. Using the slopes of the best-fit-line for these two peaks, 0.48 V analytical signal decreases at a rate of -9.85%/hr and 0.60 V analytical signal decreases at a rate of -9.05%/hr with correlation coefficients of 0.9359 and 0.9771, respectively. 1.050 V, which is slightly higher than 1.0 V, produces minor change in signal with a -0.05%/hr change in analytical signal and a correlation coefficient of 0.0048. The 1.350 V peak in canola oil produced an increase in analytical signal over time, with a 4.95%/hr change in analytical signal and a correlation coefficient of 0.6969. A similar trend is observed for the other oils assessed. It has been noted that as fatty acid double-bond saturation increases, so too does the rate of oxidation. Monosaturated oleic acid oxidized roughly 100 times faster than the unsaturated stearic acid. Linoleic acid with two double bonds and linoleic acid with three double bonds oxidize roughly 1,200 and 2,500 times faster than that stearic acid (Bockisch, 1998). It is possible the varying rates of signal change could relate to changes in the types of electrons available for oxidation during SWV analysis. For example, it has been established that autoxidation occurs through allylic carbons near the double bonds of fatty acids leading to hydroperoxide formation (Frankel, 1995). Hydroperoxides and hydroperoxide radicals have been estimated to have oxidation potentials ranging between 0.77-1.69 V versus a normal hydrogen electrode (Merenyi et al., 1994). It would be expected that the oxidation potentials for peaks reported in this study would differ from the values predicted by Merenyi et al. given the specific experimental design differences in this work. Therefore, an explanation for the increasing oxidation currents above 1.0 V observed in this study could be related to increases in hydroperoxide and hydroperoxide radical formation as oils are exposed to the high heat over time and could warrant further study. While this observation is not related to quantitation of fatty acids, it does yield insights into future modeling or machine learning opportunities toward predictive screening of polyunsaturated fatty acid oxidation using SWV in a similar vein to those currently being pursued by other researchers (Ramiro et al., 2024; Strati et al., 2024). The objective of this study was to quantitate linoleic, linolenic, and oleic acid via SWV linear regression and single-point calibration and compare the results to an established GC-FID method. Given the increase in analytical signals observed above 1.0 V, only linoleic acid was the only fatty acid able to be quantified with any confidence using SWV techniques. Linoleic acid values were normalized to mg linoleic acid per gram of oil and are presented in Table 3. Using the GC-FID values as a comparison point, the linoleic acid values quantitated using the linear regression model produced values closer to those obtained via GC-FID than the single-point calibration method. However, when compared to GC-FID, both SWV quantitation models overestimated the amount of linoleic acid in samples except for in vegetable oil where the linear regression model underestimated the linoleic acid concentration. Additionally, SWV data resulted in larger variability in calculated means compared to GC-FID data suggesting optimization in the SWV equipment and/or experimental parameters remains a priority. Although the SWV predicted values for linoleic acid concentration using either the linear regression or single point prediction models lacked accuracy, within each oil there were strong correlations between linoleic acid concentration over oxidation time calculated by SWV and FID time for both SWV quantitation models (Table 3). Conclusion: The results of this study represent an advancement in the development of an electrochemical method for assessing polyunsaturated fatty acid oxidation in edible oils. Major findings from this study include the observation of a voltage-dependent relationship between oxidation currents and sample exposure time to heat and could be used for predictive modeling of oil quality. Additionally, SWV quantitation methods for linoleic acid found a decrease in linoleic acid concentration over time similar to GC-FID analysis. Both linear regression and single-point calibration methods via SWV resulted in a higher calculated linoleic acid concentration than GC-FID quantitation for all oils except for vegetable oil where the linear regression model returned lower than expected values. Electrochemical methods for lipid oxidation in edible oils continue to represent a path towards providing low-cost, on-site possibilities for screening oil quality. Acknowledgements: The authors would like to thank Dr. Cassie Fhaner for providing valuable feedback during the revision process of this manuscript. The authors would also like to thank the University of Michigan-Flint Office of Research and Economic Development for providing stipends to support student research through the Undergraduate Research Opportunity Program and the Summer Undergraduate Research Experience program. The authors would also like to thank Julie Anderson (USDA) for her technical assistance with GC analysis. The research was funded in part by the U.S. Department of Agriculture, Agricultural Research Service. Mention of trade names or commercial products in this article is solely for the purpose of providing scientific information and does not imply recommendation or endorsement by the USDA. USDA is an equal opportunity provider and employer. Authorship: M.F. conceived and designed the study, provided training to undergraduate researchers, provided oversight to data collection and analysis, and wrote the manuscript. M.T., S.F., and J.K. conducted the heating and electrochemical experimental work. M.T. and B.S. provided electrochemical data analysis support. J. W.-M. provided chromatography experimental and data analysis support along with written contributions to the manuscript. All authors contributed to and approved the final draft of the manuscript. References: Amatore, C., Oleinick, A., Klymenko, O. V., & Svir, I. (2009). Capacitive and solution resistance effects on voltammetric responses at a disk microelectrode covered with a self-assembled monolayer in the presence of electron hopping. Analytical Chemistry , 81 (20), 8545–8556. https://doi.org/10.1021/ac901513x Bansal, G., Zhou, W., Barlow, P. J., Joshi, P. S., Lo, H. L., & Chung, Y. K. (2010). Review of rapid tests available for measuring the quality changes in frying oils and comparison with standard methods. Critical Reviews in Food Science and Nutrition , 50 (6), 503–514. https://doi.org/10.1080/10408390802544611 Bian, X., Wang, Y., Wang, S., Johnson, J. B., Sun, H., Guo, Y., & Tan, X. (2022). A Review of Advanced Methods for the Quantitative Analysis of Single Component Oil in Edible Oil Blends. In Foods (Vol. 11, Issue 16). MDPI. https://doi.org/10.3390/foods11162436 Bockisch, M. (1998). Fats and Oils Handbook . AOCS Press. https://doi.org/https://doi.org/10.1016/C2015-0-02417-0 Carrín, M. E., & Carelli, A. A. (2010). Peanut oil: Compositional data. In European Journal of Lipid Science and Technology (Vol. 112, Issue 7, pp. 697–707). https://doi.org/10.1002/ejlt.200900176 Frankel, E. N. (1995). Chemistry of autoxidation: mechanism, products and flavor significance. In D. B. Min & T. H. Smouse (Eds.), Flavor Chemistry of Fats and Oils (p. 138). AOCS. Ghobadi, S., Akhlaghi, M., Shams, S., & Mazloomi, S. M. (2018). Frying oil of fast food restaurants Acid and Peroxide Values and Total Polar Compounds of Frying Oils in Fast Food Restaurants of Shiraz, Southern Iran Restaurant Frying oil Acid value Peroxide value Total polar compounds. In Int J Nutr Sci March (Vol. 3, Issue 1). Grootveld, M., Percival, B. C., Moumtaz, S., Gibson, M., Woodason, K., Akhtar, A., Wawire, M., Edgar, M., & Grootveld, K. L. (2021). Commentary: Iconoclastic reflections on the ‘safety’ of polyunsaturated fatty acid-rich culinary frying oils: Some cautions regarding the laboratory analysis and dietary ingestion of lipid oxidation product toxins. In Applied Sciences (Switzerland) (Vol. 11, Issue 5, pp. 1–34). MDPI AG. https://doi.org/10.3390/app11052351 Ichihara, K. N. ichi, Shibahara, A., Yamamoto, K., & Nakayama, T. (1996). An improved method for rapid analysis of the fatty acids of glycerolipids. Lipids , 31 (5), 535–539. https://doi.org/10.1007/BF02522648 Jiang, K., Huang, C., Liu, F., Zheng, J., Ou, J., Zhao, D., & Ou, S. (2022). Origin and Fate of Acrolein in Foods. In Foods (Vol. 11, Issue 13). MDPI. https://doi.org/10.3390/foods11131976 Kalogiouri, N. P., Manousi, N., Mourtzinos, I., Rosenberg, E., & Zachariadis, G. A. (2022). A Rapid GC-FID Method for the Determination of Fatty Acids in Walnut Oils and Their Use as Markers in Authenticity Studies. Food Analytical Methods , 15 (3), 761–771. https://doi.org/10.1007/s12161-021-02157-3 Keene, K. A., Ruddy, R. M., & Fhaner, M. J. (2019). Investigating the Relationship between Antioxidants and Fatty Acid Degradation Using a Combination Approach of GC-FID and Square-Wave Voltammetry. ACS Omega , 4 (1), 983–991. https://doi.org/10.1021/acsomega.8b02275 Kostik, V., Memeti, S., & Bauer, B. (2013). FATTY ACID COMPOSITION OF EDIBLE OILS AND FATS . Laouni, A., El Orche, A., Elhamdaoui, O., Karrouchi, K., Karbane, M. El, & Bouatia, M. (2023). A Preliminary Study on the Potential of FT-IR Spectroscopy and Chemometrics for Tracing the Geographical Origin of Moroccan Virgin Olive Oils. Journal of AOAC International , 106 (3), 804–812. https://doi.org/10.1093/jaoacint/qsac136 Lewinska, A., Zebrowski, J., Duda, M., Gorka, A., & Wnuk, M. (2015). Fatty acid profile and biological activities of linseed and rapeseed oils. Molecules , 20 (12), 22872–22880. https://doi.org/10.3390/molecules201219887 Liu, H., Chen, Y., Shi, C., Yang, X., & Han, D. (2020). FT-IR and Raman spectroscopy data fusion with chemometrics for simultaneous determination of chemical quality indices of edible oils during thermal oxidation. LWT , 119 . https://doi.org/10.1016/j.lwt.2019.108906 Lubeckyj, R. A., Winkler-Moser, J. K., & Fhaner, M. J. (2017). Application of Differential Pulse Voltammetry to Determine the Efficiency of Stripping Tocopherols from Commercial Fish Oil. JAOCS, Journal of the American Oil Chemists’ Society , 94 (4), 527–536. https://doi.org/10.1007/s11746-017-2968-0 Madhujith, T., & Sivakanthan, S. (2019). Oxidative Stability of Edible Plant Oils. In J.-M. Merillon & K. Ramaway (Eds.), Bioactive Molecules in Food (pp. 529–551). Springer Nature. Matthaus, B., Özcan, M. M., & Al Juhaimi, F. Y. (2015). Fatty acid composition and tocopherol profiles of safflower (Carthamus tinctorius L.) seed oils. Natural Product Research , 29 (2), 193–196. https://doi.org/10.1080/14786419.2014.971316 Merenyi, G., Linda, J., & Engmanb, L. (1994). One-and Two-electron Reduction Potentials of Peroxyl Radicals and Related Species . https://doi.org/https://doi.org/10.1039/p29940002551 Montney, J. M., Siddiqui, A.-R., Flores, S. A., & Fhaner, M. J. (2023). Identifying greener solvents for square wave voltammetry applications in functional food analysis. Frontiers in Environmental Chemistry , 4 . https://doi.org/10.3389/fenvc.2023.1226960 Multari, S., Marsol-Vall, A., Heponiemi, P., Suomela, J. P., & Yang, B. (2019). Changes in the volatile profile, fatty acid composition and other markers of lipid oxidation of six different vegetable oils during short-term deep-frying. Food Research International , 122 , 318–329. https://doi.org/10.1016/j.foodres.2019.04.026 Ramirez-Montes, S., Santos, E. M., Galan-Vidal, C. A., Tavizon-Pozos, J. A., & Rodriguez, J. A. (2021). Classification of Edible Vegetable Oil Degradation Using Multivariate Data Analysis From Electrochemical Techniques. Food Analytical Methods , 14 (12), 2597–2606. https://doi.org/10.1007/s12161-021-02083-4 Ramiro, J. L., Neo, A. G., Pérez-Palacios, T., Antequera, T., & Marcos, C. F. (2024). Machine learning-enabled fatty acid quantification and classification of pork from autochthonous breeds using low-field 1H NMR spectroscopic data. Food Control , 166 . https://doi.org/10.1016/j.foodcont.2024.110753 Rodrigues de Aguiar, T., Dorocz, E. L., Dalagrana do Santos, L., Coelho Tanamati, A. A., Gozzo, A. M., & Bona, E. (2024). Mid-infrared spectroscopy and chemometrics in the detection of adulteration in chia oil (Salvia hispanica L) and α-linolenic acid content prediction. Food Control , 165 . https://doi.org/10.1016/j.foodcont.2024.110687 Sikorska, E., Gliszczyńska-Świgło, A., Khmelinskii, I., & Sikorski, M. (2005). Synchronous fluorescence spectroscopy of edible vegetable oils. Quantification of tocopherols. Journal of Agricultural and Food Chemistry , 53 (18), 6988–6994. https://doi.org/10.1021/jf0507285 Slobodianiuk, L., Budniak, L., Feshchenko, H., Sverstiuk, A., & Palaniza, Y. (2022). Quantitative analysis of fatty acids and monosaccharides composition in Chamerion angustifolium L. by GC/MS method. Pharmacia . https://doi.org/10.3897/phar Strati, I. F., Tsiantas, K., Psouni, A., Ladika, G., Cavouras, D., & Sinanoglou, V. J. (2024). Quality Assessment of Edible Plant-Based Fixed Oils Using Different Analytical Techniques and Machine Learning Approaches. Applied Sciences (Switzerland) , 14 (22). https://doi.org/10.3390/app142210305 Zhu, W., Wang, X., & Chen, L. (2017). Rapid detection of peanut oil adulteration using low-field nuclear magnetic resonance and chemometrics. Food Chemistry , 216 , 268–274. https://doi.org/10.1016/j.foodchem.2016.08.051 Figure Legends: Figure 1 Square wave voltammograms of linoleic, linolenic, and oleic acid standards at 30,000 mg/kg in 1:1 v/v ethanol-ethyl acetate with 0.24 M sulfuric acid. Primary oxidation peaks used for quantitative analysis are identified for each fatty acid by an asterisk. Figure 2 Square wave voltammograms of canola, peanut, safflower, and vegetable oils 1:1 v/v ethanol-ethyl acetate with 0.24 M sulfuric acid Figure 3 Linear regression calibration curves for linoleic (circle), linolenic (oblique square), and oleic acids (triangle) Figure 4 Repetitive square wave voltammetry analysis of linoleic (circle; 0.66 V oxidation peak), linolenic (triangle; 1.0 V oxidation peak), and oleic (square; 1.45 oxidation peak) acids at 10,000 mg/kg. Oxidation currents are presented as a percentage of the analytical signal obtained in the initial scan. Each data point is the average of at least two trials, with a single trial representing ten consecutive square wave voltammograms collected in five-minute intervals. Error bars are presented as standard deviations of total trials for a given fatty acid. Statistically significant differences for linolenic acid (*) and oleic acid (#) values are marked as calculated using a Student’s t-test with a p<0.05 Figure 5 Square wave voltammogram overlays for canola, peanut, safflower, and vegetable oils at 0- (solid), 2- (dashed), 4- (dot), 6- (dash-dot), and 8-hr (dash-dot-dot) Figure 6 Scatter plots of percent change in analytical signal for each observed oxidation peak in canola, peanut, safflower, and vegetable oils. The slope of the linear regression analysis and correlation coefficient are detailed in Table 2 Tables: Table 1: Analytical figures of merit including limit of detection (LOD), lower limit of quantitation (LLOQ), and correlation coefficient obtained using data presented in Figure 3 Linoleic Acid (0.66 V) 1152 11.27808 3840 37.5936 0.9766 Linolenic Acid (1.0 V) 1646 16.11434 5485 53.69815 0.9957 Oleic Acid (1.45 V) 3972 38.88588 13240 129.6196 0.9828 Table 2: Change in analytical signal as measure via oxidation currents for each identified peak in canola, peanut, safflower, and vegetable oils. Oxidation currents are presented as a net change in percentage normalized to 0-hr values. The slope and correlation coefficient for linear regression analysis of data presented in Figure 6 is also included. Canola Oil Percent Change in Oxidation Current 0.480 V 0 -30 -60 -65 -81 -9.85 0.9359 0.600 V 0 -18 -46 -55 -72 -9.05 0.9771 1.050 V 0 2 -2 -3 2 -0.05 0.0048 1.350 V 0 13 18 12 50 4.95 0.6969 Peanut Oil Percent Change in Oxidation Current 0.490 V 0 -34 -37 -64 -68 -8.30 0.9176 0.590 V 0 -20 -18 -39 -49 -5.85 0.9307 0.670 V 0 -28 -30 -39 -53 -5.85 0.9042 0.780 V 0 -11 -10 -15 -38 -4.00 0.8052 0.850 V 0 -10 5 -8 -21 -2.00 0.4012 1.235 V 0 4 13 21 27 3.55 0.9884 1.495 V 0 23 37 61 89 10.80 0.9885 Safflower Oil Percent Change in Oxidation Current 0.525 V 0 -28 N.D. N.D. N.D. -25.00 0.9394 0.660 V 0 -52 -84 -96 N.D. -12.20 0.8590 0.750 V 0 -31 -60 -69 N.D. -11.90 0.9763 1.170 V 0 26 13 11 -7 -1.45 0.1303 1.455 V 0 31 46 47 86 9.40 0.9152 Vegetable Oil Percent Change in Oxidation Current 0.575 V 0 -6 -2 -25 -41 -5.05 0.8156 0.660 V 0 -28 -29 -36 -54 -5.80 0.8881 0.780 V 0 -10 -9 -14 -37 -3.90 0.7943 0.850 V 0 -9 -6 -7 -20 -1.90 0.6773 1.235 V 0 5 14 22 28 3.65 0.9927 1.495 V 0 20 34 57 57 7.55 0.9464 Table 3. Linoleic acid quantitation via gas chromatography flame ionization detection, square wave voltammetry using linear regression, and square wave voltammetry using single point calibration. Values are presented as averages and standard deviations respectively and were taken from the voltammogram overlays in figure 5. Safflower Oil Linoleic Acid Concentration (mg linoleic acid/g oil) Linoleic Acid: GC-FID 163 ± 0 154 ± 1 148 ± 5 139 ± 8 126 ± 4 Linoleic Acid (0.67 V): SWV Calibration Curve 740 ± 40 370 ± 40 130 ± 40 47 ± 6 50 ± 2 Linoleic Acid (0.67 V): SWV Single Point Calibration 1920 ± 40 920 ± 30 300 ± 100 70 ± 10 N. D. Canola Oil Linoleic Acid Concentration (mg linoleic acid/g oil) Linoleic Acid: GC-FID 171 ± 2 160 ± 1 159 ± 1 154 ± 2 146 ± 2 Linoleic Acid (0.67 V): SWV Calibration Curve 1000 ± 100 830 ± 60 550 ± 90 460 ± 90 300 ± 50 Linoleic Acid (0.67 V): SWV Single Point Calibration 2600 ± 200 2160 ± 70 1400 ± 100 1200 ± 200 730 ± 50 Peanut Oil Linoleic Acid Concentration (mg linoleic acid/g oil) Linoleic Acid: GC-FID 190 ± 5 177 ± 4 167 ± 7 155 ± 5 150 ± 7 Linoleic Acid (0.67 V): SWV Calibration Curve 370 ± 50 280 ± 10 270 ± 20 240 ± 30 190 ± 20 Linoleic Acid (0.67 V): SWV Single Point Calibration 1000 ± 100 690 ± 20 670 ± 30 590 ± 90 449 ± 8 Vegetable Oil Linoleic Acid Concentration (mg linoleic acid/g oil) Linoleic Acid: GC-FID 540 ± 10 521 ± 9 503 ± 4 497 ± 3 464 ± 2 Linoleic Acid (0.67 V): SWV Calibration Curve 360 ± 60 270 ± 20 260 ± 20 240 ± 30 180 ± 20 Linoleic Acid (0.67 V): SWV Single Point Calibration 700 ± 100 493 ± 9 490 ± 20 440 ± 70 320 ± 20 Information & Authors Information Version history V1 Version 1 09 April 2025 Peer review timeline Published Journal of the American Oil Chemists' Society Version of Record 24 Jun 2025 Published Copyright This work is licensed under a Non Exclusive No Reuse License. Collection Journal of the American Oil Chemists' Society Keywords food and feed science / nutrition and health lipid chemistry / lipid analysis lipids oxidative stability thermal analysis Authors Affiliations Matthew M. Thelen 0009-0005-4736-3233 University of Michigan-Flint View all articles by this author Sebastian A. Flores University of Michigan-Flint View all articles by this author Jack A. Kelley University of Michigan-Flint View all articles by this author Brianna R. Swank University of Michigan-Flint View all articles by this author Jill Winkler-Moser 0000-0002-7210-1092 USDA-ARS Midwest Area View all articles by this author Matthew J. Fhaner 0000-0003-3008-3639 [email protected] University of Michigan-Flint College of Innovation & Technology View all articles by this author Metrics & Citations Metrics Article Usage 360 views 245 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Matthew M. Thelen, Sebastian A. Flores, Jack A. Kelley, et al. Toward Rapid Analysis of Unsaturated Fatty Acid Oxidation in Edible Oils via Square-Wave Voltammetry. Authorea . 09 April 2025. DOI: https://doi.org/10.22541/au.174421805.53368145/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.174421805.53368145/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fe91abd3bb952ad',t:'MTc3OTI1NjY1Mg=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();