The Stability of Edible Oils

Edible oils stability preoccupies many product developers using these ingredients in their recipes and formulations. Oxidative stability of an oil is an important quality factor because it is a measure of the resistance to oxidation during processing and storage. It is used to judge shelf life of the oil, its overall quality and its susceptibility to change.

Cooking processes are largely responsible for major changes to oils, especially their composition and ultimately performance. Frying technology incidentally is a process that comes in many guises. We have vacuum-frying, spray-frying, pan-frying (PF), deep-frying, roast-frying (RF) and air-frying (AF). Comparisons between the methods usually focus on the quality of the food being fried. There are few studies that look at edible oil quality It is almost a given now that deep fat frying has a major damaging effect on edible oil stability. We look at some of the chemical reactions occurring which characterise the phenomenon.

In most cases free fatty acid (FFA) content and total polar materials (TPM) are the best guides to edible oil stability and decay but there are many other measures too.  Whilst analytical methods to monitor these two parameters is probably the most effective when it comes to checking how far an oil has broken down, there are other measures worth considering alongside these too.

Here we look at the factors and then a brief survey of methods to help put them into context.

Factors Affecting Edible Oils Stability

The factors affecting the oxidation of edible oils are the following:-

• Fatty acid composition
• Energy input in the form of heat and light
• Level of oxygenation and the presence of different types of oxygen like ozone.
• Presence of peroxides
• Presence of metals especially transition metals like iron and copper
• Free radicals especially oxygen and hydroxyl radicals.
• Free fatty acids and mono- and diacylglycerols
• Thermally oxidised compounds

Edible oils such as canola (maize), corn and sunflower oil are regularly used as frying oils. Their quality needs to be regularly checked as processors like to reuse oil before it breaks down and becomes unusable. The two key values are acid value and peroxide value.

Fast food outlets such as fish and chip shops, burger outlets, kebab shops all rely on high quality oil to sustain their business. 

Prooxidants Affecting Edible Oils Stability

Lipid oxidation is often called autooxidation. However, in most foods there are several prooxidation systems that generate free radicals and lipid hydroperoxides. These compounds are additional to those formed during the classic initiation and propagation steps.

Prooxidants are found in virtually all food systems. These compounds initiate, facilitate or accelerate lipid oxidation. Many prooxidants are not true catalysts because they are altered during the reaction. Metal ions are especially good examples of prooxidants. For example, the ferrous iron is converted to ferric iron during its interactions with hydroperoxides. Singlet oxygen is converted to a hydroperoxide upon interaction with unsaturated fatty acids.

Hydroperoxides are significant and important substrates for rancidity because of the manner of their decomposition. Their breakdown results in the scission of the fatty acid to produce the low molecular weight volatile compounds that produce off-aromas. Prooxidants can accelerate lipid oxidation by directly interacting with unsaturated fatty acids to form lipid hydroperoxides. That reaction can be seen with lipoxygenases and singlet oxygen for example. Another example in the promotion and  formation of free radicals is by transition metals or the action of ultraviolet light which promotes hydroperoxide decomposition.

Mechanism Of Autooxidation in Edible Oil

Fatty acids or acylglycerols are usually in their nonradical singlet states in food and oils. The autoxidation of oils however requires fatty acids or acylglycerols to be initially present in their radical forms. The reaction of fatty acids with the radical state of atmospheric oxygen which is a triplet-state O2 is thermodynamically unfavourable due to electronic spin conservation (Min and Bradley, 1992).

Singlet Oxygen Quenchers

Singlet oxygen (¹O₂) is an excited state of oxygen that can be formed by enzymatic reactions in biological systems or in the presence of a photosensitizer, light, and triplet oxygen (Davies, 2003; Davies and Truscott, 2001; Decker, 2002).

Singlet oxygen can be inactivated through both chemical and physical quenching pathways by a wide range of compounds. The majority of these are antioxidants.  Amino acids and small peptides, tocopherols and carotenoids including β-carotene, lycopene, and lutein, polyphenols such as the catechins, and flavonoids, proteins, urate, and ascorbate (Decker, 2002; Mukai et al., 2005 a & b; Nagai et al., 2005; Shixian et al., 2005; Bradley and Min, 1999; Kanofsky, 1990).

Antioxidants For Oils

Most antioxidants in current use are synthetic such as butylated hydroxyl anisole (BHA), butylated hydroxyl toluene (BHT), propyl gallate (PG), dodecyl gallate (DOG) and tetra butyl hydroquinone (TBHQ). These are used individually or in combination but must not exceed 0.02% by weight in the final product in India for example.

Natural antioxidants include tocopherols and carotenes, various herb oils such as rosemary, thyme and sage.

Measurement Of Oxidation In Edible Oils Stability 

Numerous analytical methods exist for the routine measurement of lipid oxidation in edible oils (Shahidi and Zhong, 2005). It seems there is no uniform or standard method for detecting all the oxidation reactions that occur in food. The choice of the analytical method depends on the foodstuff.

There are four groups of measures.

• the absorption of oxygen,
• the loss of initial substrates,
• the formation of free radicals,
• the formation of primary and secondary oxidation products.

For each one there are a number of physical and chemical tests used for the measurement of various lipid oxidation parameters. The methods have also been applied to edible oil degradation.  We have:

• Weight-gain based on oxygen absorption
• Headspace oxygen uptake method for oxygen absorption
• Iodometric titration and peroxide values (PV)
• Ferric ion complexation
• Fourier transform infrared (FTIR) methods
• Dielectric spectroscopy
• Chromatographic analysis measuring changes in reactants or formation of products
• Measures of total polar compounds (TPCs)
• 2-thiobarbituric (TBA) acid also known as the TBARS method.
• p-anisidine values (p-AnV)
• carbonyl values
• spectrometric assays on conjugated dienes and trienes including the determination of the total oxidation value (Totox).
• Oxidative Stability instruments

Some techniques are based on differential scanning calorimetry (DSC) and nuclear magnetic resonance (NMR).

The final measure is sensory testing to provide both subjective and objective assessments of the level of oxidative deterioration.

The TBARS method is discussed elsewhere on this domain but we add it for comparison with other methods. The oxidative stability of edible oils is often determined using the oxidative stability index (OSI) and the differential scanning calorimeter (DSC).

The Methods

The Peroxide Value/Hydroperoxide Value

The leading method for many of the oxidation state of an oil. It is accurate, widely used, simple and economical. Suited to evaluation of the early phases and stages of lipid oxidation but not really effective for monitoring frying. There is usually an increase in the PV in the early stage of frying followed by a decline. The hydroperoxides that are formed either react further or thermally breakdown when the temperature of the frying oil for example rises above 180ºC as secondary oxidation products are formed ((Man & Hussain, 1998; Lalas, 2009)

It is an AOCS titration method which needs extracted oils and fats for analysis with solvent removal. The method is defined as ISO3960:2010.

A typical method requires dissolving the oil sample in acetic acid/chloroform and then reacting it with 15% potassium iodide. Liberated iodine is titrated with 0.N sodium thiosulphate.

A modified FOXII (xylenol orange) method is also used for liquid and solid samples. Has extremely high sensitivity. It does not need extraction of oil or fats.  The IDF variant uses thiocyanate. Good for liquid and solid samples. Has high sensitivity. Again like the FOXII it does not need extracted oils or fats.

With all these variants, the method is ideal for fresh or pure bulk oils and fats, emulsions and above all seafood. However results need interpretation as a number of factors impact measurement. It has low sensitivity and is also not easy to validate. A range of units are used from mEq/kg to micromoles/kg.

When antioxidants are added, the amount of peroxides formed is always limited because they mop up  any free radicals especially in the very early phases of frying (Zhang et al., 2012).

The Iodine Value

Iodine values are useful for measuring the degree of unsaturation i.e. level of double bonds in the fatty acid structure. Iodine adds across the double bond in unsaturated fatty acids. The higher the value the better from the perspective of edible oil quality.

Method: An oil sample is treated with a defined amount of chloroform and iodine solution. The reaction is left long enough for the iodine to react with the (unsaturated) double bonds in the fatty acids. The whole solution must be allowed to react in the dark. Any remaining iodine is converted to iodide anions using 15% aqueous potassium iodide. The whole mixture is titrated against sodium thiosulphate using starch as the end-point indicator.

Method referenced as AOCS Cd 1-25/93; AOAC 28.023.

Any form of heating causes a reduction in levels of unsaturation and has been observed in all edible oils which have such fatty acids.  

The Acid Value

A good measure of potential rancidity. For oils it is usually quoted as the free fatty acid value (FFA content).

The method is quoted as ISO660: 2009 and by the AOCS as Method Cd 3d-63.

A typical method relies on heating the oil sample with excess ethanol to boiling. The sample is cooled and then titrated with a standard potassium or sodium hydroxide solution using phenolphthalein as a standard endpoint indicator. Other methods involve titrating a solution of the oil in diethyl ether with an alcoholic solution of sodium or potassium hydroxide.

The acid value is quoted as:

  Acid value  =  (V * N * M.wt)/W

where V = volume of standard KOH solution in mL, N = normality of standard KOH solution, W = weight of oil sample in grams, M.wt (molecular weight) of KOH = 56.1 g/mol.

Typical acid values from 1 to 3 mg/KOH/g oil  need to be achieved. The permissible level for an oil is above 0.6mg KOH/g. An alternative is to quote the measures as a percentage of oleic acid.

A level of 2 per cent acidity is enough to reject an oil (Matthaus, 2006).

In some cases, the acid value can be converted to a a free fatty acid content (FFA) by multiplying the acid value with a factor equal to the molecular weight of the fatty acid of interest.

FT-IR methods have been tried for determining an acid value based on the O-H stretching band (Jiang et al., 2016). 

The Oxipres (Oxygen Bomb) or Oxidography method for oils and fats

A widely used technique for dry foods and ingredients but also suitable for bulk oils and fats. The bath temperature is set at 80-100ºC. It is usually needed at 80ºC for products rich in polyunsaturated fatty acids such as fish meals. It is set higher for baked and extruded goods, dry pet foods and fried foods. Oxidography is mainly used for measuring bulk oils and fats.

The Checkmate or Headspace Oxygen Measurement

Ideal for solid and liquid samples. It involves analysis of headspace oxygen levels at ambient temperature. Ideal for shelf-life studies of samples stored at different temperatures. The issue with the Checkmate method is its difficult to validate with complex products because a number of ingredients, especially proteins also absorb oxygen which produces higher values than expected.

OSI (Oil Stability Index)

A good solid and widely used method for bulk oils and fats. Used at a measuring temperature of 100 to 120 Centigrade and mainly for oils containing saturated, mono-, di-unsaturated oils such as sunflower oil and chicken fat. Use at 60 to 80 Centigrade for polyunsaturated oils such as fish oil. There is a modified method available with AMVN using a lower temperature of 50 Centigrade.

The main limitations are the high measuring temperature. Extraction is also needed for solid samples and it has low sensitivity.

The Rancimat method

A well established technique which is very rarely used now given the advent of more advanced methods. Refer to method ISO6886:2009.

TBA Method For Edible Oils Stability

Used for foods containing fat especially meat products and for samples which contains polyunsaturated fatty acids with 3 or more double bonds.

The main limitations are its lack of specificity and sensitivity. Different analytical procedures result in different analytical results. It is interfered with when analysing oxidised samples generating colour.

p-Ansidine Value (p-AV)

Simple and reliable and mainly used for bulk oils and fats. It is a measure of secondary oxidation of lipids. It relies on the formation of a yellow coloured Schiff base between the reactant and the carbonyl group of compounds formed in the degradation of the oil. These will be non-volatile aldehydes (RCHO) and are formed as a product of the secondary degradation of the hydroperoxides of fatty acids (O’Keefe & Pike, 2010).

The lower the value of p-ansidine (p-AV), the lower the level of oxidation. It increases in proportion to the number of frying events. The more frying takes the place, the more the oil is oxidised hence the rise in the p-AV value. Adding antioxidants lowers the p-AV value.

Very much a part of routine analysis and is quoted as reference method ISO6885:2008. 

Generally, a limited method as it is only possible to use with pure bulk oils and fats plus it has low validity. It has been used however as means of quantifying aldehydic compounds.

The hexanal or 2,4-decadienal, hexanoic acid methods

Employed in analysing samples rich in omega-6 fatty acids. Can be correlated to sensory measurements. If a gas chromatography static headspace is used then the method is rapid and good for routine analysis. When a dynamic headspace method is used – has lower reproductivity. Not great for routine analysis. SPME also has good sensitivity and is used for routine analysis.

The limitations mainly revolve around the different pretreatments meted out on the samples and can impact the final result. A standardised method has long been overdue. There are problems with SPME in terms of the space on the fibre. Suffers the same issues as the hexanal method listed above.

Propanal or 1-penten-3-one or 2,4-heptadienal, propanoic acid technique

Used on samples rich with n-3 fatty acids – fish oils, algal oils and Krill oils.

Total Polar Material (TPM)

During oil breakdown compounds are formed which are more polar than the triacylglycerols in fat. These are the total polar materials (TPM).

TPMs are formed by hydrolysis of the cooking fat releasing fatty acids, monoglycerides and diglycerides. As hydrolysis proceeds various derivatives are formed which are more reactive than the triglycerides from which they were derived. The free radical reactions taking place all promote further oil degradation.

The level of TPM formation is dependent on a few factors: 

• the amount of water in a food that reacts with the frying oil or fat
• the surface to volume ratio of the fried food. The greater the value of this ration the higher the contact between the oil and the water from the product.
• any residual solid particles remaining in the oil accelerate the formation of free fatty acids
• the temperature of the frying temperatures of 360 to 400 ºF produce insignificant hydrolysis because moisture is removed as steam. The highest (and thus worst situation) level of reaction occurs during heating and cooling when the temperature is less than 200 °F. 

The TPM mass concentration is a good indicator of frying fat quality. It is often measured before frying commences to check the quality of the oil being used (Fritsch, 1981; Al-Kahtani, 1991). The formation of these compounds rises with repeated frying and the level depends on the degree of oil unsaturation to begin with (Takeoka et al., 1997; Romero et al., 1998).

There is a situation that when oxidation reactions predominate thermal changes to the oil that sensory issues such as fishiness appear before a critical TPM value is reached. Linoleic acid is notable for this effect.

TPM formation is also influenced by the important measure of the ratio of the surface oil area to oil volume. Specific surface is also important (Bracoo et al., 1981).

The current analytical method relies on fat and oil being separated into polar and non-polar fractions using preparative column chromatography (PCC). 

The standard method is defined in ISO 8420:2002 and as AOCS Cd 20-91. The frying oils are separated into their nonpolar and polar components. The nonpolar compounds are eluted away. The polar compounds are determined by calculating the difference between the weights of the oil samples prepared and added to the column and the weight of the eluted nonpolar fraction.

The main issue with the method is the need to standardise the silica gel used in the column to the correct condition so that it completely separates the polar compounds. It is time consuming and uses hazardous reagents especially solvents (Melton et al., 1994).  The chromatographic step has been replaced with accelerated solvent extraction (ASE) (Zainal & Isengard, 2010) to some limited extent.

Commercial monitoring systems are available for TPMs where batch fryers are being operated. The Testo 270 (Testo Ltd. Alton. Hamps. UK) is a compact hand-held device which relies on a sensor that is placed in the hot oil during frying. It relies on a metal probe inserted into the oil to measure capacitance change. It is sturdy and easily cleaned and gives a reading in seconds. It produces an audible and optical alarm to show when the TPM level has been exceeded. The measuring range is between 0 and 40% TPM. It is claimed to reduced cooking oil use by up to 20%.

An alternative supplier of equipment is suppled by Vito AG (Tuttingen, Germany).

The Oleotest ™ is another rapid method of value which relies on off-line analysis. A sample of oil is added to the pre-conditioned test tube. This is heated in a microwave if the sample is collected from cooled oil. A colour reaction is produced after thorough mixing and the results compared with a standard colour chart which indicates the amount of TPM present in the oil sample.

Total Polar Compounds (TPCs)

The total polar compounds could be considered a more extensive collection of compounds described earlier as TPMs. The TPCs collectively represent various products resulting from oxidation reactions in the oil medium. These include but are not limited to triglycerides and other partial glycerides, free fatty acids, ketones, aldehydes, and other oxidized components.

Like TPMs they are polar and frequently impact the total electrical polarization of the oil medium.

They are monitored using electrochemical impedance spectroscopy (Grossi & Riccò, 2017).  

Conjugated Dienes

Polyunsaturated fatty acids often contain the basic 1,4-pentadiene structure which can be analysed for. The molecular structure has a carbon atom at the bisallylic position. The hydrogen atom at the carbon is highly reactive compared with the other hydrogen atoms in the hydrocarbon chain. 

The universities use this measure for oil and fats measurement when extracted from other lipids. Usually applied in a clinical sense for analysis of dienes in tissue samples. Really not good for quantitative comparison and only sometimes exploited in the food industry.

The Schaal Oven Test

Used mainly for bulk oils and fats. It enables measurement of PV values over time at 60 Centigrade and is coupled to sensory testing. The issue is that some solid samples do not work with the test.

Antioxidant Measurement

Generally the HPLC methods are used for measuring antioxidant levels. You can see changes such as a loss in oils and fats in a range of foods. Ideal for evaluating changes in stored foods. For example used for tocopherols, carnosic acids, epigallocatechins, BHA and BHT and ethoxyquin.

Stability Testing And Accelerated Shelf-Life Of Edible Oils

Storage tests which examine the degradation of oils in real time are often too long for most product developers. Accelerated shelf-life testing which examines the rate of oxidation is often employed. Here changes in temperature or the addition of prooxidants helps. The Schaal Oven Test (carried out at 40–60°C) gives results close to storage conditions, but the test takes several weeks or even months.

A number of combination devices are available which record p-anisidine values, peroxide values and free fatty acids in a few minutes such as the CDR FoodLab system.

Polymeric Triacylglycerols (PTGs)

As with TPMs, the polymeric triacylglycerols which form during frying have to be measured off-line using high performance liquid chromatography (HPLC). The oil is treated with a tetrahydrofuran solution and the PTGs monitored using a refractive index detector.

Determination Of Fatty Acid Contents, Sterol Contents And Pesticide Residues

The determination of componentry in edible oils which can alter during cooking and frying is an important measure where specific properties regarding abuse are concerned. Abuse includes adulteration with other oils but is also a useful quality control method and an aid to identification.

Sterols, pesticides and the fatty acid composition are widely measured using gas chromatography (GC) and high performance liquid chromatography (HPLC). Oils can also be screened rapidly using vibrational spectroscopy including Fourier transform-infrared (FTIR) or FT-Raman for both fatty acid composition and for free acidity in relatively few minutes.

FT-NIR Methods

Near infrared spectrometry (NIR) has also been useful mainly for identification of adulteration but also to monitor changes in acidity, bitterness and a host of analytical measures all referenced earlier (Calero et al., 2018). 

Fourier-transform NIR methods are much more affordable and easily operated in the food industry. It is simple to use, there are no complex sample pretreatments, the measurements are as rapid as any current chemical tests on the market and there is no pollution (Hein et al., 1998). It can be applied to solids and pastes as well as liquids with simultaneous determination of many different components.

The analysis is usually performed off-line where samples are withdrawn and passed into disposable vials with temperature control at 50ºC.

The FT-NIR method was used to measure four analytical criteria such as TPM, acid value, the formation of dimeric and polymeric triglycerides. The main issue was poor statistical calibration on the TPCs and the various polymeric triglycerides (Buning-Pfaue & Kehraus, 2001).

There is a great deal of spectral information between 4590 and 7000 cm-1 which has bearing on FFA content. It is thought to be due to the formation of the -COOH group. It is not clear if these regions are related to the release of FFAs from triglycerides through lipid hydrolysis due to frying. The spectra for total polar material usually shows a characteristic hump at a wavenumber at 4570/cm^-1.

There is a strong hump or rough peak at 6780 cm-1 which is thought to be due to the -OH group in the forming hydroperoxide. This will disappear of course as these radicals react further or evaporate away (Holman & Edmondson, 1956; Nawar, 1996).

Bands at  around 5796 and 5678 cm⁻¹ are due to the combination bands and first overtone of C–H of methylene of aliphatic groups of oil, and their second overtone is observed at around 8264 cm⁻¹ (Workman & Meyer, 2009). The bands at around 4663 and 4591 cm⁻¹ may be attributed to combination bands of C = C and C–H stretching vibrations of cis unsaturated fatty acids, and at 7074 and 7180 cm⁻¹ are attributed to C–H combination band of methylene (Cozzolino et al., 2005).

Online analysis is possible of free fatty acids and polar materials in oils withdrawn from the fryer at temperatures from 80°C and above. Withdrawn samples are initially characterised using a T-filter with an 8 channel mechanical multiplexer, cooled down and then passed to a tube heater at 65ºC before monitored using a 100mm flow cell. The measurement is made and analysed accordingly.  

A rapid near-infrared (NIR) spectroscopy was used to monitor the changes in total polar materials (TPMs) and free fatty acid (FFAs) levels. This application examined changes to soy frying oils and has been tested with other frying applications. Models using both forward stepwise multiple linear regression (FSMLR) and partial least-squares (PLS) regression techniques (Kazemi et al., 2005; Ng et al., 2007, 2011).

The formation of polymerized triacylglycerides has for example been monitored with NIR using PLS regression (Kuligowski et al., 2012). The calibration samples are probably the most important features of the method generally.

One study has investigated the use of a portable NIR spectrometer for quantifying adulteration of olive oils. It would also have some value as a cheap method for analysing samples taken from frying operations (Borghi et al., 2020).

NIR in whatever form appears to be a technique for the future and there is the added bonus that it can be used to monitor changes in oil quality on a continuous basis.

Mid-Infrared Spectroscopy

A technique that complements NIR. As with the latter range on light absorption, MIR relies on a dipole moment which is usually present in diatomic and complex molecules. Any absorptions of light here are due to fundamental molecular vibrations. Mid-IR offers detailed information on molecular structures including the trans configuration of C=C bonds in unsaturated fats.

It appears that using FT-IR generally in a single reflection ATR system is most appropriate. 

The key absorption band is at 1743 cm⁻¹ and due to C = O stretching of aliphatic esters. There are strong bands at around 2922 and 2852 cm⁻¹. These are ascribed to asymmetrical and symmetrical C–H stretching vibrations of CH₂ groups and the band at around 1159 cm⁻¹ may be assigned to the stretching of the C–O bonds of aliphatic esters  (Koca et al., 2010; Sinelli et al., 2010; Du et al., 2012).

Du et al., (2012) describe a weak absorption at 966 cm⁻¹ which may be the C–H out‐of‐plane deformation of isolated trans double bonds or some trans conjugated unsaturated fatty acids (Mossoba et al., 2009). This peak increases to some extent as the oil is increasingly heated. It is much more apparent as the second derivative transformed spectra of frying oils. Soybean oils were used by Du et al., (2012) and they speculated that as they are rich in unsaturated fatty acids such as oleic (C18:1) and linoleic acids (C18:2), these would become hydrogenated at frying temperatures or isomerize to form trans fatty acids.

One disadvantage of MIR is that it cannot penetrate far into the sample and cannot even penetrate glass, plastic or other materials used to hold the sample. This is one of the principal reasons why NIR is preferred.

If it is possible the technique picks up on the overtones of O-H, N-H, C-H, and S-H bands. 

Proton (¹H)NMR Spectroscopy 

The compounds in oil can be both qualitatively and quantitatively characterized (Guillén & Ruiz, 2003, 2004; Guillén & Uriarte, 2012; Martínez‐Yusta et al., 2014).  The technique has very strong possibilities as an off-line monitoring procedure  because not only is the rate of degradation of oil followed but so is the formation of both primary and secondary oxidation products.

The technique is used to monitor changes especially of unsaturated acyl groups. The technique of ³¹P NMR can be used to monitor changes in the diacylglycerols (Zhu et al., 2020).

A study using low-field nuclear magnetic resonance spectroscopy (Low-field NMR) and non-probe-based fluorescence spectroscopy. also appears to have some credibility (Gu et al., 2020a). The fluorescence probe uses iron tetraphenylporphyrin (FeTPP). Another  probe-based three-dimensional fluorescence spectroscopy method was developed for monitoring TPCs (Gu et al. 2020b). This uses tetraphenylporphyrin manganese (MnTPP) because of its selectivity.

A method using a  D-A type triphenylamine derivative (CPA-TPA) has also been tested with claimed success in quantitative determinations of TPC (Cui et al., 2020).

Dielectric Spectroscopy

Frying oils can be monitored using dielectric spectroscopy over a frequency range of between 40 Hz–100 MHz (Yang et al., 2016). Polar components of any sort derived from the oil itself or from the frying food will alter the dielectric constant of the oil. A Food Oil Sensor (FOS) is an instrument that detects these changes and has long been explored for its value (Wegmüller, 1994).

A capacitive sensor which can monitor the quality change of edible oil during repeated cooking has also been explored. It measures the changes of oil permittivity with so described low-cost and fast detection (Paiter et al., 2015, Stevan et al., 2015).

Frequency domain dielectric spectroscopy (FDDS) provides information about the dielectric response of materials to the electromagnetic fields. The fundamental electrical response can be reflected by the complex dielectric permittivity of the material ε*. 

Effect Of Heating/Frying On Edible Oils Stability

A number of reviews are available exploring frying effects on edible oil stability. One review (Aladedunye, 2015) stability and performance into two broad groups. There were external factors such as frying temperature, frying time, presence of oxygen, and type of fryers, among others, which are factors that can be manipulated by operators. The second category were internal or endogenous factors which are oil‐specific and include fatty acid composition (as in the type of oil) and their distribution on triacylglycerols, and the amounts and composition of other minor components. Excellent reviews are available such as that by Nayak et al., (2016).

Man and Hussin (1998) explored and defined the physicochemical properties of refined, bleached, and deodorized palm olein and coconut oil during intermittent frying of potato chips at 180ºC for 5 hours per day for 5 consecutive days.

In terms of kinetics of oxidation of vegetable oils, it is easier to treat the reaction as roughly first order. One study examined the breakdown of crude palm oil, refined canola oil and a 1:1 blend. The oil was deep fried at 170, 180 and 190°C  over a period of 20 hours (Mba et al., 2016). The kinetics behind changes in  free fatty acid (FFA), peroxide value (PV), anisidine value (p-AV), total polar compounds (TPC) and color index (CI) were monitored.

In this simple model, the FFA and PV followed a first-order kinetic model whilst p-AV, TPC and CI followed the kinetic zero order model. All the rate constants increased with increasing frying temperature. The effect of temperature could be modelled using an Arrhenius equation.

What was most noticeable was that the peroxide value had the lowest activation energy Ea (kJ/mol) and the highest was the formation of the free fatty acids. The 1:1 blend was found to be superior to the individual frying oils used. Clearly, the type of frying oil is a high priority factor and dictates oil stability (Kazemi et al., 1999).

There is for example an increase in the degree of hydrogenization with heating.

One investigation used ATR-FTIR on four different oils which were heated for 36 hours at 170 °C and sampled every 6 hours. As well as using standard analysis (Mahboubifar et al., 2016). Partial least squares (PLS) regression combined with a genetic algorithm and an orthogonal signal correction was performed on the spectroscopic data. It led to a predictive model for the various parameters and was suggested as a rapid and economical technique for oil analysis.

The total amount of polar compounds is a good measure of degradation and this can be assessed using a capacitive sensor probe (Khaled et al.,2015). There is also good agreement between the oil electrical capacitance and the level of these compounds.

One of the consequences of heating such as frying are the changes that take place to the main unsaturated fatty acids in oil. Typical examples include oleic acid, linoleic acid and α-linolenic acid. They are precursors to the production of aldehydes which are often derived from frying. All three have different susceptibilities to heat-induced oxidation and they all produce different types and levels of lipid oxidation products (Choe et al., 2006).

A variety of aldehydes are generated which in turn go on to react with food components that are being fried too. Unsaturated aldehydes formed in frying oil will react with primary amines which are formed by breakdown in vegetables such as potatoes (as in french fries) and with amino-acids (Kim et al., 1996; Kim & Ho, 1998; Uchida, 2003). Thus fried food also modifies the oil flavour by removing compounds like aldehydes (Wang et al., 2016).

A number of compounds formed from oils involve the degradation of polyunsaturated acyl groups, such as oxygenated α,β-unsaturated aldehydes. A particularly toxic aldehyde from this group is called 4-hydroxynonenal (Seppanen et al., 2002; Csallany et al., 2015) which is formed at frying oil temperatures can be removed if it is allowed to combine with amines. Other toxic aldehydes to look out for in frying oils include malondialdehyde (MDA) and 4-hydroxy-2-hexenal (HHE). All these are thought to be involved in the generation of cancer and various degenerative diseases such as Alzheimer’s, Parkinson or even atherosclerosis.

The presence of all these are best monitored using reversed-phase-high-performance liquid chromatography (RP-HPLC) coupled with photodiode array detector (PAD) (Ma & Liu, 2017). The best recourse is to add antioxidants which significantly reduce aldehyde formation (Liu et al., 2020). 

Some aldehydes formed from fats eventually evaporate because they are relatively volatile. For example acrolein and trans-2-pentenal will start to disappear.

Glycidyl esters (GEs) of fatty acids also decrease with frying. One study looked at palm oil and the frying of potato chips which were heated daily for 8 hours over five consecutive days and three frying temperatures: 150, 165 and 180 °C. The GEs were monitored using liquid chromatography-mass spectrometry. In this frying study, hydrolysis, oxidation and polymerization reactions all increased with increasing frying temperature as expected.

In palm oil, the most abundant GEs are esters of palmitic and oleic acid. As temperature and frying time increase, the amount of GEs declines and is correlated with degree of oil degradation (Aniołowska & Kita, 2016).

Some of the minor compounds from oil oxidation have been monitored using gas chromatography/mass spectrometry (DI-SPME-GC/MS)  (Alberdi-Cedeño et al., 2019). 

Dielectric spectroscopy as we referenced is not often discussed but has been tested on soybean oil used for frying dough with different moisture levels (Yang et al., 2016). This technology exploits the way oils breakdown in relation to their level of dielectric relaxation. The relaxation time is dictated by the number and size of the polar components in the oil.

Oil degradation can also be seen with changes in its viscosity and absorption characteristics as indicated by changes to oil surface tension. Measures of this type might be extremely useful in assessing how well an oil continues to fry products like chips for example (Sahasrabudhe et al., 2017; 2019; Yang et al., 2020).

Oxidative stability is measured using a number of the parameters mentioned above. The increase in stability here is associated with an increase in hydrogenization (Kazemi et al., 1999). The type of oil dictates the level of oxidative stability. Extra virgin olive oil is one of the most stable followed by high oleic acid sunflower oils with added antioxidants.

One study checking the effectiveness of viscometry used a 3D-Printed multichannel viscometer which was claimed to be highly effective (Oh et al., 2018).

An online system using a UWB integrated reflectometer is feasible (Ziga et al., 2015).

Regulations On the Composition Of Oils And Fats Especially For Frying

A number of European countries now have composition regulations on oils and fats for frying. Germany, Belgium, Austria, Hungary, Spain, Sweden, France and Switzerland have standards mainly related to TPC levels but a few have included the formation of polymers and triglycerides.

TPMs are thought to be harmful for health. Some countries have set a maximum limit of TPM in frying oil of between 24 and 30%. In Germany the amount of TPM has to be lower than 24% and the oil is discarded if above this value.

Another measure used regularly is the polymeric triglyceride (PTG) level and is often quoted in combination with the TPM value. The PTG value should not exceed 12% for frying oils and a few European countries consider the TPM and the PTG values together should not exceed 24 to 27%.

In South Korea the legal rejection limit of frying oil is 3.0 AV (MFDS, 2016).

One of the current issues is a lack of standardization between the various countries. 

References On Edible Oils Stability

Aladedunye, F. A. (2015). Curbing thermo‐oxidative degradation of frying oils: Current knowledge and challenges. European Journal of Lipid Science and Technology, 117(11), pp. 1867-1881.

Alberdi-Cedeño, J., Ibargoitia, M. L., & Guillén, M. D. (2019). Monitoring of minor compounds in corn oil oxidation by direct immersion-solid phase microextraction-gas chromatography/mass spectrometry. New oil oxidation markers. Food Chemistry, 290, pp. 286-294.

Al-Kahtani, H.A. (1991) Survey of quality of used frying oils from restaurants. J. Am. Oil Chem. Soc. 68 pp. 857-862

Aniołowska, M., & Kita, A. (2016). The effect of frying on glycidyl esters content in palm oil. Food Chemistry, 203, pp. 95-103

Borghi, F. T., Santos, P. C., Santos, F. D., Nascimento, M. H., Corrêa, T., Cesconetto, M., … & Filgueiras, P. R. (2020). Quantification and classification of vegetable oils in extra virgin olive oil samples using a portable near-infrared spectrometer associated with chemometrics. Microchemical Journal, 159, 105544. 

Buning-Pfaue, H., Kehraus, S. (2001) Application of near infrared spectroscopy (NIRS) in the analysis of frying fats. Eur. J. Lipid Sci.Tech. 103 pp. 793-797

Chaiyasit, W., Elias, R. J., McClements, D. J., & Decker, E. A. (2007). Role of physical structures in bulk oils on lipid oxidation. Critical Reviews in Food Science and Nutrition, 47(3), pp. 299-317

Calero, A. M., Muñoz, E., Pérez-Marin, D., Riccioli, C., Pérez, L., & Garrido-Varo, A. (2018). Evolution of Frying Oil Quality Using Fourier Transform Near-Infrared (FT-NIR) Spectroscopy. Applied Spectroscopy, 72(7), pp. 1001-1013.

Chen, X., Yu, X., Wang, Y., Yang, Y., & Zhang, J. (2015). Determination of polar components in frying oils by Fourier-transform near-infrared spectroscopy. Journal of Oleo Science, ess14227 (Article).

Choe, E., & Min, D. B. (2006). Mechanisms and factors for edible oil oxidation. Comprehensive Reviews In Food Science and Food Safety, 5(4), pp. 169-186.

Costa Pereira, A.F., Coelho Pontes, M.J., Gambarra Neto, F.F., Bezerra Santos, S.R., Harrop Galva, R.K., Ugulino Araujo, M.C. (2008) NIR spectrometric determination of quality parameters in vegetable oils using iPLS and variable selection. Food Res. Int. 41 pp. 341–348

Cozzolino, D., Murray, I., Chree, A., Scaife, J.R. (2005). Multivariate determination of free fatty acids and moisture in fish oils by partial least‐squares regression and near‐infrared spectroscopy. LWT 38 pp. 821–8 (Article)

Csallany, A. S.; Han, I.; Shoeman, D. W.; Chen, C.; Yuan, J. Y. (2015) 4-Hydroxynonenal (HNE), a toxic aldehyde in French fries from fast food restaurants. J. Am. Oil Chem. Soc. 92, pp. 1413−1419

Cui, S., Wang, B., Yan, X., Li, Y., Zhou, X., Wang, Y., & Chen, L. (2020). A novel emitter: Sensing mechanical stimuli and monitoring total polar materials in frying oil. Dyes and Pigments, 174, 108020 (Article).

Decker, E.A. 2002. Antioxidant mechanisms. In: Akoh, C.C., and Min, D.B., Eds., Food Lipids: Chemistry, Nutrition, and Biotechnology. Marcel Dekker, New York

Du, R., Lai, K., Xiao, Z., Shen, Y., Wang, X., Huang, Y. (2012) Evaluation of the quality of deep frying oils with Fourier transform near-infrared and mid-infrared spectroscopy. J Food Sci. Feb;77(2):C261-6. (Article) Epub 2012 Jan 17. PMID: 22251019.

Fritsch, C. (1981) Measurements of frying fat deterioration: a brief review. J. Am. Oil Chem. Soc. 58 pp. 272-274

Gu, H., Huang, X., Sun, Y., Chen, Q., Wei, Z., & Lv, R. (2020). Intelligent evaluation of total polar compounds (TPC) content of frying oil based on fluorescence spectroscopy and low-field NMR. Food Chemistry, 128242 (Article)

Gu, H., Huang, X., Chen, Q. et al. (2020b) Rapid Assessment of Total Polar Material in Used Frying Oils Using Manganese Tetraphenylporphyrin Fluorescent Sensor with Enhanced Sensitivity. Food Anal. Methods 13, pp. 2080–2086 (Article)

Guillén, M. D., & Ruiz, A. (2003). 1H nuclear magnetic resonance as a fast tool for determining the composition of acyl chains in acylglycerol mixtures. European Journal of Lipid Science and Technology, 105(9), pp. 502-507 (Article)

Guillén, M. D., & Ruiz, A. (2004). Formation of hydroperoxy‐and hydroxyalkenals during thermal oxidative degradation of sesame oil monitored by proton NMR. European Journal of Lipid Science and Technology, 106(10), pp. 680-687

Guillén, M. D., & Uriarte, P. S. (2012). Study by 1H NMR spectroscopy of the evolution of extra virgin olive oil composition submitted to frying temperature in an industrial fryer for a prolonged period of time. Food Chemistry, 134(1), pp. 162-172 (Article).

Hein, M., Henning, H., Isengard, H.D. (1998) Determination of total polar parts with new methods for the quality survey of frying fats and oils. Talanta 47, pp. 447-454  

Holman, R.T., Edmondson, P.R. (1956). Near‐infrared spectra of fatty acids and some related substance. Anal. Chem. 28 pp. 1533–8.

Jiang, X., Li, S., Xiang, G., Li, Q., Fan, L., he, L., Gu, K. (2016) Determination of the acid values of edible oils via FTIR spectroscopy based on the O-H stretching band. Food Chem., 212 pp. 585-589 (Article)

Juarez, M.D., Osawa, C.C., Acuna, M.E., Samman, N., Goncalves, L.A.G. (2011) Degradation in soybean oil and partially hydrogenated fats after food frying, monitored by conventional and unconventional methods. Food Control 22 pp. 1920–1927

Kalogianni, E.P., Karapantsios, T.D., Miller, R. (2011) Effect of repeated frying on the viscosity, density and dynamic interfacial tension of palm and olive oil. J. Food Eng. 105 pp. 169–179

Kazemi, S., Wang, N., Ngadi, M., Prasher, S.O., (2005) Evaluation of frying oil quality using VIS/NIR hyperspectral analysis. Int. Comm. Agr. Eng. 7 pp. 1-12

Khaled, A. Y., Abd Aziz, S., & Rokhani, F. Z. (2015). Capacitive sensor probe to assess frying oil degradation. Information Processing in Agriculture, 2(2), pp. 142-148 (Article)

Kim, Y. S.; Hartman, T. G.; Ho, C. T. (1996) Formation of 2-pentylpyridine from the thermal interaction of amino acids and 2,4-decadienal. J. Agric. Food Chem. 44, pp. 3906−3908

Kim, Y. S.; Ho, C. T. (1998) Formation of pyridines from thermal interaction of glutamine or glutamic acid with a mixture of alkadienals in aqueous and oil media. J. Food Lipids 5, pp. 173−182

Koca, N., Kocaoglu‐Vurma, N.A., Harper, W.J., Rodriguez‐Saona, L.E. (2010) Application of temperature‐controlled attenuated total reflectance‐mid‐infrared (ATR‐MIR) spectroscopy for rapid estimation of butter adulteration. Food Chem. 121 pp. 778–82

Kuligowski, J., Carrion D, Quintas G, Garrigues S, de la Guardia, M. (2012) Direct determination of polymerised triacylglycerides in deep-frying vegetable oil by near infrared spectroscopy using
partial least square regression. Food Chem. 131 pp. 353–359

Lalas., S. (2009) Quality of Frying Oil. In: S.G. Sahin, S. Sumnu, editors. Advances in Deep-Fat Frying of Foods. Boca Raton, FL: CRC Press, pp.57–81.

Liu, Z. Y., Zhou, D. Y., Li, A., Zhao, M. T., Hu, Y. Y., Li, D. Y., … & Shahidi, F. (2020). Effects of temperature and heating time on the formation of aldehydes during the frying process of clam assessed by an HPLC-MS/MS method. Food Chemistry, 308, 125650.

Ma, L., & Liu, G. (2017). Simultaneous analysis of malondialdehyde, 4-hydroxy-2-hexenal, and 4-hydroxy-2-nonenal in vegetable oil by reversed-phase high-performance liquid chromatography. Journal of Agricultural and Food Chemistry, 65(51), pp. 11320-11328 (Article).

Mahboubifar, M., Yousefinejad, S., Alizadeh, M. et al. (2016) Prediction of the acid value, peroxide value and the percentage of some fatty acids in edible oils during long heating time by chemometrics analysis of FTIR-ATR spectra. J. Iran Chem Soc. 13, pp. 2291–2299 (Article)

Man, C. and Hussain, W. (1998). Comparison of frying and performance of refined bleached and deodorised palm oliein and coconut oil. J. Food Lipids 5, pp. 197–210 (Article)

Martínez‐Yusta, A., Goicoechea, E., & Guillén, M. D. (2014). A review of thermo‐oxidative degradation of food lipids studied by 1H NMR spectroscopy: influence of degradative conditions and food lipid nature. Comprehensive Reviews in Food Science and Food Safety, 13(5), pp. 838-859

Matthaüs, B. (2006) Utilization of High-Oleic Rapeseed Oil for Deep-Fat Frying of French Fries Compared to Other Commonly Used Edible Oils. Eur. J. Lipid Sci. Technol. 108(3): pp. 200–211  

Mba, O.I., Dumont, M.J., Ngadi, M. (2016) Deterioration kinetics of crude palm oil, canola oil and blend during repeated deep-fat frying. J. Am. Oil Chem. Soc. 93: pp. 1243–1253

Melton, S., Jafar, S., Sykes, D., Trigiano, M. (1994) Review of stability measurements for frying oils and fried food flavor. J. Am. Oil Chem. Soc. 71 pp. 1301-1308

MFDS (2016) Ministry of Food and Drug Safety. Food Code. Available from: http://www.foodsafetykorea.go.kr/foodcode/index.jsp. Accessed 17th November 2020

Min, D.B., Bradley, G.D. (1992) Fats and oils: flavors. In: Hui YH, editor. Wiley Encyclopedia of Food Science and Technology. New York: John Wiley & Sons. Pp. 828–32

Mossoba, M.M., Seiler, A., Kramer, J.K.G., Milosevic, V., Milosevic, M., Azizian, H., Steinhart, H. (2009) Nutrition labeling: rapid determination of total trans fats by using Internal reaction infrared spectroscopy and a second derivative procedure. J. Am. Chem. Soc. 86 pp. 1037–45 (Article).

Mukai, K., Nagai, S., and Ohara, K. 2005. Kinetic study of the quenching reaction of singlet oxygen by tea catechins in ethanol solution. Free Radic. Biol. Med., 39(6): pp. 753–761.

Mukai, K., Noborio, S., and Nagaoka, S. 2005. Why is the order reversed? Peroxyl-scavenging activity and fats-and-oils protecting activity of vitamin E. Int. J. Chem. Kinetics., 37(10) pp. 605–610

Nawar, W.W. (1996) Lipids. In: Fennema, O.R., editor. Food Chemistry. New York : Marcel Dekker. pp. 292–5.

Ng, C.L., Wehling, R.L., Cuppet, S.L. (2007) Method for determining frying oil degradation by near-infrared spectroscopy. J Agric. Food Chem. 55 pp. 593-597

________________________________ (2011) Near-infrared spectroscopic determination of degradation in vegetable oils used to fry various foods. J Agric. Food Chem. 59 pp. 12286–12290

Öğütcü, M., Aydeniz, B., Büyükcan, M. B., & Yılmaz, E. (2012). Determining frying oil degradation by near infrared spectroscopy using chemometric techniques. Journal of the American Oil Chemists’ Society, 89(10), pp. 1823-1830

Oh, S., Kim, B., Choi, S. (2018) A 3D-Printed Multichannel Viscometer for High-Throughput Analysis of Frying Oil Quality. Sensors. 18(5):1625 (Article).

O’Keefe, S.F., O.A. Pike. O.A. (2010) Fat Characterization. In: S.S. Nielsen, editor. Food Analysis. New York: Springer Science, pp.239–260. .

Romero A., Cuesta C. and Sanchez-Muniz J. 1998. Effect of oil replenishment during deep-fat frying of frozen foods  in sunflower oil and high-oleic acid sunflower oil. J. Am. Oil Chem. Soc., 75: pp. 161-167.

Sahasrabudhe, S. N., Staton, J. A., & Farkas, B. E. (2019). Effect of frying oil degradation on surface tension and wettability. LWT, 99, pp. 519-524 (Article).

Seppanen, C. M.; Csallany, A. S. (2002) Formation of 4-hydroxynonenal, a toxic aldehyde, in soybean oil at frying temperature. J. Am. Oil Chem. Soc.  79, pp.1033−1038. .

Shahidi, F., & Zhong, Y. (2005). Lipid Oxidation: Measurement Methods. Bailey’s industrial oil and fat products. (Article)

Sinelli N, Cerretani L, Egidio VD, Bendini A, Casiraghi E. (2010) Application of near (NIR) infrared and mid (MIR) infrared spectroscopy as a rapid tool to classify extra virgin olive oil on the basis of fruity attribute intensity. Food Res. Int. 43 pp. 369–75.

Takeoka G.R., Full G.H. and Dao L.T. 1997. Effect of heating on the characteristics and chemical composition of selected frying oils and fats. J. Agric. Food Chem., 45: pp. 3244-3249

Tan, C. P., Man, Y. C., Selamat, J., & Yusoff, M. S. A. (2002). Comparative studies of oxidative stability of edible oils by differential scanning calorimetry and oxidative stability index methods. Food Chemistry, 76(3), pp. 385-389

Uchida, K. (2003) Histidine and lysine as targets of oxidative modification. Amino Acids 25, pp. 249−257

Wang, L., Csallany, A. S., Kerr, B. J., Shurson, G. C., & Chen, C. (2016). Kinetics of forming aldehydes in frying oils and their distribution in French fries revealed by LC–MS-based Chemometrics. Journal of Agricultural and Food Chemistry, 64(19), pp. 3881-3889 (Article).

Workman, J.J., Meyer, L. (2009). Alkanes and Cycloalkanes. In: Ren H‐M, editor. Practical Guide to Interpretive Near‐Infrared Spectroscopy. 1st ed. Beijing : Chemical Industry. pp. 25–34.

Yang, J., Zhao, K. S., & He, Y. J. (2016). Quality evaluation of frying oil deterioration by dielectric spectroscopy. Journal of Food Engineering, 180, pp. 69-76.

Yavari, A.A., Hamedi, M., Haghbin, S. (2009) Investigation of frying oil quality using vis/NIR hyperspectral analysis. J. Am. Oil Chem. Soc. 86 pp. 941–947 (now retracted because the authors wholly plagiarised the paper by Kazemi et al., of 2005).

Zainal, Isengard, H.-D. (2010) Determination of total polar material in frying oil using accelerated solvent extraction. Lipid Technol,. 22 (6) pp. 134-136 (Article)

Zhang, Q., Saleh, A.S.M., Chen, J., Shen. Q. (2012) Chemical Alterations Taken Place During Deep-Fat Frying Based on Certain Reaction Products: A Review. Chem. Phys. Lipids. 165(6): pp. 662–681.

Žiga, M., Galajda, P., Slovák, S., & Kmec, M. (2015, June). Determination of the quality of frying oil based on UWB impedance spectrometer. In 2015 16th International Radar Symposium (IRS) (pp. 955-960). IEEE.

Zribi,A., Jabeur, H., Aladedunye, F., Rebai, A., Matthaus, B., Bouaziz. M. (2014) Monitoring of Quality and Stability Characteristics and Fatty Acid Compositions of Refined Olive and Seed Oils during Repeated Pan- and Deep-Frying Using GC, FT-NIRS, and Chemometrics. J. Agric. Food Chem. 62(42): pp. 10357–10367

Amendment: Original article of 7th September 2022. Addition of extra sections on new analytical methods and updated references.