Changes during deep fat frying

Frying fats: understanding chemical processes occurring during frying and the methods used to control oil degradation can greatly improve product flavor and shelflife

Prepared Foods, August, 2006

The following article has been condensed and adapted from Lipid Oxidation by Edwin N. Frankel, Chapter 12, "Frying Fats." For more information, see this article's end.--Eds.

Deep-fat frying is perhaps the most complex of food processing operations because of the multiplicity of reactions that occur and the vast quantity of chemical products that are generated. Understanding the chemical processes that occur during frying and the methods available to control oil degradation can have a big impact on finished product flavor and shelflife.

Frying is usually done in the 180[degrees]-200[degrees]C range. There are three major chemical reactions that occur: oxidation, polymerization and hydrolysis. Each of these reactions produces a wide range of compounds which contribute to the formation of both desirable and undesirable products in fried foods.

Thermal lipid oxidation and hydrolysis produce complex mixtures of volatile and non-volatile monomeric and polymeric substances. Polar materials are polymers, which are useful indicators of oil quality. The food itself also releases proteins, carbohydrates, fats, oils and metals into the frying medium that may promote reactions affecting oil performance as well as its health and nutritional properties.

The rapid decomposition of hydroperoxides and aldehydes during frying produces an intricate assortment of volatile compounds. However, they are found in very low concentrations because the frying process acts like a large steam distillation operation. Those that remain are important because many at extremely low levels (in parts per million) contribute to the distinctive flavors and aromas so typical of fried foods.

The chart "Various Volatiles" shows some of the many compounds found in frying oils and/or French fries. One of the compounds listed in this table, 2,4-decadienal, a decomposition product of linoleate that produces rich, fried food flavor, was a significant flavor noted during sensory evaluation of French fries. The presence of tins linoleate in vegetable oils contributes to the formation of 2,4-decadienal. When frying potato chips, research indicates that a certain level of linoleate is essential for desirable flavor development. The downside is that the presence of linoleate also increases the probability that rancidity will occur. Because decadienal imparts a desirable "fried flavor" at low concentrations, but too much can cause an undesirable "rancid" flavor, the frying process also must be carefully controlled by a turnover process to introduce fresh make-up oil during frying.

Oil Sources and Evaluations

Both fats and oils of animal and vegetable origin are used for frying. General considerations involved in the selection of frying oil include the turnover rate (i.e., oil added to a fryer to replace that removed by the fried products), the amount of fat absorbed by the food and the shelflife of the food. For high-volume frying, oils should not exceed 2%-3% linolenic acid, a polyunsaturated fatty acid that is readily oxidized and polymerized. The process of partial hydrogenation was aimed at reducing the linolenic acid content of soybean and canola oils. However, partial hydrogenation has the problem of producing trans isomers that are nutritionally undesirable. Many frying fats are developed by blending oils with much lower degrees of hydrogenation to meet the recent FDA requirement of labeling the content of trans fats in foods. In response to this issue, plant breeders have produced zero-trans oils with higher oleic acid content and reduced linolenic acid to improve stability during frying without hydrogenation.

There are many ways to measure the changes that occur in oils during frying. As frying progresses, the oil darkens, becomes more viscous and tends to foam, and the degree of unsaturation drops. Chemical methods which estimate the amount of polar materials, total polymeric materials and/or free fatty acids also are routinely used to monitor oil degradation.

Polar materials are measured by column chromatography. This is a time-consuming, expensive method and consequently is not suitable for quality control. To reduce costs and minimize the amount of solvent used, rapid micro chromatographic methods have been developed.

The rate of polar material formation varies with the type of frying operation. Oils used for restaurant frying can contain up to 30% polar materials after 40 hours of frying. The reason for this rapid buildup is that batch frying is done intermittently and oil turnover is slow (i.e., oil is not frequently replaced). Turnover in industrial frying operations is much greater due to the volume of food being prepared. The frying fat generally reaches a steady-state condition after a few hours of continuous operation. The amount of polar material produced by the three common frying operations can be compared in the chart "Frying Operations and Amount of Polar Compounds."

There also are a number of rapid test methods available. These include a range of commercial colorimetric tests based on redox indicators, carbonyl compounds, free fatty acids and alkaline materials. Unfortunately, these methods are not specific, and [this author suggests they are not to be] recommended as tools for quality monitoring. The ultimate arbiter of frying, however, is in the taste and quality of the fried food. The most sensitive and reliable methods for monitoring frying oil degradation are based on sensory evaluation of the foods and gas chromatographic analysis of the volatile flavor compounds in the foods, preferably potatoes fresh and after storage.

Control Measures

A number of control mechanisms are available to enhance food quality and extend the usable frying life of the cooking oil. These include monitoring and controlling frying oil temperatures,

increasing turnover rates, filtering the oil, using additives such as antioxidants and polymerization inhibitors and controlling fat uptake by the foods.

Optimal frying is conducted at temperatures ranging from 160[degrees]-190[degrees]C, and frying at lower temperatures will minimize thermal degradation of the oil. However, sufficiently hot oil is essential to ensure that food is thoroughly cooked. The continuous emanation of bubbles from frying food is proof that moisture is escaping as steam. Steam release helps strip volatile decomposition products from the oil, which delays fat deterioration. To reduce the potential for thermal degradation and oil breakdown, the temperature of the oil can be lowered when not in use.

Filtration is used to remove charred food particles and batter that accumulate in the oil. Removal of these materials can slow oil degradation, maintain oil color and minimize the development of off-flavors. Metal screens, paper and plastic are used as filter materials. Filtration can be augmented by using filter aids such as diatomaceous earth.

The use of additives can enhance both oil life and the shelflife of flied foods. Antioxidants such as BHA, BHT, propyl gallate and TBHQ may be included in commercially used frying fats and oils. These oils also may include naturally occurring or added tocopherols, which generally are depleted during the frying process. Natural antioxidants such as rosemary and sage also may be added to flying oils. Like the synthetic antioxidants noted above, these also can protect the oil during frying and enhance shelflife of fried food.

The amount of fat absorbed by the food during frying affects its sensory perceptions and acceptability. The chemical makeup of the frying oil and that picked up by the food are not significantly different, and minimizing the rate of oil degradation can reduce the amount of oil picked up by the food.

Deep-fat frying is perhaps the most complex processing operation known. The frying oil is a dynamic entity that changes depending upon the food being fried and the time of frying. And although oil degradation helps produce some of the delicious flavors that make fried foods so palatable, understanding its processes and the means to control them is essential to producing healthy and nutritious food.

This article was condensed and adapted, in part, by Rick Stier from Lipid Oxidation, 2nd ed., written by Edwin N. Frankel, University of California, USA, and published by The Oily Press, PJ Barnes & Associates. Published March 200.5. 486 pages, 152 tables, 148 figures, 87 equations, 849 references. Volume 18 in The Oily Press Lipid Library. For more information, call: 44-1823-698973; e-mail: sales@pjbames.co.uk. See also www.pjbarnes.co.uk/ol)/lo2.htm.

Various Volatiles

Compounds Found Description (b)in French Fries (a)

Hexanal Greentrans, trans-2,4-decadienal Deep-fried2-octenal Tallowy, nutty

Nonanal Fatty, greentrans-2-nonenal Tallowy, greencis, trans-2,4-decadienal Deep-friedDimethyltrisulphide Cabbage2-ethyl-3,6-dimethylpyrazine Roasty, earthy

Source: Derived from Table 12.2, pg. 359, Lipid Oxidation,The Oily Press, Bridgwater, England

a) From Belitz and Grosch, 1986

b) From Wagner and Grosh, 1997COPYRIGHT 2006 BNP Media COPYRIGHT 2008 Gale, Cengage Learning

http://users.auth.gr/~karapant/tdk/Research/Project15c.html

Influence of gravity conditions on mass and heat transfer in porous media

Contact person: J. Lioumbas [lioumbas@gmail.com] and A. Zamanis [agzamanis@yahoo.gr] 

Transport through porous media has gained great attention in the last decades due to its relevance to a wide range of applications such as electronics cooling, thermal insulation engineering, water movement in geothermal reservoirs, heat pipes, underground spreading of chemical waste, nuclear waste repository, geothermal engineering, grain storage and enhanced recovery of petroleum reservoirs. In many of the aforementioned problems of interest, the porous medium consists of single-phase (liquid) as well as two-phase (liquid + vapor) regions as a result of evaporation due to vapour pressure i.e. below the liquid’s boiling temperature. One common and easy practice in everyday life process which is closely related to transport phenomena through porous media is frying. Frying is an extremely complex process and involves unsteady state coupled heat and mass transfer problem in porous media (the food item can be considered as such), phase change of water, bubble formation and bubbly flow combined with unsteady state heat transfer in the oil bulk.

During frying we are dealing with “unconventional” boiling on the surface of a fluid-saturated porous material, since the bubbles are not formed solely because of the rapid vaporization of a liquid -heated to its boiling point- on the porous material surface, but are formed mainly due to the heat transfer from the hot oil (that surround the porous medium) to the -entrapped within the porous structure- water (with temperature below the boiling point) in a region close to porous interface (Figure 1). Specifically, the penetrating liquid not only wets and fills the porous network but also heats it up, while gases or vapors need to escape from the porous network to allow room for liquid penetration. Such gas/vapor departure occurs in the form of bubbles which depart from the outside surface of the cold porous material to the surrounding hot liquid. The bubble formation and departure from the porous material as well as natural convection of the unevenly heated liquid is expected to significantly be affected by gravity. The problem becomes even more complicated if one takes into account the dramatic changes in the properties of the food being fried (e.g. thermal properties, change in the concentration of components inside the porous food and structural changes such as porosity and pore sizes changes) and the frying medium (e.g. changes of the physical and physical - chemical properties of oil). Figure 2 presents schematically the scientific areas interrelated with the process of frying.

We expect that the experimental results of the present activity (obtained by meticulous on-line temperature recordings inside natural and artificial porous materials, on-line water loss measurements and high speed video recordings of the generated bubbles during the frying process in terrestrial, hypergravity and microgravity conditions) in conjunction with modeling efforts would contribute to elucidate the observed phenomena during the conventional heat and mass transfer in porous media in one hand, and on the other hand will offer a new insight in

studying the conventional frying process with a novel perspective that includes heat and mass transfer in porous media.

 

Figure 1: Geometry of the porous layer, the crust is a porous region whereas the core is an amorphous region.

Figure 2: Schematic diagram of present project various interrelated scientific

fields.

International Journal of Signal System Control and Engineering Application Year: 2009 | Volume: 2 | Issue: 2 | Page No.: 35-39 DOI: 10.3923/ijssceapp.2009.35.39  

Use of Ultrasonics for the Quality Assessment of Frying Oil Driss Izbaim, Bouaazza Faiz, Ali Moudden, Naima Taifi and Adil Hamine  

http://www.medwelljournals.com/fulltext/?doi=ijssceapp.2009.35.39

Abstract: The main objective of this study is to evaluate quality changes of Soybean Oil (SBO) during frying. Traditional methods are expensive and time consuming techniques, requiring technical personnel and laboratory facilities. Fatty Acids (FA) value, total Polar Component (PC) and Free Fatty Acids (FFA) of oil samples were used as chemical indicators of different quality levels of oil and then compared with ultrasonic measurements. The study demonstrated that using ultrasonic properties, we obtain reliable results to monitor and control oil quality.

   

INTRODUCTION

Frying has become the most popular food preparation technology due to its flavour and the easy preparation. Important characteristics of industrial frying oils are high oxidative stability, high smoke-point, low foaming, low melting point, bland flavour and nutritional value. However, complex physical and chemical changes occur during deep-fat frying leading to thermal and oxidative decomposition. Physical changes are mainly increased viscosity and foaming, colour changes and decreased smoke-point. Main chemical changes are increased free fatty acids and polar components as well as decreased levels of unsaturation, flavour quality and nutritive value (Warner, 2002).

Pokorny (1989) has demonstrated that increases in the polar fraction resulted in an eventual degradation in food quality. Many countries consider the polar compounds measurement to be the single most important test for the degradation state of oil and have established the value of around 25% as a its regulatory limits in frying oils (Firestone, 1996), some other countries also use free fatty acids. Free fatty acids produced in the frying operation of an oil contribute to the smoke haze and therefore has a substantial effect on its smoke point, which affects oil absorption by the fried food (Orthoefer et al., 1996).

There are several methods for controlling and assessing degradation of oil during deep frying, but are time-consuming, costly and usually require analytical expertise. In this respect, ultrasonic techniques are fast, non-invasive and convenient. Many studies have been conducted to assess the composition of different types of food products using ultrasound (McClements, 1997; Mulet et al., 1999). Velocity is the ultrasonic parameter most used because of its reliable results.

The objective of this study is about the changes in ultrasonic measurements during thermoxidation of SBO and the relationships between ultrasonic measurements and chemical results.

MATERIALS AND METHODS

SBO was obtained from Cristal S.A. (Casablanca, Morocco). The oils was heated in a domestic fryer at 180°C for 8 h day-1 over 4 days, for a total of 32 h. Samples of 150 mL were periodically removed and kept at -18°C for further analysis.

Ultrasonic measurements: The experimental setup (Fig. 1) used for the experiments consisted of ultrasonic transducer (5 MHz, 0.5 inch crystal diameter, A309S-SU model, panametrics, Olympus), attached to a cubic container (50x50x50 mm), where the oil samples were placed. The container was introduced into a temperature controlled bath to maintain the sample temperature and the oil was moderately stirred to avoid formation of bubbles. The ultrasonic measurements were carried out, while the oil sample was cooled from 50-30°C. The transducer was linked to a pulser-receiver (Sofranel Model 5073 PR, Sofranel Instruments), which sent the electrical signal to a digital storage oscilloscope (LeCroy 9310 M, LeCroy Cor).

The ultrasonic impulse is propagated in water and crosses the oil sample contained in the vessel before reflecting on a Plexiglas surface of plate 2 (Fig. 2). Figure 3 shows, the experimental signal composed of echoes A1 to A4.

Fig. 1: Experimental set-up

Fig. 2: Schematic of echoes reflected by the sample (A1 to A6 are observed echoes reflected from the interfaces between media Mi)

Fig. 3: Typical waveform of the reflected ultrasonic signal with the 5 MHz transducer

For ultrasonic velocity measurement, eight signal acquisitions were taken and averaged. It can be written as:

(1)

with ω = 2πv

To obtain the phase experimentally, the FFT of signals A2 and A4 are calculated (Bakkali et al., 2001).

The attenuation coefficient α was computed by fitting the experimental data to equation:

(2)

Where:

d = The distance travelled by the wave

A0 = The initial amplitude of the signal measured as the peak-to-peak voltage

A = The amplitude of the signal at distance (d)

Eight ultrasonic echoes were considered to compute attenuation.

Chemical analysis:Fatty acids: The FA profile analysis was performed by derivatization to their corresponding methyl esters (Hartman and Lago, 1973) prior to the analysis by GC. Oil samples (50 mL) are diluted in the hexane to obtain a solution to be analyzed (about approximately 0.1%).

A VARIAN 3800 chromatograph on a CP Select CB (VARIAN) capillary column (50 m x 0.25 mm i.d., film thickness 0.25 μm), was used under the following temperature program: 185°C (40 min), 15-250°C/10 min. Samples injection is 1 μL (split ratio 1:100) at 250°C and the flow rate of Helium, used as carrier gas, was 1.2 mL min-1. Temperature of both split injector and flame ionization detector was 250°C.

Free fatty acids: FFA content as the percentage of oleic acid was determined using AFNOR NF T 60-204 standard method. Acid value was defined as the amount (mg) of KOH required to neutralize FFA in 1 g of oil sample dissolved in a mixture of diethyl ether and ethanol in the presence of phenolphthalein.

Polar compounds: The content of total polar compounds was determined following the method proposed by the IUPAC (1992).

RESULTS AND DISCUSSION

Chemical characteristics: Table 1 shows the results for the Fatty Acids (FA) profiles of the SBO carried out on the oil samples at different heating periods. The FA profile of the frying oils changed as a result of cyclization, polymerization and hydrolytic, oxidative and other chemical reactions promoted by frying conditions (Nawar, 1996). The degradation affected unsaturated fatty acids more than saturated fatty acids. The FA composition of particular oil has marked effects on its frying performance as well as on its physical and chemical behaviour. The linoleic acid level in deep-frying oils appears to be an obviously negative factor in oil stability. Indeed, previous studies indicated that a lowered linoleic acid content in soybean oil resulted in improved oil quality during cooking and frying (Tompkins and Perkins, 2000). Changes in the FA profile during frying provide only limited information about these compositional changes, which are associated with oil degradation.

The regular limits in many countries for PC on frying fats and oils are around 25% and in some others are around 0.3% for FFA as the official regulations.

During oxidation and hydrolysis, FFA are formed, as a result of the cleavage of triglycerides (Perkins and Erickson, 1996). FFA content is the most frequently used test, but it is not recommended to use it as the only indicator of oil quality. Oils with high FFA are known to have a lower smoke point (Augustin et al., 1987) and the surfactant effect of FFA contributes to the foaming, which leads to further oxidation of the oil. Previous studies of frying oils have shown that the FFA content increases during deep-frying (Kalapathy and Proctor, 2000). As expected, FFA content of the SBO increased significantly during frying and after 20 h, its content is 0.31% (Table 2).

Many researchers consider measurement of TCP to be one of the most reliable indicators of the state of the oil deterioration (Fritch, 1981; Gere, 1982). According to Billek et al. (1978) and Paradis and Nawar (1981), polar compounds indicate the degradation of oils and the breakdown of triglycerides. Table 2 shows that TPC in SBO increased significantly during frying. After 32 h of frying, the final TPC level is 25%. The TPC-based stability is 32 h of frying for SBO not to exceed the established limit. This would have occurred at even shorter time if the oil was used for frying foods swamp.

Ultrasonic measurements: The ultrasonic velocity and attenuation depend on the physico-chemical properties of the medium. Ultrasonic velocity decreases with the temperature in fat (McClements, 1997). Figure 4 shows the influence of temperature on the ultrasonic velocity measurements, for 8, 16, 24 and 32 h.

Table 1:

Soybean oil FA compounds at several frying times (%)

Table 2:

Soybean oil polar compounds and free fatty acids at several frying times (%)

Fig. 4: Variation of velocity with temperature for different frying times

Fig. 5: Variation of velocity with time of cooling for different frying times

As expected, the ultrasonic velocity decreases in line as the temperature increases. The average velocity temperature coefficient is -3.78 m/sec/°C. On the other hand, the ultrasonic velocity increases as the time of cooling increases (Fig. 5).

The time of cooling allows making the difference between the various frying. Every sample of the oil is differently cooled from others according to the duration of frying. Therefore, we can distinguish two fryings without using the temperature.

Fig. 6: Relationship between the temperature and the time of cooling for different frying times (±1°C)

Fig. 7: Variation of attenuation with temperature for different frying times

The ultrasonic velocity is related to the temperature, which in its turn is related to the time of cooling. Figure 6 shows the relationship between the temperature and the time of cooling. This curve is almost the same one for the various fryings.

The ultrasonic attenuation is also affected by temperature (Fig. 7). However, velocity is the ultrasonic parameter most used because of its reliable results (Benedito et al., 2002).

It is not easy to determine the useful life of oil because, it depends on many factors, especially the composition of the oil and the type of oil used. Significant polynomial fits were found when relating the ultrasonic measurements and the chemical parameters.

Figure 8a and b show the possibility to detect the limits of the degrees of the polar components and free fatty acids (PC = 22%, FFA (as oleic acid) = 0.37%) using the ultrasonic measurements. The ultrasonic measurements are related to the chemical results, which show the feasibility of using an automated ultrasonic system to monitor the oil quality.

Fig. 8a: Relationship between the ultrasonic measurements and the percentage of free fatty acids at 30°C

Fig. 8b: Relationship between the ultrasonic measurements and the percentage of polar compounds at 30°C

CONCLUSION

The chemical changes are linked to changes in ultrasonic parameters, which show that ultrasound can be used to assess and monitor oil quality. The velocity evolution obtained according the time of cooling appears to be also an indicator of oil degradation, in addition to the evolution of velocity according to the temperature.

The feasibility of using ultrasonic techniques to rapidly evaluate quality parameters for heated edible oils was investigated.

Degradation and Nutritional Quality Changes of Oil During Frying

JAOCS, Journal of the American Oil Chemists' Society, Feb 2009 by Aladedunye, Felix A, Przybylski, Roman http://findarticles.com/p/articles/mi_hb5762/is_200902/ai_n32316879/

Abstract

The changes in regular canola oil as affected by frying temperature were studied. French fries were fried intermittently in canola oil that was heated for 7 h daily over seven consecutive days. Thermo-oxidative alterations of the oil heated at 185 ± 5 or 215 ± 5 °C were measured by total polar components (TPC), anisidine value (AV), color components formation, and changes in

fatty acid composition and tocopherols. Results showed that TPC, AV, color and trans fatty acid content increased significantly (P

Keywords Canola oil * Frying performance * Total polar component * Anisidine value * Color * Frying temperature * Tocopherols * French fries * Fatty acids

Introduction

Deep-fat frying is probably one of the most dynamic processes in all of food processing. Essentially, the process involves immersing a food item in a large quantity of heated oil or fat, which is normally replenished and reused several times before being disposed. Deep-fat frying produces a product with desired sensory characteristics, including fried food flavor, golden brown color, and a crisp texture [1].

Most frying operations are conducted at temperatures of 175-195 °C, nevertheless German regulations allow maximal frying temperatures of up to 165 °C, to limit formation of acrylamides [2]. Extruded products and pellets are typically fried at 190-215 °C [2]. This high temperature requirement and the presence of air and moisture, from the food, initiate several chemical and physical changes affecting oxidative degradation of oil used. Published studies described chemical reactions involved and various volatile and non-volatile oxidation products were identified [3-6]. The chemical changes in the frying fats also affect the physical characteristics of the oil and fried product [7]. For instance, the color of frying oil was reported to darken as a result of oxidation and the formation of browning pigments when potato chips were fried [8, 9].

A number of studies have been undertaken to assess various chemical reactions and extent of oxidative deterioration as affected by frying temperature, many of the published data were obtained by heating an oil but not during actual frying [10-12]. Meanwhile, it has been observed that the chemical reactions that take place during deep-fat frying are different from those during continuous heating [13, 14], Besides, different oils have been found to behave differently regarding the rate of formation of polar components and secondary oxidation products. Guillen and Cabo [15] reported that secondary products were formed immediately after hydroperoxide formation in olive and rapeseed oils, whereas in sunflower and safflower oils, secondary products were formed when the concentration of hydroperoxides reached level of 180 and 270 meq/kg, respectively. Consequently, the need to study the frying performance of individual oil as a function of frying temperature during actual frying of food becomes imperative.

Oxidized short-chain fatty acids are secondary oxidation products formed through thermal degradation of lipid hydroperoxides. Recently, much concern has been on the biological effects of oxidized lipids, and there is increasing evidence that they may be detrimental to health, especially in connection with the development of atherosclerosis, liver damage, and promotion of intestinal tumors [16].

The objective of this study was to evaluate the effect of frying temperature on the degradation of canola oil by monitoring the accumulation of total polar components, oxidized short-chain fatty acids, polymers formation, p-anisidine value, color components formation, changes in fatty acid composition, and tocopherol contents.

Materials and Methods

Materials

Oil and French Fries

Commercially refined regular canola oil without antioxidants added was obtained from Richardson Oilseed Processing (Lethbridge, Canada). Frozen par-fried French fries in an institutional pack were obtained from a local food store.

Chemicals

All solvents and chemicals of analytical grade used in this study were purchased from Sigma-Aldrich (St Louis, MO). Standards of tocopherols were obtained from CalbiochemNovabiochem (San Diego, CA). Standards of fatty acid methyl esters were purchased from Nu-Check-Prep (Elysian, MN).

Frying Procedure and Oil Sampling

The frying was simultaneously conducted in two 8-L capacity restaurant style stainless steel deep fryers (General Electric Company, New York, USA). Regular canola oil (3.75 L) was heated at 185 ± 5 and 215 ± 5 °C for 7 h daily for 7 days. A batch of 200 g of frozen French fries was fried for 5 min for a total of eight batches per frying day. At the end of each frying day, fryers were shut off and left to cool overnight. Two 25-mL samples of oil from each of the fryers were taken daily and kept frozen at -16 °C until analyzed. Before commencing frying each day, oils were filtered to remove solid debris. Oil was replenished every second day of frying with 500 mL of fresh oil.

Fatty Acid Analysis

Fatty acids were methylated following the AOCS Official Method Ce 1-62 [17]. The resulting fatty acid methyl esters (FAME) were analyzed on Trace GC Ultra gas Chromatograph (Thermo Electron Corporation, Rodano, Italy) using a Trace TR-FAME fused silica capillary column (100 m × 0.25 mm × 0.25 µm; ThermoFisher Scientific, Waltham, MA, USA). Hydrogen was used as the carrier gas with a flow rate of 1.5 mL min^sup -1^. The column temperature was programmed from 70 to 160 °C at 25 °C min^sup -1^ and held for 30 min, and further programmed to 210 °C at 3 °C min^sup -1^. Starting and final temperatures were held for 5 and 30 min, respectively. Splitless injection was made using a PTV injector. Detector temperature was set at 250 °C. FAME samples, 1 µL, were injected with an AS 3000 autosampler (Thermo Electron Corporation, Rodano, Italy). Fatty acids were identified by comparison of retention time with authentic standards. Oxidized short-chain fatty acids methyl esters (OFAME) were identified as a group as described by Velasco et al. [18]. Trans isomers of fatty acids were assessed according to ISO method 15304.

Total Polar Compounds and Anisidine Value

TPC were determined by gravimetric method after column chromatography separation of non-polar fraction following AOAC Method 982.27 [19]. Polar components were eluted from the column with diisopropyl ether and further analyzed for polar components composition by size exclusion chromatography.

Anisidine values (AV), a measure of secondary oxidation products, was determined according to ISO Method 6885:2004 [20].

Tocopherols

Tocopherols were analyzed by AOCS Official Method Ce 8-89 [17]. Briefly, oil samples (75 mg) were weighed directly into vials and dissolved in 1.5 mL hexane. Analysis was performed on a Finnigan Surveyor liquid Chromatograph (LC) (Thermo Electron Corporation, Rodano, Italy) with a Finnigan Surveyor Autosampler Plus and Finnigan Surveyor FL Plus fluorescence detector, set for excitation at 292 nm and emission 394 nm. The column was a normal-phase Microsorb 100 silica column (3 µm; 250 × 4 mm; Varian, CA). Of each sample, 10 µL was injected. Mobile phase consisted of 7% methyl-teri-butyl-ether in hexane with a flow rate of 0.6 mL/min. The amounts of tocopherols were quantified using calibration curves for each isomer separately.

Size Exclusion Chromatography

The composition of polar components was analyzed using high performance size exclusion chromatography (HPSEC) according to ISO Method 16931 [21]. Separation was performed on a Finnigan Surveyor LC. Components were separated on three size exclusion columns in series (Phenogel 500 Å, 100 Å and 50 Å, 5 µm, 300 × 4.60 mm; Phenomenex, Torrance, CA), with tetrahydrofuran (THF) as the mobile phase at a flow rate of 0.3 mL/min, and column temperature of 30 °C. A 10 µL sample was injected, and components were detected with a Sedex 75 evaporative light scattering detector (Sedere, Alfortville, France), operated at 30 °C with an air pressure of 2.5 bar. Polar components were identified according to the method described by Marquez-Ruiz et al. [22].

Color Analysis

Color of the frying oils was determined according to AOCS Official method Cc 13c-50 [17] using a DU®-65 spectrophotometer (Beckman, Fullerton, CA).

Statistical Analysis

Samples from three repetitions of frying at each temperature were collected and were analyzed in triplicate. Data are presented as mean values ± SD. Data were analyzed by single factor analysis of variance (ANOVA) and regression analyses using Minitab 2000 statistical software (Minitab Inc., PA, ver. 13.2). Statistically significant differences between means were determined by Duncan's multiple range tests. Statistically significant differences were determined at the P

Results and Discussion

The fresh oil had 0.06% of free fatty acids (FFA), 1.0 meq/kg of peroxide value (PV), 4.2% of polar components and an AV at the level of 4.2, indicating good quality oil [23]. Thus, changes in these values during frying would indicate a degradation in oil quality.

Total Polar Compounds

The determination of TPC in frying oil provides the most reliable measure of the extent of oxidative degradation [14, 24]. In this study, the contents of TPC increased almost linearly with the frying time at a rate affected by frying temperature (Fig. 1). Total polar content during frying at 185 °C was 19.8% at the end of frying time, which was still below the 24% oil discard level set in many European countries [25]. However, the total amount of polar components reached the discard level after 4 days of frying at 215 °C. The TPC reached 38% by the end of the frying time. The extent of oxidative deterioration, as measured by TPC formation, was 2.6 times faster during frying at 215 °C compared to 185 °C.

Composition of Polar Components

The composition of polar compounds formed during frying was analyzed using HPSEC. Diacylglycerides (DG), oxidized triacylglycerides (OTG), dimers and polymers were separated, and their contribution calculated using peak area. The contribution of polymers in total polar material increased consistently with frying time at both frying temperatures achieving maximum values of 8 and 15.6% for frying at 185 and 215 °C, respectively (Figs. 2 and 3). The amount of polymers generated at 185 °C at the end of the 7 days frying period was comparable to the third day of frying at 215 °C. Comparable increase in the amount of dimers for oil fried at 185 °C was observed throughout the frying period (Fig. 2). However, when frying at 215 °C, a 16-fold increase in the contribution of dimers at the end of first day followed by slight increase for the next 2 days of frying, then decreased until the end of the frying period (Fig. 3). This is probably due to the conversion of the dimers to polymers and thermal degradation of these components [16]. As expected, the contribution of OTG decreased consistently over the frying period at both tested temperatures, as a consequence of thermal degradation. However, a more pronounced decrease in the contribution of OTG, 1.5 times faster degradation was observed during frying at 215 °C (Fig. 3).

Anisidine Values

Aldehydes formed during oxidative degradation are secondary decomposition products, and the non-volatile portion of carbonyls remains in the frying oil [4, 13]. At the two testing temperatures, AV was not well correlated with frying time (Fig. 4). The maximum was reached on the second day of frying for both frying temperatures and then decreased consistently until the end of frying time. Apparently, oil replenishment played some role in the changes in carbonyls content but elevated temperatures was probably the main cause for the reduction in the amount of these labile and reactive components with time [26]. This result could be explained by the thermal degradation of the aldehydes formed at higher temperature, which results in a lower accumulation in the oil at the higher frying temperature. Regardless of the general knowledge

that the decomposition of hydroperoxides increases with increasing temperature and potentially the amount of carbonyls, AV followed an opposite trend in the frying tests. Carbonyl chemical reactivity, involvement in the formation of other compounds and thermal decomposition explains decreasing in AV [10, 25]. Houhoula et al. [27] reported a significant increase in AV as a function of temperature during frying potato chips in cottonseed oil. Thus, the initial increase in AV observed (Fig. 4) agrees with these authors. Furthermore, the increase was also observed as a function of temperature.

Fatty Acid Composition

The fatty acid composition of the fresh canola oil and the resulted changes during the 7 days of frying at 185 and 215 °C are presented in Table 1. The results indicate a progressive decrease in both linoleic and linolenic acids contributions throughout the frying period [24], Linoleic acid decreased by 8.5 and 13.3% during frying at 185 and 215 °C, respectively. The deterioration of linolenic acid was more pronounced and was reduced by 24.0 and 47.1% during frying at 185 and 215 °C, respectively. White et al. [24] reported decreases of 7-11.5% in linoleic acid and 27-46% in linolenic acid when soybean oils were heated at 180 °C for 40 h.

The amount of trans fatty acids formed during frying increased when temperature and time increased (Figs. 5 and 6). At the lower frying temperature applied, the amount of trans isomers increased from 2.4 to 3.3%. The increase in frying temperature to 215 °C caused extensive trans isomerization of fatty acids (Figs. 5 and 6). The total contribution of trans isomers in oil increased 2.5-fold, from 2.4 to 5.9% (Fig. 5). This indicates the importance of temperature on trans isomers formation during frying, and explains the amount of trans isomers observed in the initial oil (Fig. 5). The deodorization of canola oil is usually performed at temperatures above 200 °C under vacuum, where the main amount of trans isomers is formed [28]. The amount of individual trans fatty acids decreased in the following order: linolenic > linoleic > oleic (Figs. 5, 6). The quantity of trans isomers formed at elevated temperature indicates that a specific amount of energy is required to transfer double bonds from cis to trans configuration. Data from this work are supported by published results, confirming that activation energy for isomerization decreasing when the numbers of cis double bonds increases [29].

Increasing amounts of trans isomers during frying at higher temperature, can have practical implications related to nutritional claims about zero trans content in a serving portion of fried products. When the amount of these isomers increases 2.5-fold during frying at the higher temperature, then the amount of trans isomers in fried products will increase by the same amount and may exceed the specified definition limit, making the claim for zero trans fat invalid. The amount of trans isomers in oil can affect the trans level in fried product due to fast exchange of fats during frying. We observed that the oil in fried food had a similar composition of fatty acids as frying oil, even when par-frying was done in different oil (data not included). These data clearly indicate the importance of controlling frying temperature and keeping it below 190 °C.

The ratio of linoleic acid to palmitic acid (C ^sub 18.2^/C^sub 16:0^) has been suggested as a valid indicator of the level of PUFA deterioration [34]. Our result showed a decrease in this ratio from 4.73 to 3.87 and 4.28 to 3.23 during frying at 185 and 215 °C, respectively (Table 1). This implies that the decrease in this ratio was 1.2 times greater in oil heated at 215 °C as compared to

185 °C. Örnal and Ergin [30] reported a decrease in the ratio from 4.04 to 3.49 at the end of frying time. Houhoula et al. [27] reported a reduction of the ratio from 2.39 to 2.03 for cottonseed oil heated at 185 °C for 12 h. The decrease in the ratio of linolenic acid to palmitic acid was more pronounced, reducing it 1.9 times faster in oil heated at 215 °C compared to 185 °C (Table 1).

Short-chain glycerol-bound aldehydes, acids, ketones and alcohols are non-volatile secondary oxidation products formed during oxidative degradation of lipids [18]. They are of particular chemical and nutritional interest since they remain in the frying oil, and are absorbed and subsequently ingested. Analysis of the oxidized short-chain fatty acid methyl esters (OFAME) as a group revealed a consistent increase in their contribution for the first 5 days of frying at 185 °C, reaching a maximum at 1.83% (Fig. 7). However, for the oil heated at 215 °C, the amount of oxidized fatty acids increased in the first 3 days of frying with the maximum at 2.20%. Thereafter, a decrease for the next 3 days of frying was observed. Velasco et al. [31] observed that the amount of polar FAME decreased drastically as compared to the amount of total polar compounds in used frying fats.

Peers and Swoboda [32] suggested the quantification of methyl octanoate, product of linoleic acid degradation, as an oxidation index during frying. In this study, the amount of octanoic acid increased significantly (P Color Analysis

Tocopherols

The tocopherol profile of the fresh canola oil used in this study was found to be: 214 µg/g of a-tocopherol, and 347 µg/g of y-tocopherol. Tocopherols degradation increased as a function of frying temperature. At the end of the seventh day of the frying period, approximately 31% of the total tocopherols remained after frying at 185 °C (Fig. 10). For frying at 215 °C the entire amount of tocopherols was gone at the end of sixth day of frying. The calculated half life of tocopherols for oil heated at 185 °C was 8 h while for frying at 215 °C was 5.3 h. Consistent with previously published results [35], a strong inverse relationship was observed between TPC formation and the reduction of tocopherol at both frying temperatures. In this study, y-tocopherol degraded at a faster rate than a-tocopherol at the lower frying temperature, but the order was reversed during frying at 215 °C (data not shown).

Correlation Between Assessment Parameters

Although TPC remains the best assessment parameter for evaluating frying oil performance and oxidative stability, a faster and yet objective alternative is desirable. Assessment methods that correlate well with TPC may provide the much needed alternative. In this study, a low correlation was found between AV and TPC, and AV and color at both frying temperatures (Table 2). However, a good correlation was observed between color and TPC at both frying temperatures (Table 2). Lopez-Varela et al. [35] reported a correlation coefficient of 0.885 between color and TPC for sunflower oil used in 75 successive fryings of potatoes. In summary, the effect of temperature and frying time on performance of canola oil as measured by TPC, AV, fatty acid composition, tocopherols amount and color was significant.

Conclusions

The rate of the thermal and oxidative degradation of PUFA was drastically higher at the elevated temperature tested, forming larger amounts of components with potential detrimental health effects. This study showed that increasing frying temperature above 195 °C can cause intensive isomerization of PUFA and the amount of trans isomers may increase above the threshold level described by "zero trans definition" annulling the trans fat free claim for fried product.

References

1. Warner K (2004) Chemical and physical reactions in oil during frying. In: Gupta MK, Warner K, White PJ (eds) Frying technology and practice. AOCS, Champaign, pp 16-28

2. Gupta MJ (2004) The effect of oil processing on frying oil stability. In: Gupta MK, Warner K, White PJ (eds) Frying technology and practice. AOCS, Champaign, pp 76-90

3. Chang TY, Eiserich JP, Shibamoto T (1993) Volatile compounds identified in headspace samples of peanut oil heated under temperature ranging from 50 to 200°C. J Agric Food Chem 41: 1467-1470

4. Perkins EG (1996) Volatile odor and flavor components formed in deep frying. In: Perkins EG, Erickson MD (eds) Deep frying: chemistry, nutrition and practical applications. AOCS, Champaign, pp 43-48

5. Stevenson SG, Jeffery L, Vaisey-Genser M, Fyfe B, Hougen FW, Eskin NAM (1984) Performance of canola and soybean fats in extended frying. Can Inst Food Sci Technol J 17:187-194

6. Przybylski R, Eskin NAM (1995) Methods to measure volatile compounds and the flavor significance of volatile compounds. In: Warner K, Eskin NAM (eds) Methods to assess quality stability of oils and fat-containing foods. AOCS, Champaign, pp 107-133

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

8. Al-Kahtani HA (1991) Survey of quality of used frying oils from restaurants. J Am Oil Chem Soc 68:857-862

9. Che Man YB, Tan CP (1999) Effects of natural and synthetic antioxidants in refined, bleached, and deodorized palm olein during repeated deep-fat frying of potato chips. J Am Oil Chem Soc 76:331-339

10. Achir N, Kara W, Chipeaux C, Trezzani I, Cuvelier ME (2006) Effect of energy transfer conditions on the chemical degradation of frying oil. Eur J Lipid Sci Technol 108:999-1006

11. Valdes AF, Garcia AB (2006) A study of the evolution of the physicochemical and structural characteristics of olive and sunflower oils after heating at frying temperatures. Food Chem 98:214-219

12. Fiselier K, Bazzocco D, Gama-Baumgartner F, Grob K (2006) Influence of frying temperature on acrylamide formation in French fries. Eur Food Res Technol 222:414-419

13. Chang SS, Peterson RJ, Ho CT (1978) Chemical reactions involved in deep-fat frying of foods. J Am Oil Chem Soc 55:718-727

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

15. Guillen MD, Cabo N (2002) Fourier transform infrared spectra data versus peroxide and anisidine values to determine oxidative stability of edible oils. Food Chem 77:503-510

16. Dobarganes C, Marquez-Ruiz G (2003) Oxidized fats in foods. Curr Opin Clin Nutr Metab Care 6:157-163

17. Firestone D (1999) Official methods and recommended practices of the American Oil Chemists' Society, 5th edn. AOCS, Champaign

18. Velasco J, Berdeaux O, Marquez-Ruiz G, Dobarganes MC (2002) Sensitive and accurate quantification of monoepoxy fatty acids in thermooxidized oils by gas-liquid chromatography. J Chromatogr A 982:145-152

19. Association of Official Analytical Chemists (1990) Official methods of analysis of the association of official analytical chemists' 15th edn. AOAC Inc. Arlington, Method 982.27

20. International Organization for Standardization (2004) Animal and vegetable fats and oils-determination of anisidine value, ISO, Geneva, Standard No. 6885

21. International Organization for Standardization (2007) Animal and vegetable fats and oils-determination of polymerized triglycerides content by high-performance size-exclusion chromatography (HPSEC), ISO, Geneva, Standard No. 16931

22. Marquez-Ruiz G, Jorge NI, Martfn-Polvillo M, Dobarganes MC (1996) Rapid, quantitative determination of polar compounds in fats and oils by solid-phase extraction and size-exclusion chromatography using monostearin as internal standard. J Chromatogr A 749:55-60

23. Orthoefer FT, Cooper DS (1996) Initial quality of frying oil. In: Perkins G, Erickson MD (eds) Deep frying: chemistry, nutrition and practical applications. AOCS, Champaign, pp 29-42

24. White PJ (1981) Methods for measuring changes in deep-fat frying oils. Food Technol 45:75-80

25. Firestone D, Stier RF, Blumenthal M (1991) Regulation of frying fats and oils. Food Technol 45:90-94

26. Tyagi VK, Vasishtha AK (1996) Changes in the characteristics and composition of oils during deep-fat frying. J Am Oil Chem Soc 73:499-506

27. Houhoula DP, Oreopoulou V, Tzia C (2002) A kinetic study of oil deterioration during frying and a comparison with heating. J Am Oil Chem Soc 79:133-137

28. Fournier V, Destaillats F, Juaneda P, Dionisi F, Lambelet P, Sebedio JL, Berdeaux O (2006) Thermal degradation of longchain polyunsaturated fatty acids during deodorization of fish oil. Eur J Lipid Sci Technol 108:33-42

29. Augustin MA, Asap T, Heng LK (1987) Relationships between measurements of fat deterioration during heating and frying in RBD olein. J Am Oil Chem Soc 64:1670-1675

30. Önal B, Ergin G (2002) Antioxidative effects of α-tocopherol and ascorbyl palmitate on thermal oxidation of canola oil. Nahrung 46:420-426

31. Velasco J, Marmesat S, Marquez-Ruiz G, Dobarganes MC (2004) Formation of short-chain glycerol-bound oxidation products and oxidized monomeric triacylglycerols during deep-frying and occurrence in used frying fats. Eur J lipid Sci Technol 106:728-735

32. Peers KE, Swoboda AT (1982) Deterioration of sunflower seed oil under simulated frying conditions and during small-scale frying of potato chips. J Sci Food Agric 33:389-395

33. Pritchard MK, Adam LR (1994) Relationship between fry color and sugar concentration in stored Russet Burbank and Shepody potatoes. Am Potato J 71:59-68

34. Normand L, Eskin NAM, Przybylski R (2001) Effects of tocopherols on the frying stability of regular and modified canola oils. J Am Oil Chem Soc 78:369-373

35. Lopez-Varela S, Sanchez-Muniz FJ, Garrido-Polonio C, Arroyo R, Cuesta C (1995) Relationship between chemical and physical indexes and column HPSE chromatography methods for evaluating frying oil. Z Ernahrungswiss 34:308-313