Beef (Longissumus dorsi) loin from a single three-year-old cow was obtained from a local supermarket in Gaziantep, Turkey and used as the sample. Longissumus dorsi is the muscle that runs along the top of the spine from hip to shoulder. The meat was frozen at –70 °C until use for cooking. The required meat samples were removed from the freezer and allowed to thaw for one day at 4 °C.

Standard PAH, which is the EPA Method 8310 PAH mixture, and a Pinnacle II PAH 4 μm column (LC columns, 50 × 3.2 mm, USA) were purchased from RESTEK. All chemicals required for calibrations and analyses were UPLC grade. Thiobarbituric acid (TBA), tetramethoxypropane, trichloroaceticacid (TCA) and the rest of the reagents and solvents used in this study were purchased from Sigma-Aldrich.

A standard calibration was prepared using 1,1,3,3-tetramethoxypropane (TMP). First of all, a 166 µl TMP solution was completed to 1 liter with pure water to prepare the stock solution. Then 0.1 ml, 0.2 ml, 0.3 ml, 0.4 ml, 0.5 ml, 0.7 ml and 1.0 ml were taken from the stock solution and each of them was completed to 10 ml with pure water. After that, a 0.5 ml solution was taken into tubes from each dilution and homogenized with 5 ml 10% TCA. Then the homogenized solution was centrifuged for 20 minutes. About 2 ml of supernatant were taken and mixed with 2 ml of 1% thiobarbituricacid (TBA). The mixture was placed in a 95 °C water bath and kept for 30 minutes. At the end of this step, shock cooling was applied to the standard samples. Finally, the absorbance was read at 532 nm against a blank in a spectrophotometer so that a standard curve was obtained.

Stock solutions of the PAH standard (EPA Method 8310 PAH mixture, Restek) were prepared by dissolving it in ultrapure water (EMD Millipore, Merck KGaA, Germany). First of all, 50 µl PAH standard mixture were completed to 50 ml with ultrapure water in order to prepare a stock solution of 500 ppb concentration. Then, one by one 6 ml, 4 ml, 2 ml, 1 ml and 0.5 ml were taken from the stock solution and each of them was completed to 10 ml with pure water. In this way, 300 ppb, 200 ppb, 100 ppb, 50 ppb and 25 ppb concentrations of stock solution were prepared. The 300 ppb stock solution of the PAH chromatogram is shown in Figure 1.

For the designation of the ohmic cooker, a 5 kw conductor, a 220 Volt input – 24 Volt output AC power supply, a limit switch, a 10 A fuse, a voltmeter and an ampermeter were used. The ohmic cooking design is shown in Figure 2. A cylindirical test cell with the dimensions of 5 cm length and 3.5 cm diameter was designed with open edges to hold the sample. Then, electrodes with 3.5 cm diameter that contact and cover the meat’s surface completely were needed. Stainless steel electrodes (430 type) were tested and cooking was very successful. For this system, the meat was cut by a 5 cm diameter sharp cutter and then placed into the test cell. Next, 430 type stainless steel electrodes were placed into the open gaps at the edges of the test cell to completely cover and be in contact with the surface of the meat sample. The homogeneous and uniform cooking of the meat was achieved. 40, 55 and 70 Volt levels were used and the application time was adjusted to 7 minutes for ohmic cooking. Two replicates were performed for each level.

Infrared cooking was performed in a Frigidaire Professional Stainless Programmable 6 Slice Infrared Convection Oven (Frigidaire, FPCO06D7MS 120V-60 Hz 1500 Toaster Oven, US). In infrared cooking, three different temperature settings were used: 162.77 °C (325 °F), 190.55 °C (375 °F) and 218.33 °C (425 °F), as the oven temperature and the temperature in the center of the meat sample was 72 °C. For ohmic cooking followed by infrared cooking, the time of infrared cooking was 3 minutes for each side of the meat sample. Two replicates made for each treatment at every level.

The experimental set-up design is shown in Figure 3. A flow diagram of the experimental design is shown in Figure 4.

The determination of TBARS is the most common method for the measurement of lipid oxidation in meat and consists of the spectrophotometric determination of MDA formation as an index of oxidative status. About 1 g of the meat sample was taken and homogenized with 5 ml of 10% TCA. This homogenized solution was centrifuged for 20 minutes. About 2 ml supernatant were taken and mixed with 2 ml of 1% TBA. The mixture was placed in a water bath at 95 °C and kept for 30 minutes. At the end of this step, shock cooling was applied to standard samples. Finally, absorbance values were read at 532 nm against a blank and TBARS values of the sample were determined with the calibration curve equation. The analyses were performed on duplicate samples. The result of the calibration curve equation was found to be: y= 0.0391x, where y is the absorbance and x is the mg MDA/kg.

The extraction and clean-up procedures were carried out according to the method described by Chung et al., (2011) and Kendirci et al., (2014). A 30 g sample was placed in a 500 ml round flask, and 100 ml KOH (2 M, prepared by using a methanol in water [9:1]) solution and 100 ml of hexane were added. The flask was held in a water bath at 80 ± 2 °C under reflux for 2 hours. Then, it was cooled to approximately 40 °C by adding 100 ml of cold water and allowed to stand overnight in the dark. The hexane phase was transferred into a 250 ml Erlenmeyer flask and concentrated to 2 ml at 50 ± 2 °C using a vacuum oven. The concentrate was purified by moving it through a Sep-Pak florisil cartridge (Vac C18 6 cc, Waters Corporation, USA), which had been pre-conditioned with 10 ml of dichloromethane and 20ml hexane. Elution solvents involving 10 ml hexane and 8 ml hexane:dichloromethane (3:1) were moved through the cartridge. All the eluates were collected, dissolved in 1 ml of acetonitrile, filtered through a 0.45 μm membrane filter (Minisart RC 25, Germany) and collected into 2 ml amber vials.

The PAH analysis was carried out with a UPLC device equipped with a fluorescence detector. A Pinnacle-II PAH 4 μm column (Restek LC columns, 50 × 3.2 mm, USA) was used. The gradient solvent system was started with 50% acetonitrile in water (v/v) and linearly increased to 100% acetonitrile in 30 min at a flowrate of 0.5 ml/min. The excitation and emission wavelengths were 360/460 nm. The external calibration curves of the PAHs were prepared by injecting 10 μl of working solutions of the standards to HPLC. For the quantification of the PAHs, a 10 μl sample extract was injected into the HPLC. The analyses were performed on duplicate samples. The retention times (Rt) and limit of detections (LODs) of PAHs are shown in Table 3.

An ANOVA was performed for TBARS and PAH levels as a function of voltage value for ohmic cooking alone and cooking temperature for the combined ohmic-infrared cooking. SPSS version 21.0 was used to evaluate significant differences (P < 0.05) among the samples. For the identification of the homogeneous groups of cooking methods’ variables, Duncan’s multiple range test was applied. Trends were considered as significant when the means of the compared parameters differed at a P < 0.05 significance level.

Lipid oxidation values in meat, after ohmic cooking alone and ohmic cooking followed by infrared cooking are expressed as mg MDA/kg muscle and presented in Tables 1 and 2. In the ohmic cooking process alone, increasing the voltage value increased the TBARS value significantly (P < 0.05). In ohmic processing, the increase in temperature could be due to the agitation of molecules due to the movement of charges within the material by the flow of electric current through a conductor (Ramaswamy et al., 2014). Min et al., (2006) investigated the effect of different ohmic power intensities (0, 10, 20, 30, 40, 50 V) on the TBARS levels of hamburger patties. They argued that thermal processing can promote lipid oxidation by disrupting cell membranes and releasing pro-oxidants. They concluded that increasing ohmic power intensity tended to increase the TBARS levels. Also, higher ohmic power intensity at 50 V increased the lipid oxidation in all patties. Piette et al., (2004), studied bologna samples where they applied ohmic cooking and reached a fixed target temperature (70 °C), with 3 different voltage densities (2.9, 3.5, and 4.7 V cm⁻¹). They observed that higher ohmic voltage values caused faster heating rates. Higher heating rates facilitate free radical production, greatly accelerating lipid oxidation.

During the consecutive application of ohmic and infrared cooking, at the same voltage level, increasing infrared temperature caused a significant (P < 0.05) increase in TBARS value (Table 2). Infrared is a form of electromagnetic energy that can cause heating on the surface of objects when absorbed (Huang and Sites, 2012). As shown in Table 2, higher TBARS values were obtained from the consecutive application of ohmic-infrared cooking compared to the ohmic cooking alone process. This could be due to the exposure of the meat’s surface to higher temperatures and to the cooperative effects of thermal damage to the membrane phospholipids and heat denaturation of proteins during exposure to ohmic- infrared temperatures (Zell et al., 2010a).

In this study, TBARS levels were found at 0.31 mg MDA/kg for raw beef and 0.59–1.74 mg MDA/kg for cooked products. In this respect, Broncano et al., (2009) emphasized that there are many studies available which demonstrate how different kinds of cooked meat have higher levels of TBARS in comparison to raw meat. Broncano et al., (2009) studied the influence of different cooking methods [grilled (GR), fried (FP), microwaved (MW) and roasted (RO)] on lipid oxidation, and found TBARS value of 0.26 mg MDA/kg for raw, 0.86 mg MDA/kg for grilled, 1.19 mg MDA/kg for fried, 1.03 mg MDA/kg for microwave, and 1.35 mg MDA/kg for the roasted Latissimus dorsi muscle of Iberian pigs. Soladoye et al., (2017) reported that the average TBARS value in raw bacon was about 0.19 mg MD/kg and increased by over three-fold and four-fold after microwave and frying pan cooking (0.69 and 0.93 mgMD/kg), respectively. The cooked beef showed an increased TBARS value compared to the raw meat, and the highest values were found when the beef was broiled at a high temperature, cooked by microwave and boiled in water in comparison to cooked in a domestic oven (Peiretti et al., 2012). Cheng, (2016) reported that pork contains more unsaturated fatty acids than beef. Therefore pork is more sensitive to lipid oxidation than beef. In our study, although TBARS results after ohmic cooking alone and ohmic-infrared cooking at 40 and 55 Volt are closer to Broncano et al., (2009)’s study, ohmic-infrared cooking at 70 Volt caused higher TBARS formation. The cooking process gains importance at this point. The ohmic-infrared cooking process caused higher TBARS formation in beef, especially at 70 Volt. Maillard et al., (1996) indicated that lipid oxidation could decrease the nutritional value by the formation of potentially toxic products during cooking and processing. Wu et al., (1991) reported that if the TBARS value is higher than 1 mg/kg, off-odors are usually formed and it is considered the beginning of organoleptic perceptibility of lipid oxidation. Meat products that have a TBARS value higher than 1 mg/kg could be considered “rancid”. In this case, using 40 and 55 Volt values for ohmic cooking alone and 40 V-325 F, 40V-375 F for ohmic-infrared cooking is much better for maintaining the TBARS value below this generally acceptable threshold level.

As can be seen in Tables 4–7, the PAH levels of cooked samples were found between 0.0233-2.9941 μg/kg. It can also be observed that despite the fact that 1-methylnapthalin, fluorene, anthracene and benzo(a)pyrene were detected at different levels in almost all samples treated with both methods, the other PAHs were not determined in ohmic cooking alone and detected in lower levels at consecutive application of ohmic and infrared cooking. In this sense, the lower center temperature reached and the direct energy application of ohmic cooking alone make it safer for the PAH formation compared to the consecutive application of the ohmic and infrared cooking methods.

EFSA (2008) assessed that the sum content of four PAH compounds benzo(a)pyrene, chrysene, benz(a)anthracene and benz(b)fluoranthene (PAH4) is the most suitable indicator of PAH in food. The PAH4 levels of the cooked meat samples in this study were found to be between 1.2514- 3.7844 μg/kg, as shown in Tables 5 and 7. Kendirci et al., (2014) determined PAH4 levels between 0.62- 6.21 μg/kg in meatball samples treated with ohmic cooking after infrared heating using different fluxes, application distances and application durations. Jira (2010) studied the PAH in representative samples of smoked meat products (raw sausages, raw ham, cooked ham, frankfurter-type sausage, liver sausages) in Germany. PAH4 levels were found for frankfurter-type sausages with a mean value of 0.6 μg/kg, which was the highest, in the range of 0.3 μg/kg for raw ham and liver sausages, a 0.2 μg/kg mean value for raw sausages and the lowest mean value of 0.1 μg/kg for cooked ham. Öz and Yüzer (2016) studied the effect of type of barbecue and cooking levels on the formation of PAHs in beef steak. They found PAH4 levels which ranged between not detected and 1.30 ng/g. The PAH4 results in our study were higher than these values but lower than the European Commision (EC) PAH limits. According to EC directives, the maximum limit of 4 PAH which is permitted in smoked meat and smoked meat products is 12 μg/kg (EC, 2011).

The total PAH contents varied between 2.5685 and 13.6536 μg/kg in this study. This is relatively low in comparison with the study of Duedahl-Olesen et al., (2006). They examined smoked meat samples produced in Denmark and found that the sum of PAH ranged from 24 μg/kg for salami to 64 μg/kg for bacon. Also, Elhassaneen (2004) studied the broiling of beef burgers by charcoal and detected 11 PAHs in the range of 0.31-14.95 mg/kg. Panalaks (1976) indicated that barbecued meat had the greatest PAHs level at 164 mg/kg total PAHs and 30 mg/kg of benzo(a)pyrene.

BaP is the most common PAH that is present in an extensive variety of foods. Since BaP is one of the most powerful carcinogenic PAHs, it is very important to identify and study its level. Moreover, BaP is a very good indicator of other PAHs in food items; the correlation coefficient was reported as 0.87 between the total PAHs and BaP level and as 0.98 between the carcinogenic PAHs and BaP level (Kazerouni et al., 2001; Reinik et al., 2007). According to the European Commission regulation, the maximum acceptable level of BaP in smoked meat and smoked meat products was reported as 5 mg/kg until 31.08.2014; but from 01.09.2014 the maximum acceptable level is 2 mg/kg (EC, 2011). As can be observed from Tables 5 and 7, ohmic cooking alone and the consecutive application of ohmic and infrared cooking caused lower levels of BaP (1.25–1.43 μg/kg) formation than the EC limits stated above.

In general, the PAH levels of cooked beef reached their maximum values at high ohmic voltage level and high infrared temperature values in this study. However, the detection levels in all treatments were below the EC legal limits. For this reason, the ohmic cooking procedure followed by infrared cooking can be considered as safe in the context of PAH level.