Food Chemistry 160 (2014) 90–97

Contents lists available at ScienceDirect

Food Chemistry

journal homepage: www.elsevier .com/locate / foodchem

Analytical Methods

Optimization of microwave-assisted extraction of cottonseed oiland evaluation of its oxidative stability and physicochemical properties

http://dx.doi.org/10.1016/j.foodchem.2014.03.0640308-8146/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel./fax: +98 171 4426 432.E-mail addresses: smjafari@gau.ac.ir, jafarism@hotmail.com (S.M. Jafari),

m.taghvaii@gmail.com (M. Taghvaei).

Mostafa Taghvaei a, Seid Mahdi Jafari a,⇑, Elham Assadpoor a, Shahram Nowrouzieh b, Omran Alishah b

a Department of Food Materials and Process Design Engineering, Faculty of Food Science and Technology, University of Agricultural Sciences and Natural Resources, Gorgan, Iranb Cotton Research Institute, Gorgan, Iran

a r t i c l e i n f o a b s t r a c t

Article history:Received 14 December 2012Received in revised form 5 March 2014Accepted 12 March 2014Available online 21 March 2014

Keywords:Microwave-assisted extractionCottonseed oilThermal stabilityResponse surface methodology

Microwave assisted extraction (MAE) is a novel method, which can reduce the extraction time andsolvent consumption. This study aimed to evaluate the influence of MAE on oxidative stability andphysicochemical properties of cottonseed oil. We found that the optimum extraction conditions were:irradiation time 3.57 min; cottonseed moisture content 14% and cottonseed to solvent ratio 1:4, whichresulted in an extraction efficiency of 32.6%, 46 ppm total phenolic content, 0.7% free fatty acids, peroxidevalue of 0.2 and 11.5 h of Rancimat oxidative stability at 110 �C. GC analysis for MAE cottonseed oildetermined palmitic acid (23.6%), stearic acid (2.3%), oleic acid (15.6%) and linoleic acid (55.1%), whichwere not significant different (P > 0.05) than conventionally-extracted (control) cottonseed oil. MAE oilsamples from whole cottonseed (without dehulling) had the greatest long-term stability, more than oilsamples containing BHT.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Microwave assisted extraction (MAE) is a modern process thathas been studied extensively (Amarni & Kadi, 2010; Camel,2000). The efficiency of MAE process depends on the time and tem-perature of extraction, sample ratio and nature of both the solventand the solid matrix (Terigar et al., 2010). It has been well docu-mented that microwaves destroy biological cell structure in planttissues such as oilseeds resulting in better extraction (Azadmard,Habibi, Hesari, Nemati, & Fathi, 2010; Chemat, Amar, Lagha, &Esveld, 2005; Jun & Chun, 1998). This is because the heat generatedby movement of the polar molecules, which denatures cellular pro-teins. Thus, the moisture of oilseeds before extraction has animportant role in extraction efficiency.

Azadmard et al. (2010) investigated the effect of microwavepre-treatment of rapeseeds on cold press extraction efficiency withthe aim of enhancing oil extraction yield. They concluded thatmicrowave pre-treatment could increase oil extraction by 10%. Inanother study, Amarni and Kadi (2010) obtained oil from olive cakeusing microwave assisted extraction and hexane as the solvent.They concluded that MAE gave better yields in less time and usedless solvent.

In a recent study by Taghvaei, Jafari, Nowrouzieh, and Alishah(2013) has been revelled that microwave can destroy the structureof oil cells during process and facilitate the oil extraction withoutany heat treatment before extraction. Hence, applying MAE, thereis no need to cook the oilseeds before extraction, which could havea potential benefit for the industry.

Oxidation reactions are the main reason of deterioration inedible oils and fats during storage or heat treatments such asfrying and cooking. Autoxidation, which is the most commonoxidation phenomena in edible oils, happens through a reactionbetween oxygen and unsaturated fatty acids via an auto-catalyticprocess consisting of a free radical chain mechanism. This chainincludes initiation, propagation, and termination stages, whichmay be cyclical once started (Shahidi, 2005). Antioxidants pre-vent autoxidation of oils and fats by giving their hydrogen to freeradicals formed during initial stages of autoxidation. There aretwo main groups of antioxidants, natural and synthetic. Thereis still doubt about safety and the approval, usage level, typeand application of synthetic antioxidants and they are regulatedin most countries (Akoh & Min, 2008; Gunstone, 2011; Shahidi,2005). In recent years, a global trend has been made towardthe substitution of synthetic antioxidants with natural alterna-tives, which has been reviewed by Taghvaei and Jafari (2013).The main goal of these research studies was to reduce the appli-cation of synthetic compounds as antioxidants because of theirpotential negative health effects and as a result of consumerdemand.

M. Taghvaei et al. / Food Chemistry 160 (2014) 90–97 91

It has been proved that natural antioxidants like phenolic com-pounds are more susceptible to microwave (Proestos & Komaitis,2008), and much research has been undertaken to extract phenoliccompounds such as isoflavones (Terigar et al., 2010), polycyclicaromatic hydrocarbons (Camel, 2000), carvone and limonene(Chemat et al., 2005), gallic acid, vannilic acid, catechin, p-coumaricacid, ferulic acid and many others phenolic compounds (Proestos &Komaitis, 2008), pigments (Jun & Chun, 1998), tea polyphenols andcaffeine (Pan, Niu, & Liu, 2003), tocopherols and tocotrienol(Zigoneanu, Williams, Xu, & Sabliov, 2008), olive leaf polyphenols(Rafiee, Jafari, Alami, & Khomeiri, 2012; Taghvaei et al., 2014)and many others from various plant resources and most of themconcluded that MAE decreased the extraction time, solvent usageand increased the amount of extracted phenolic compounds. Also,Taghvaei et al. (2013) concluded that phenolic compounds fromcottonseed, such as gossypol, could be extracted more easily usingMAE than conventional extraction methods. Therefore, MAE of edi-ble oils can lead to extraction of more natural phenolic compounds,which can improve the oxidative stability of final products(Azadmard et al., 2010; Rafiee, Jafari, Alami, & Khomeiri, 2011).In a study by Azadmard et al. (2010), microwave pre-treatmentof rapeseed caused an increase in the oil phytosterols (by 15%)and tocopherols (by 55%) and, as a result, the stability of rapeseedoil (analysed by Rancimat) increased from one hour (which was foruntreated rapeseed) to eight hours after MAE.

The cottonseed oil, which contains around 30% saturated fattyacids (palmitic and stearic acid) is stable and is consumed mainlyas frying oil formulations (Shahidi, 2005), hence it is subjectedmore frequently to high temperatures and moisture (duringfrying). The presence of food moisture, atmospheric oxygen andhigh temperatures could cause various chemical changes and lossof antioxidants such as steam distillation of antioxidants, oxidationof phenolic compounds, reduction of their pro-oxidative activitydue to reaction with fried materials and polymerisation (Pokorny,Yanishlieva, & Gordon, 2000). Synthetic antioxidants such asbutylated hydroxyanisole (BHA) and butylated hydroxytoluene(BHT) are more susceptible to steam distillation (Warner, Brumley,Daniels, Joe, & Fazio, 1986). Losses of natural antioxidants arecomparatively smaller since their volatility is much lower thanthat of common synthetic antioxidants (Pokorny et al., 2000).

The objectives of this study were, firstly, to increase cottonseedoil extraction efficiency through MAE and optimize extractiontime, solvent usage and moisture content of seeds using responsesurface methodology (RSM) to achieve the highest extraction effi-ciency, thermal and oxidative stability, and most natural phenoliccompounds as well as the lowest free fatty acid and peroxidecontent. Secondly, to investigate the effect of natural phenoliccompounds from cottonseed hull (extraction of cottonseed oilwithout de-hulling) on long-term oxidative stability of cottonseedoil compared with soybean oil with and without 100 and 200 ppmsynthetic antioxidant.

2. Materials and methods

2.1. Sample preparations and reagents

The cottonseeds (pak variety) were obtained from CottonResearch Institute of Iran (2011). Pak is a variety of cotton, whichhas trace amounts of gossypol, and was selected to eliminate theeffect of gossypol on stability and total phenolic content of the finaloil. Gossypol is a phenolic compound with antioxidant and toxiceffects, and should be removed from cottonseed oil during therefining process (Taghvaei et al., 2013). Solvents and chemicalswere obtained from Merck (Darmstadt, Germany). Gallic acid,gossypol standards and Folin–Ciocalteu reagent were purchasedfrom Sigma–Aldrich Co. (St. Louis, MO, USA)

2.2. Microwave-assisted extraction

Cottonseeds were de-hulled and powdered with a laboratorymill (Sunny, SFP-820). For evaluating the influence of cottonseedmoisture content on MAE efficiency, three levels of moisturecontent (1%, 7% and 14% wet basis) were applied before extrac-tion. The initial moisture content of cottonseeds was 7% wetbasis. The cottonseed powders were placed as a uniform layeron an aluminum tray with 5 mm diameter and were placed ina vacuum oven (Memmert VO-200, Germany) under 10,000 Papressure at 50 �C for about 5 h until the moisture content wasreached to 1%. For increasing the powders moisture content to14%, sufficient amount of distilled water was sprayed on samplesaccording to the following equation. In order to have a uniformdiffusion of water, the moisturised powders were maintained at4 �C for 24 h.

Q ¼W iðMf �MiÞ=100�Mf ð1Þ

Q: amount of water which must be added (kg).Wi: initial weight of cottonseed powder (kg).Mi: initial moisture content (% wet basis).Mf: final moisture content (% wet basis).

A microwave oven (Samsung, model: CF3110N-5, Korea) wasmodified for oil extraction. The modified MAE system consistedof a volumetric flask (500 ml) coupled with a condenser at thetop and a magnetic stirrer beneath. The microwave outputwas 900 W with 2450 MHz frequency and its inner cavitydimensions were 400 mm � 300 mm � 250 mm. For each extrac-tion, 100 g cottonseed powder, 200, 300 or 400 ml of solvent(n-hexane) and a magnet were placed in microwave oven. After1, 3.5 and 6 min of irradiation and simultaneous magneticstirring, the extraction process was stopped. Then, thecottonseed meal removed from miscella by means of filtrationfollowed by centrifugation (Sigma, 3K30, Germany) at 80,500gfor 5 min. Also, the solvent removed under reduced pressureat 50 �C by a rotary evaporator (IKA RV 10 basic, Japan). Thecontrol (blank) oil was extracted by soaking of 100 g cottonseedpowder in 200 ml of n-hexane at 50 �C for 30 min without anymicrowave treatment for comparison. Refined soybean oilwithout any additives was purchased from a local oil refiningfactory (Alia Golestan Co.) to compare its oxidative stabilitywith our samples.

For evaluating the influence of phenolic compounds containedin cottonseed hull on oxidative stability of the final MAE oil, thewhole cottonseeds (without dehulling) were milled and mixedwith n-hexane by a ratio of 1:3 and the oil was extracted throughMAE after 3.5 min.

2.3. Oil physicochemical properties

In order to comparison of the effect of various microwavetreatments on physical properties of resulting oil, the follow-ing experiments were carried out: free fatty acid content(AOCS Ca 5a-40), melting point (AOCS Cc 3-25), smoke point(AOCS Cc 9a-48), refractive index (AOCS Cc 7-25) and themoisture content of cottonseed powder (AOCS Aa 3-38). Thetotal oil content of cottonseeds was determined by Soxhletapparatus for 16 h using n-hexane. The colour was evaluatedby transferring cottonseed oil samples into a micro-plate cell(4 mm diameter) and analysing using a Lovibond Tintometermodel cam-system 500 in the L, a, b, mode of CIE (L, a, b,indicate lightness, redness/greenness, and yellowness/blueness,respectively).

92 M. Taghvaei et al. / Food Chemistry 160 (2014) 90–97

2.4. Oil stability analysis

For evaluation of oxidative stability of cottonseed oil, the Schaaloven test was used. The stability of oil samples during 20 days ofoven storage was evaluated by Rancimat test, peroxide value,thiobarbituric acid (TBA) value and p-anisidine value usingspectrophotometer WPA (S2000, UK). The peroxide value wasperformed according to AOCS Cd 8b-90 method. The p-anisidinvalue was carried out by AOCS Cd 18-90 method, and the TBAvalue was performed according AOCS Cd 19-90 procedure. Theoxidative stability index was analysed by Rancimat (Metrohm,743, Switzerland) apparatus according to AOCS Cd 12b-92 methodat the temperature of 110 �C (AOCS, 2007). The oil samples weretransferred to a series of open glass bottles and placed in an ovenat 55 �C for 20 days with forced air circulation. Samples of soybeanoil without additives and containing 100 and 200 ppm BHT werealso selected as control samples. Periodically (every 4 days),samples were removed from oven, flushed with nitrogen, andstored at �20 �C until analysis.

2.5. Total phenolic content

The phenolic compounds of oils were extracted by methanol–water as described by Bail, Stuebiger, Krist, Unterweger, andBuchbauer (2008) and total phenolic content were performedaccording Folin–Ciocalteu method.

2.6. GC analysis of fatty acids

Preparation of methyl esters of fatty acids for GC analysis wasperformed by AOCS Ce 2-66 and GC analysis was carried out byAOCS Ce 1-62. A fused-silica capillary column with inner dimen-sion of 0.25 mm, outer dimension of 0.39 mm and film thicknessof 0.2 lm was connected to a Chrompack CP 3800 gas chromato-graph (Middleburg, The Netherlands) equipped with a flame ioni-sation detector. GC conditions were: injector temperature 250 �C,detector temperature 270 �C, column temperature 130 �C andnitrogen as the carrier gas with a flow rate of 25 ml/min.

2.7. Statistical analysis

The statistical analyses were carried out by Minitab 16(Mini-tab, Inc., State College, PA, USA). The extraction optimizationwas performed by response surface methodology using two-levelfactorial (full factorial) design, 3 factors of moisture, MAE timeand sample ratio, 16 cube points, 2 centre points in cube, 12 axialpoints and no centre point in axial. The optimum extraction condi-tions were specified based on maximum extraction efficiency,Rancimat stability and phenolic content and minimum free fattyacid content and peroxide value having the weight of 10, 7, 2, 1and 1, respectively. The long term oxidative stability statisticalanalyses (ANOVA) were carried out using full factorial design. Allexperiments were performed in triplicate and the mean valueswere reported.

3. Results and discussion

3.1. Extraction efficiency of MAE cottonseed oil

The initial oil content of Pak variety of cottonseed was 34.7% drybasis as it was evaluated by Soxhlet method after 16 h extraction.Through microwave irradiation, 32.6% (dry basis) of cottonseed oilwas extracted in 3.5 min under the optimized conditions.

We found that the extraction efficiency for MAE of cottonseedoil enhanced by increasing cottonseed moisture content (Fig. 1b).The effectiveness of microwave on rapid extraction of oil from

oilseeds is related to destruction of oil cell structures within theplant tissues through denaturation of cell proteins by heat gener-ated from movement of polar molecules including water (Chematet al., 2005; Jun & Chun, 1998; Azadmard et al., 2010). Thus, thehigher the moisture content, the better the extraction efficiency.

Another interesting result was that the extraction efficiencyalso increased with longer irradiation times (up to 3.5 min). Irradi-ation for longer than 3.5 min had an adverse effect on extractionefficiency (Fig. 1a). The best irradiation time was 3.5 min, basedon extraction efficiency. As illustrated in Fig. 1(a), more solvent in-creased extraction efficiency. By increasing the solvent content inextraction flask, more oil can be dissolved in the solvent beforeoilseeds and solvent reach to a balance but regarding the cost ofsolvent recovery in oil extraction industry, the usage amount ofsolvent is always a limitation (Shahidi, 2005).

3.2. Total phenolic content of cottonseed oil after MAE

The results of our experiments showed that total phenolic con-tent of cottonseed oils (as it shown in Fig. 1e and f) increased athigher irradiation times and moisture contents. It shows that phe-nolic compounds are more exposed to microwave and microwaveirradiation had a suitable effect on extraction of phenolic com-pounds. These results concurred with previously published results(Zigoneanu et al., 2008; Proestos & Komaitis, 2008; Rafiee et al.,2011). Enhancing the extraction solvent led to a slight decline intotal phenolic content. The reason could be related to repellingeffects of solvent. N-hexane is a non-polar solvent which is notinfluenced by microwave, therefore increasing amount of n-hexanein extraction flask could led to a poor penetration of microwaveinto plant cells and a lower efficiency. The whole cottonseedextract (extracted oil without removing the cottonseed hulls) had70 ppm of total phenolic content as gallic acid equivalent (GAE),which was the highest phenolic content among all the samples.This shows cottonseed hull is a good source of phenolics that canbe extracted with the oil using MAE.

3.3. Influence of MAE on extracted cottonseed oil stability

Our results showed that oil stability, which was analysed usingthe Rancimat method, is directly related to total phenolic content.Fig. 1(d) shows that the cottonseed oil stability increased at highermoisture contents. The reason is related to total phenolic contentsince the total phenolic content of oil also increased at highermoisture contents (Fig. 1f), which enhances oil stability as a result.As has been explained, a decline in the amount of n-hexane in theextraction flask, resulted in an increase in total phenolics of cotton-seed oil, which increased thermal stability (Fig. 1c). Oil stabilityincreased with longer higher MAE times and reached a maximumafter 3.5 min, but this trend was reversed subsequently (Fig. 1c).Greater oil stability could be related to the increasing in phenoliccompounds, but reduced oil stability after 3.5 min, as shown inFig. 1(g), could be because of more peroxides following longerirradiation.

In Fig. 2, the oxidative stability of MAE cottonseed oil was com-pared with blank cottonseed oil (extracted by soaking in n-hexaneat 50 �C for 30 min), refined soybean oil without any additives, andwhole cottonseed extract. The thermal stability of blank cotton-seed oil was greater than refined soybean oil with or without100 and 200 ppm BHT. The saturation of cottonseed comparedwith soybean oil could explain this phenomenon. Autoxidationhappens through a reaction between oxygen and unsaturated fattyacids. Thus, a degree of high saturation cottonseed oil makes itmore resistant to autoxidation and suitable for frying (Akoh &Min, 2008; Shahidi, 2005). Interestingly, the whole cottonseed oil

Fig. 1. The surface plot of parameters affecting extraction efficiency (a,b), oil stability (c,d), total phenolic content (e,f), and peroxide value (g) of MAE cottonseed oil.

M. Taghvaei et al. / Food Chemistry 160 (2014) 90–97 93

(without dehulling) had a high oxidative stability (12.91 h), whichprobably was due to its higher total phenolic content.

3.4. Peroxide value of MAE cottonseed oil

This index was examined immediately after MAE to evaluate theoxidation rate of the oil during processing. The peroxide value forcottonseed oil increased with higher moisture levels and reached amaximum at 7% (w/w) with a decreasing trend subsequently

(Fig. 1g). The peroxide value indicates progression of oxidation reac-tions and is influenced by total phenolic content of the oil. As shownin Fig. 1(f), total phenolics were not significantly changed (P > 0.05)up to 7% moisture. Thus, higher values of peroxide could be a result ofhigher extraction temperatures generated by the presence of morewater. Moisture greater than 7% resulted in a significant increase(P < 0.05) in total phenolic content, which decreased the oxidationrate and peroxide formation in cottonseed oil. The amounts of bothperoxide and free fatty acids were at the maximum after 6 min

Fig. 2. The influence of moisture content (MC), MAE time (T), and cottonseed to solvent ratio (SR) on thermal stability of cottonseed oil evaluated by Rancimat at 110 �C.

94 M. Taghvaei et al. / Food Chemistry 160 (2014) 90–97

MAE. This suggested MAE longer than 3.5 min was inappropriate interms of extraction quality.

3.5. Physical properties of extracted cottonseed oil through MAE

The free fatty acids (FFA) content was analysed as an index forextraction quality and handling of cottonseeds before extraction.

Table 1The effect of MAE through different moisture content (MC), irradiation time (T) and sample to solvent ratio (SR) on physical properties of cottonseed oil.

Treatment Fatty acid composition (%) Free fatty acids (%) Refractive index Colour

Palmitic acid Stearic acid Oleic acid Linoleic acid L⁄ a⁄ b⁄

MC: 7%, T: 3.5, SR: 1:3 23.6 2.6 15.8 54.9 0.544 1.4663 30.2 5.9 5.1MC: 7%, T: 1, SR: 1:3 24.2 2.1 15.6 55.3 0.321 1.4658 30.6 3.5 6.7MC: 1%, T: 3.5, SR: 1:3 23.7 2.3 15.7 55.1 0.642 1.4651 31.4 5.1 5.9MC: 14%, T: 6, SR: 1:4 23.8 2.6 15.5 55.1 0.479 1.4656 29.8 5.1 5.9MC: 14%, T: 1, SR: 1:2 24.1 2.5 15.9 54.8 0.439 1.4656 30.6 4.3 6.7MC: 14%, T: 1, SR: 1:4 23.5 2.4 15.6 55.3 0.421 1.4660 31.4 3.5 7.5MC: 14%, T: 6, SR: 1:2 24 2.5 15.8 54.9 0.923 1.4661 29.4 5.9 6.7MC: 1%, T: 6, SR: 1:4 23.9 2.3 15.4 55.4 0.478 1.4658 30.6 4.3 6.7MC: 7%, T: 6, SR: 1:3 23.7 2.1 15.7 55.3 0.359 1.4661 29.8 5.9 5.1MC: 1%, T: 1, SR: 1:4 23.5 2.2 15.6 55.4 0.461 1.4652 30.2 5.9 6.7MC: 7%, T: 3.5, SR: 1:2 23.7 2.1 15.7 55.6 0.458 1.4660 31.0 4.3 7.5MC: 14%, T: 3.5, SR: 1:3 24.1 2.6 15.3 54.9 0.681 1.4668 30.2 5.1 8.2MC: 1%, T: 6, SR: 1:2 23.5 2.3 15.4 55.5 0.894 1.4668 29.4 5.9 6.7MC: 1%, T: 1, SR: 1:2 23.4 2.4 15.6 55.6 0.641 1.4660 29.8 4.3 6.7MC: 7%, T: 3.5, SR: 1:4 23.9 2.3 15.7 55.3 0.894 1.4657 31.0 5.1 5.9Whole cottonseed oil 23.9 2.4 15.8 55 2.476 1.4649 29.4 7.5 5.1Blank 23.7 2.5 15.4 55.1 0.673 1.4656 29.8 5.9 6.7

Table 2The colour changes of MAE cottonseed oils after different exposure times (T) and with various moisture contents (MC) compared with control sample during 20 days storage at55 �C.

Day 1 Day 4 Day 8 Day 12 Day 16 Day 20

L a b L a b L a b L a b L a b L a b

Blank 29.8 5.9 6.7 29.4 5.1 5.9 29 5.9 5.1 29.8 5.9 4.3 29.8 5.1 4.3 29 5.9 3.5T: 3.5 min, MC: 14% 30.2 5.1 8.2 27.8 7.5 3.5 28.2 8.2 4.3 29 8.2 1.2 28.2 8.2 1.2 29 8.2 0.4T: 3.5 min, MC: 1% 31.4 5.1 5.9 29 6.7 5.9 29.4 6.7 5.9 30.2 6.7 5.1 29.8 8.2 4.3 29 8.2 3.5T: 1 min, MC: 14% 30.2 5.1 5.9 28.6 6.7 4.3 30.6 6.7 4.3 30.2 5.9 5.1 29.8 7.5 4.3 30.6 8.2 3.5Whole cottonseed oil 29.4 7.5 5.1 28.2 7.5 5.1 29.4 7.5 4.3 29 8.2 4.3 30.2 8.2 3.5 29.8 8.2 2BHT 100 ppm 34.1 5.9 �2 32.2 5.9 �0.4 32.9 5.9 �1.2 32.9 5.1 �0.4 33.3 5.1 0.4 34.1 5.1 2BHT 200 ppm 33.7 5.9 �2 31.4 5.9 �0.4 32.2 5.9 �0.4 32.9 6.7 �0.4 33.7 5.1 0.4 31.8 5.9 2Soybean oil 34.1 5.9 0.4 31 5.9 0.4 31.8 6.7 0.4 33.7 5.1 �0.4 33.1 5.9 �0.4 33.6 6.1 0.4

All the samples had less than 1% free fatty acids, which showhandling (storage, grinding, drying, etc.) of cottonseeds beforeextraction was appropriate. We observed that 3.5 min MAE, asample ratio of 1:4, and moisture content of 7% resulted in an oilwith the best quality in terms of FFA (Table 1).

Refractive index was used as an indicator of MAE influence ontriglyceride composition of cottonseed oil. As shown in Table 1,

M. Taghvaei et al. / Food Chemistry 160 (2014) 90–97 95

there was no significant difference (P > 0.05) between refractiveindex of MAE-treated and control samples. We can concludedifferent moisture contents, irradiation times, and sample ratiosand, generally speaking, MAE processing had no significant effecton triglyceride and fatty acid composition. Also, the triglyceridecomposition of whole cottonseed oil (without dehulling) wassimilar to its dehulled counterpart.

Fig. 3. The peroxide value (a), TBA value (b) and p-anisidine value (c) of M

The colour of edible oils is one of the most important physicalparameters for marketing purposes. Therefore, it was importantto show the influence of MAE on the colour of the final oil. Asshown in Table 2, there was no significant difference (P > 0.05) incolour between the various oil samples. The whole cottonseed oilwas redder than other samples, which is because some pigmentsare extracted from cottonseed hull.

AE cottonseed oil and control samples during 20 days storage at 55 �C.

96 M. Taghvaei et al. / Food Chemistry 160 (2014) 90–97

3.6. Long term oxidative stability of MAE oils

Four MAE oil samples were selected for long term stability anal-ysis along with blank sample (conventionally extracted oil) andthree soybean oil samples, one without any additives and two sam-ples containing 100 and 200 ppm BHT for comparison. During20 days storage in an oven at 55 �C, peroxide value, p-anisidine va-lue, TBA value and colour of oil samples were evaluated every4 days.

The peroxide value of all soybean oils was very low (less than3 meq/kg oil) until day 8, but increased suddenly after day 12.The peroxide values of soybean oil samples after 20 days storagewere in the order: soybean oil without any additives > BHT100 ppm > BHT 200 ppm (Fig. 3a). This is quite straightforwardsince the highest stability was observed in samples that containedthe greatest amounts of antioxidants. The cottonseed oil samplesshowed a gradual increasing trend in peroxide values (Fig. 3a).The oxidation rate of cottonseed samples was high during firstdays, but slowed after day 4 and was consistent until day 20,except the whole cottonseed oil, which was highly resistant tooxidation (3.8 meq/kg maximum, 20 days storage, 55 �C). Afterwhole cottonseed oil, the sample extracted by 3.5 min irradiationand 1% moisture content showed the best oxidative stability.

TBA value can give a measure of lipid oxidation. TBA resultsconcurred with peroxide values; whole cottonseed oil showedthe best oxidative stability and the MAE oil following 3.5 min irra-diation and 1% moisture was more stable than alternative samples(Fig 3b).

The results for p-anisidine value are shown in Fig. 3c. This valuerepresents secondary oxidation products (principally 2-alkenalsand 2, 4-alkadienals) generated during decomposition of hydroper-oxides. Initial p-anisidine value were around 6 for all soybean oils,but the sample without any additives quickly increasing comparedwith other samples and reached 14.8 after 20 days. For cottonseed

Fig. 4. GC analysis of fatty acid composition of MAE cottonseed oil after 1 min exposure time, 7% moisture content and sample to solvent ratio of 1:3. (1) Palmitic acid, (2)stearic acid, (3) oleic acid and (4) linoleic acid.

samples, the same trends for peroxide and TBA value were re-peated in p-anisidine value.

It can be concluded that the influence of MAE time on cotton-seed oil oxidative stability is greater than moisture content, andbased on oil stability, the best irradiation time is 3.5 min. Theoxidative stability of crude cottonseed oil was much higher thanrefined soybean oil samples. The highest oxidative stability wasobserved in cottonseed oil extracted without dehulling (wholecottonseed oil).

3.7. GC analysis of fatty acids composition of MAE oils

The fatty acid composition of cottonseed oil, as illustrated inFig. 4, was: palmitic acid (23.6%), stearic acid (2.3%), oleic acid(15.6%) and linoleic acid (55.1%). We found that there was nosignificant difference (P > 0.05) between blank sample and MAEcottonseed oils (Table 1). Also, the GC results revealed thatdifferent MAE times at various moisture contents and differentamounts of n-hexane had no significant effect (P > 0.05) on fattyacid composition of cottonseed oil. Our GC analysis resultsconcurred with the data provided by Azadmard et al. (2010) andLopez, Velasco, Dobarganes, Ramis-Ramos, and Luque de Castro(2003).

4. Conclusion

After optimization of extraction parameters to achieve the max-imum oil extraction efficiency, total phenolic content and oxidativestability, and the minimum peroxide value and free fatty acid con-tent, it could be concluded that the ideal MAE time was 3.5 min,the best moisture content for cottonseeds 14%, and the best cotton-seed to solvent ratio was 1:4. Together, these factors resulted in anextraction efficiency of 32.6%, 46 ppm of total phenolic GAE, 11.5 h

of

M. Taghvaei et al. / Food Chemistry 160 (2014) 90–97 97

oxidative stability, peroxide value of 0.2 meq/kg oil and 0.7% freefatty acids with composite desirability of 0.33. Our results suggestcottonseed oil total phenolic content had the greatest influence onoil stability; the highest oxidative stability was observed in cotton-seed oil without dehulling (whole cottonseed extract), which alsohad the greatest phenolics content. To conclude, MAE is a suitableprocess for rapid extraction of edible oils, which results in a prod-uct with favourable physical properties and oxidative stability.

References

Akoh, C. C., & Min, D. B. (2008). Food lipids chemistry, nutrition, and biotechnology.Boca Raton, London, New York: CRC Press, Taylor & Francis Group.

Amarni, F., & Kadi, H. (2010). Kinetics study of microwave-assisted solventextraction of oil from olive cake using hexane: Comparison with theconventional extraction. Innovative Food Science and Emerging Technologies, 11,322–327.

AOCS (2007). Official methods and recommended practices of the American OilChemist’s Society (5th ed.). Champaign.

Azadmard, D. S., Habibi, N. F., Hesari, J., Nemati, M., & Fathi, A. B. (2010). Effect ofpretreatment with microwaves on oxidative stability and nutraceuticalscontent of oil from rapeseed. Food Chemistry, 121, 1211–1215.

Bail, S., Stuebiger, G., Krist, S., Unterweger, H., & Buchbauer, G. (2008).Characterisation of various grape seed oils by volatile compounds,triacylglycerol composition, total phenols and antioxidant capacity. FoodChemistry, 108, 1122–1132.

Camel, V. (2000). Microwave-assisted solvent extraction of environmental samples.Trends in Analytical Chemistry, 19, 229–248.

Chemat, S., Amar, H. A., Lagha, A., & Esveld, D. C. (2005). Microwave-assistedextraction kinetics of terpenes from caraway seeds. Chemical Engineering andProcessing, 44, 1320–1326.

Gunstone, D. F. (2011). Vegetable oils in food technology: Composition, properties anduses (2nd ed.). John Wiley & Sons Ltd..

Jun, S. J., & Chun, J. K. (1998). Design of u-column microwave-assisted extractionsystem and its application to pigment extraction from food. Trans IChemE, 76,231–236.

Lopez, E. P., Velasco, J., Dobarganes, M. C., Ramis-Ramos, G., & Luque de Castro, M. D.(2003). Focused microwave-assisted Soxhlet extraction: An expeditiveapproach for the isolation of lipids from sausage products. Food Chemistry, 83,143–149.

Pan, X., Niu, G., & Liu, H. (2003). Microwave-assisted extraction of tea polyphenolsand tea caffeine from green tea leaves. Chemical Engineering and Processing, 42,129–133.

Pokorny, J., Yanishlieva, N., & Gordon, M. (2000). Antioxidants in food: Practicalapplications. New York: CRC Press.

Proestos, C., & Komaitis, M. (2008). Application of microwave-assisted extraction tothe fast extraction of plant phenolic compounds. LWT, 41, 652–659.

Rafiee, Z., Jafari, S. M., Alami, M., & Khomeiri, M. (2011). Microwave-assistedextraction of phenolic compounds from olive leaves: A comparison withmaceration. Journal of Animal & Plant Sciences, 21(4), 46–51.

Rafiee, Z., Jafari, S. M., Alami, M., & Khomeiri, M. (2012). Antioxidant effect ofmicrowave-assisted extracts of olive leaves on sunflower oil. Journal ofAgricultural Science and Technology, 14(4), 1497–1509.

Shahidi, F. (2005). Bailey’s industrial oil and fat products. New Jersey: John Wiley &Sons.

Taghvaei, M., & Jafari, S. M. (2013). Application and stability of natural antioxidantsin edible oils in order to substitute synthetic additives. Journal of Food Scienceand Technology. http://dx.doi.org/10.1007/s13197-0131080-1. in press.

Taghvaei, M., Jafari, S. M., Nowrouzieh, S., & Alishah, O. (2013). The influence ofcooking process on the microwave-assisted extraction of cottonseed oil. Journalof Food Science and Technology. http://dx.doi.org/10.1007/s13197-013-1125-5.in press.

Taghvaei, M., Jafari, S. M., Sadeghi-Mahoonak, A., Mehregan Nikoo, A. R., Rahmanian,N., Hajitabar, J., et al. (2014). The effect of natural antioxidants extracted fromplant and animal resources on the oxidative stability of soybean oil. LWT-FoodScience and Technology, 56, 124–130.

Terigar, B. G., Balasubramanian, S., Boldor, D., Xu, Z., Lima, M., & Sabliov, C. M.(2010). Continuous microwave-assisted isoflavone extraction system: Designand performance evaluation. Bioresource Technology, 101, 2466–2471.

Warner, C. R., Brumley, W. C., Daniels, D. H., Joe, J. F. L., & Fazio, T. (1986). Reactionsof antioxidants in foods. Food Chemistry Toxicology, 24, 10–15.

Zigoneanu, I. G., Williams, L., Xu, Z., & Sabliov, C. M. (2008). Determination ofantioxidant components in rice bran oil extracted by microwave-assistedmethod. Bioresource Technology, 99, 4910–4918.