PaPer Ital. J. Food Sci., vol. 27 - 2015 1 - Keywords: Microwave, air drying, antioxidant activity, total phenolic content, response surface methodology - OptimizatiOn Of micrOwave and air drying cOnditiOns Of quince (Cydonia oblonga, miller) using respOnse surface methOdOlOgy cem BaltaciOğlu*, nurhan uslu and mehmet musa Özcan Food Engineering Department, Faculty of Agriculture, Selcuk University, Konya, Turkey *Corresponding author: cembaltacioglu@gmail.com AbstrAct Effects of slice thickness of quince (Cydonia oblonga Miller) , microwave incident power and air drying temperature on antioxidant activity and total phenolic content of quince were investigat- ed during drying in microwave and air drying. Optimum conditions were found to be: i) for mi- crowave drying, 285 W and 4.14 mm thick (maximum antioxidant activity) and 285 W and 6.85 mm thick (maximum total phenolic content), and ii) for air drying, 75 ºc and 1.2 mm thick (both maximum antioxidant activity and total phenolic content). Drying conditions were optimized by using the response surface methodology. 13 experiments were carried out considering incident microwave powers from 285 to 795 W, air temperature from 46 to 74 ºc and slice thickness from 1.2 to 6.8 mm. 2 Ital. J. Food Sci., vol. 27 - 2015 1. IntrODuctIOn Edible fruits are sources of nutrients such as carbohydrates, vitamins, and minerals as well as non-nutrient compounds such as poly- phenols. nowadays, it is commonly admit- ted that there is a positive relationship be- tween a diet rich in vegetables and fruits and a reduced incidence of degenerative diseas- es such as cancer and cardiovascular events (GIbnEY et al., 2009). Health beneficial prop- erties of quince fruit (Cydonia oblonga Mill- er) have known from ancient times. Quince is the only species in the genus cydonia, which falls into Pomoideae subfamily of the rosace- ae along with apple and pear (PAcIFIcO et al., 2012). Quince is used extensively in Europe as a dwarfing rootstock for pear. total world production of fresh quince was 540.337 tons in 2010 and about 25% of this was produced in turkey (tsI, 2012). Drying process is one of the most important preserving operations that causes time and en- ergy consumption in the food industry. that is why new methods are aimed to decrease dry- ing time and energy consumption. new meth- ods combined different systems such as using microwave drying together with traditional dry- ing methods to reduce drying time (sEcMELEr, 2003). Over the past two decades, there has been an increasing attraction in microwave drying to reduce drying time and increase the removal of water from agricultural products. Microwave drying has several advantages such as short drying time, higher drying rate, better quality of the dried products and decrease energy con- sumption (sAnGA et al., 2000). response surface methodology (rsM) is one of the most commonly used optimization tech- nique in food science. this method is preferred because of the simplicity and high efficiency. rsM covers a group of techniques used to study the relationship between one or more measured responses and input variables. (ArtEGE et al., 1994). It has been successfully applied to op- timize food processing operations by many re- searchers (FrAnK, 2001; LEE et al., 2006; LucIA- nE et al., 2001; MIrHOssEInI et al., 2008; PIEt- rAsIK and LI-cHAn, 2002). 2. MAtErIAL AnD MEtHODs 2.1. Material Fresh quinces were obtained from a local market in Konya. these samples were trans- ferred to laboratory in cool bags and they stored in refrigeration temperature (4 °c) un- til the assay, Initial moisture content of fruits was detected as 80% in average. Prior to dry- ing, round shaped samples (2 cm in diameter) were obtained from fruit slices. thickness var- ied from 1.2 to 6.8 mm according to experi- mental design. 2.2. Drying Quinces were dried until the moisture content decreased to 40% of the initial moisture content, since burning was observed on the quince slices in microwave drying below this moisture content level. Microwave drying experiments were per- formed using a domestic microwave oven (ArcE- LIK ArMD 580, turkey). the dimensions of the microwave cavity were 345 mm x 340 mm x 225 mm. three power levels were selected as high (720 W), medium (540 W) and low (360 W) for drying experiments. One dish containing 1 slice of sample to make effective drying was placed on the centre of a turntable fitted inside the micro- wave oven. Quinces were placed uniformly as a thin layer onto the stainless steel trays (0.3 m x 0.2 m) and dried using air oven (nüve Fn055 Ankara, turkey, 55 L volume) at three different temperatures ( 50, 60 and 70 °c). Quince slices were placed uniformly as a thin layer onto the stainless steel trays (0.3 m x 0.2 m) and dried under direct sunlight in April in Konya, turkey (bALLADIn & HEADLEY, 1999). 2.3 Extraction the phenolic compounds were isolated from samples using a modified version of the meth- od described by sHAHIDI et al. (2001). One gram of sample was extracted 3 times using 10 ml of 70% (v/v) aqueous methanol (Merck, Germany) at room temperature by a homogenizer (Ika ul- tra turrax tube Disperser) for 1 min. the slur- ry was centrifuged at 4000 rpm for 15 min. su- pernatants were collected and combined in a ro- tary flask and then evaporated at 45 °c under vacuum by a rota vapor r-3000 rotary evap- orator (Laborato 4001, Heidolph). the extract- ed phenolics were dissolved in 25 ml metha- nol and then filtered using filter paper. Metha- nolic solutions of phenolic were stored -25 °c until analysis. 2.4 Determination of total phenolic content and antioxidant activity total phenolic compounds were prepared by using Folin-ciocalteu reagent (YOO et al., 2004). the free radical scavenging activity of the ex- tract was determined using 1, 1-diphenyl-2-pic- rylhydrazyl (DPPH) in order to determine anti- oxidant activity (LEE et al., 1998). total pheno- lic contents were calculated by a standard cal- ibration curve prepared using gallic acid. total phenolic content results were given as gallic acid equivalents in milligrams per 1000 g fruit. Anti- oxidant activity results were expressed as per- centage activity (%). Ital. J. Food Sci., vol. 27 - 2015 3 2.5 Experimental Design response surface Methodology (rsM) was used to optimize drying conditions, based on better preserve the antioxidant activity and total phenolic content of quince fruit. box– behnken design was selected for rsM analysis. box–behnken designs require only three lev- els, coded as −1, 0 and +1. the effects of the two independent processing parameters: slice thickness (X 1 , mm), incident microwave pow- er (X 2 , Watt) in microwave operation and ad- ditionally slice thickness (X 1 , mm), processing temperature level (X 2 , °c) in air drying on two dependent variables (antioxidant activity and total phenolic content) were investigated us- ing rsM. the total number of experiments in this study was 13 based on two levels and a two factor experimental design, with five rep- licates at the centre of the design for estima- tion of a pure error sum of squares. Minitab 16 (Minitab Inc. state college, PA) was used for the experimental design, data analysis and re- gression modeling. the independent variables were; X 1 (2–4 mm), X 2 (360-720 W) in micro- wave process and X 1 (2 – 4 mm), (X 2 50 - 70 °c) in air drying. Experimental data from the box– behnken design was fitted into a second-order polynomial model. Y= b o + b 1 X 1 + b 2 X 2 + b 1 2 X 1 2+ b 2 2 X 2 2 + b 1 b 2 X 1 X 2 (1) Where Y is the predicted response, X 1 and X 2 are independent variables, b 0 is a constant; b 1 , b 2 , b 1 2, b 2 2, b 1 b 2 are linear, quadratic and inter- action coefficients, respectively. 3. rEsuLts AnD DIscussIOn 3.1 Microwave drying Independent variables [thickness of slice (X 1 ) and microwave power level (X 2 )] observed and predicted values of antioxidant activity and total phenolic content for microwave drying were given in table 1. During drying in microwave oven an- tioxidant activity decreased in the studied varia- bles. the influences of interaction between thick- ness-thickness were determined as statistically important (p≤0.05) whereas the other parame- ters were found insignificant (p>0.05) (table 2a). statistical analysis indicated that the fitted mod- el (Eq.1) to experimental results displayed high performance to predict the antioxidant activi- ty of quince samples within the studied ranges of variables. regression coefficient (R2) and ad- justed regression coefficient (R2 adj ) were calculat- ed as 0.994 and 0.990, respectively (table 2a). 3D surface plots for the significant terms were shown in Figs. 1 and 2 (MInItAb 16). surface plot given in Fig. 1a indicated that antioxidant activity was not affected by thickness and pow- er level. thickness-thickness square parameter was the only parameter which affected on anti- oxidant activity (p≤0.05) (table 2a). Antioxidant activity of quince increased up to 4 mm of slice thickness and then decreased (Fig. 1a). the interaction between thickness-thickness, thickness-power, power-power, power and thick- ness alone showed significant effects (p≤0.05) (ta- ble 2b) on total phenolic content of quince dried in microwave oven. total phenolic content of quince dried in microwave oven was fitted in Eq.2. re- 4 Ital. J. Food Sci., vol. 27 - 2015 gression coefficient (R2) and adjusted regression coefficient (R2 adj ) were calculated as 0.880 and 0.795, respectively (table 2b). the linear coeffi- cient for thickness was over seventy times greater than that of power level. there was interaction of significance between thickness and power level. Y 1 = 80.13 + 5.58 X 1 – 0.66 X 1 2 (1) Y 2 = 2088.05 + 290.44 X 1 – 4.87 X 2 + -7.86 X 1 2 -0.39 X 1 X 2 (2) Antioxidant activity and total phenolic content values were found higher in microwave drying than that of air drying. Drying time decreased 60-120 times in microwave drying. bErtELI & MArsAIOLI (2005) have studied the influence of drying methods on moisture content of product. Very short heating–cooling cycles take place in microwave drying. the use of microwave for dry- ing has become common because it enhances the product quality and processing speed (DIAZ et al., 2003). Fresh fruits are well known for their an- tioxidant activity which is usually attributed to the polyphenol content (WAnG et al., 1996). to- tal phenolic content of 2020 mg GAE/kg sample and 78% inhibition for antioxidant activity were reported in fresh quince samples before the dry- ing experiment (HAMAZu et al., 2005). KArADE- nIZ et al. (2005) reported that antioxidant activ- ity of some quince varieties in turkey found be- tween 51-68%. Ital. J. Food Sci., vol. 27 - 2015 5 Fig. 1 - Effect of thickness and power on antioxidant activity during microwave drying (A) and on total phenolic content dur- ing microwave drying (b). Fig. 2 - Effects of thickness and temperature on antioxidant activity oven drying (A) and on total phenolic content oven dry- ing (b). 6 Ital. J. Food Sci., vol. 27 - 2015 3.2 Air drying Independent variables [thickness of slice (X 1 ) and temperature of air (X 2 )], observed and pre- dicted values of antioxidant activity and total phenolic content for air drying were given in table 3. Antioxidant activity of quince sam- ples decreased after air drying. thickness of samples and process temperature had con- siderable effect on antioxidant activity. Addi- tionally, the influences of interaction between thickness-thickness and temperature-tem- perature were statistically significant (p≤0.05) whereas the other parameters were insignif- icant (p>0.05) (table 4a). Effect of thickness and temperature on the antioxidant activity in air drying was given in Fig. 2a. Increase in thickness from 2 to 4 mm resulted in lower an- tioxidant activity at different drying tempera- tures. the fitted model to antioxidant activity results of quince samples after air drying was given in Eq. 3. Model parameters for Eq. 3 (R2 and R2 adj ) were calculated as 0.953 and 0.920, respectively (table 4a). the interaction between thickness-thickness, temperature-temperature, thickness-tempera- ture and temperature of process showed sig- nificant effects (p≤0.05) on total phenolic con- tent after air drying (table 4b). changes in to- tal phenolic content was given in Fig. 2b and increase in thickness and power level caused firstly decrease and then increase in total phe- Ital. J. Food Sci., vol. 27 - 2015 7 nolic content similar to the antioxidant activi- ty. total phenolic content of quince by air dry- ing was fitted in Eq. 4. Model parameters for Eq 4 (R2 and R2 adj ) were calculated as 0.977 and 0.960, respectively (table 4b). the high values of regression coefficient indicate a high degree of correlation between the experimental and fitted values. Y 3 = 449.60 – 20.10 X 1 – 12.54 X 2 + + 2.59 X 1 2+ 0.10 X 2 2 (3) Y 4 = 4925.38 – 6.43 X 1 – 161.10 X 2 + + 42.22 X 1 2+ 1.63 X 2 2- 6.27 X 1 X 2 (4) 3.3 Sun drying sun drying was chosen as traditional drying method and this process maintained about 24 h. the use of microwave oven seems to be more advantageous considering the time factor. In sun drying, total phenolic content and antioxidant activity were determined as 1544 mg GAE/ 1000 g and 74%, respectively. these values were close to the results observed after microwave drying and higher than that of air drying. In order to determine the optimal residual activity, response optimizer tool in MInItAb 16 (Minitab Inc. state college, PA) was used. the op- timum conditions were found as 4.14 mm thick- ness and 285 W power level for maximum anti- oxidant activity, 6.85 mm thickness and 285 W power level for maximum total phenolic content in microwave drying. When air drying was ana- lyzed, 1.2 mm thickness and 75 °c temperature were selected for maximum antioxidant activity and total phenolic content. 4. cOncLusIOns the effect of power level and slice thickness on the antioxidant activity and total pheno- lic content were investigated for quince sam- ples after drying. rsM was used to optimize the factors in order to obtain maximum level of antioxidant activity and total phenolic con- tent of quince samples. All independent vari- ables including thickness of slice (mm), pow- er level (W) and processing temperature (°c) had significant effects on the response val- ues. Furthermore square and interaction pa- rameters showed significant effects (p≤0.05). A desirable quadratic mathematical model was built by using box–behnken design. the an- tioxidant activity and total phenolic content decreased after drying. nevertheless optimum drying conditions were obtained for both mi- crowave and air drying. Additionally, the use of microwave provides time saving compared to the other drying methods. rEFErEncEs Arteaga G.E., Li-chan E., Vazquez-Arteaga M.c. and na- kai s. 1994. systematic experimental designs for produ- ct formula optimization. trends in Food science and te- chnology. 5: 243. balladin D.A. and Headley O. 1999. Evaluation of solar dried thyme (Thymus vulgaris L.) herbs. renewable En- ergy. 17: 523. berteli M.n. and Marsaioli A. Jr. 2005. Evaluation of short cut pasta air dehydration assisted by microwaves as com- pared to the conventional drying process. Journal of Food Engineering. 68:175. Dı´az G.r., Martìnez-Monzò J., Fito P. and chiralt A. 2003. Modelling of dehydration–rehydration of orange slices in combined microwave/air drying. Innovative Food science & Emerging technologies. 4: 203. Frank r. 2001. blending response surface methodology and principal components analysis to match a target produ- ct. Food Quality and Preference. 12: 457. Gibney M.J., Lanham-new s.A., cassidy A. and Vorster H.H. 2009. Introduction to Human nutrition. 2nd ed. Wiley-blackwell. Hamazu Y., Yasui H., Inno t., Kume c. and Omanyuda M. 2005. Activity of chinese Quince (Pseudocydonia sinen- sis schneid.), Quince (Cydonia oblonga Mill.), and App- le (Malus domestica Mill.) Fruits. Journal of Agricultural and Food chemsitry. 53: 928. Karadeniz F., burdurlu H.s., Koca n. and soyer Y. 2005. An- tioxidant Activity of selected Fruits and Vegetables Grown in turkey. turk. J. Agric. For. 29: 297. Lee W.c., Yusof s., sheikh Abdul Hamid n. and baharin b.s. 2006. Optimizing conditions of enzymatic clarifica- tion of banana juice using response surface methodology (rsM). Journal of Food Engineering. 73: 55. Lee s.K.. Mbwambo Z.H.. chung H.s.. Luyengi L.. Games E.J.c. and Mehta r.G. 1998. Evaluation of the antioxi- dant potential of natural products. combinational chem- istry and High throughput screening. 1: 35. Luciane c.M., Hilary c.M., Aparecida M. and De silva A.P. 2001. Optimization of the roasting of robusta coffee (Cof- fea Canephora, conillon) using acceptability tests and rsM. Food Quality and Preference. 12: 153. Mirhosseini H., tan c.P., sheikh Abdul Hamid n. and Yusof s. 2008. Optimization the contents of Arabic gum, xant- han and orange oil affecting on turbidity, cloudiness, average particle size, polydispersity index and density in orange beverage emulsion. Food Hydrocolloids. 22: 1212. Pcifico s., Gallicchio M., Fiorentino A., Fischer A., Meyer u. and stintzing c. 2012. Antioxidant properties and cyto- toxic effects on human cancer cell lines of aqueous fer- mented and lipophilic quince (Cydonia oblonga Mill.) preparations. Food and chemical toxicology. 50: 4130. Pietrasik Z. and Li-chan E.c.Y. 2002. response surface met- hodology study on the effects of salt, microbial transglu- taminase and heating temperature on pork batter gel pro- perties. Food research International. 35: 387. sanga E., Mujumdar A.s. and raghavan G.s.V. 2000. Prin- ciples and application of microwave drying. In: Mujum- dar Ed. Drying technology in agriculture and food scien- ces. science Publication, Enfield, usA. secmeler O. 2003. comparison of microwave drying and microwave mixed-bed drying of red peppers. M.sc. the- sis, Department of Food Engineering, Middle East tech- nical university, Ankara, turkey. turkish standardization Instutue bulletin. 2012. num- ber :10950. Yoo K.M., Lee K.W., Park J.b., Lee H.J. and Hwang I.K. 2004. Variation in major antioxidants and total antioxidant ac- tivity of Yuzu (Citrus junos sieb ex tanaka) during matu- ration and between cultivar. Journal of Agricultural and Food chemistry. 52: 5907. Wang H., cao G.H. and Prior r.L. 1996. total antioxidant ca- pacity of fruits. Journal of Agricultural and Food chem- istry. 44: 701. Paper Received December 11, 2013 Accepted May 8, 2014