INTRODUCTIONFortified dairy products appeal to a wide varietyof consumers and help them increase their intake ofbioactive components. Kashk is a local name for atraditional low-fat dried yogurt in Iran. It is obtainedfrom boiled and concentrated yogurt and is available ina semi-liquid or dried form. Liquid kashk contains 20–25% nonfat solids, 1% fat, 3% salt, and at least 13%protein [1].Coffee is one of the most consumed and commercializedfood products in the world and one of themost traded commodities, second only to petroleum [1].It has various biological and pharmacological properties,such as antioxidant, anti-inflammatory, and antimicrobialproperties, as well as other health benefits [2].Natural therapy is a modern approach to preventingcommon diseases. Therefore, coffee bean extract orpowder can be used in cosmetic products and numerousfunctional foods [1, 3–5]. However, natural antioxidantsare usually heat sensitive and susceptible to oxidation,which limits their application in food industry [6].Microencapsulation is a promising technique thatprotects bioactive materials and controls the releaseof entrapped ingredients [1]. Over the last few years,encapsulation has attracted much attention in food,pharmaceutical, and cosmetic industries due to its wideapplication in the design of functional products [1].Encapsulation techniques are often based on dryingprocesses, such as spray-drying or freeze-drying, dueto the liquid nature of extracts containing bioactivecompounds [7]. Several studies have investigated theprotective role of these techniques against adverseconditions to which extracts can be exposed. They foundthat encapsulation promoted better volatile retentionResearch Article DOI: http://doi.org/10.21603/2308-4057-2020-1-40-51Open Access Available online at http://jfrm.ru/en/Effects of encapsulated green coffee extractand canola oil on liquid kashk qualityElnaz Rahpeyma1 , Seyed Saeed Sekhavatizadeh2,*1 Department of Food Science, Sarvestan branch, Islamic Azad University, Sarvestan, Iran2 Fars Agricultural and Natural Resources Research and Education Center, AREEO, Shiraz, Iran* e-mail: s.sekhavati@areeo.ac.irReceived November 01, 2019; Accepted in revised form January 09, 2019; Published February 25, 2020Abstract: In this study, we used a water-in-oil (W/O) emulsion encapsulation technique to enhance green coffee extractin the novel kashk product and protect it against hot filling. Green coffee extracts (GCE) in free (1%, 0.5%, and 0.25%) andencapsulated form (EGCE) (5%, 2.5%, and 1.25%) were added to kashk during hot filling, and their physicochemical and sensoryproperties were investigated. The EGCE kashk had higher oxidative stability (0.43 h) than the control due to the extract’s highphenolic content and DPPH radical scavenging activity (74%). Although a high concentration of GCE caused a higher pH (4.02),the latter declined in all the samples during the storage period. Further, the size of droplets in the emulsion varied from 3.20 to8.51 μm, confirming the well-encapsulated GCE by Fourier transform infrared. In addition, palmitic acid and oleic acid weredetected in GCE by gas chromatography as the main saturated and unsaturated fatty acids, respectively. All the treatments hadsimilar rheological properties and the highest flow index was observed in the samples with EGCE 5% on day 60. The sensoryevaluation showed that the assessors preferred the kashk formulated with 1% GCE. Finally, GCE encapsulation protectedthe color of the samples, and the b* value remained unchanged, whereas the lightness (L*) increased. We suggest that a W/Oemulsion is a successful technique for GCE encapsulation in kashk and can offer the latter to consumers as an alternative type offlavored dairy product with a better shelf life and health benefits.Keywords: Antioxidant activity, encapsulation, green coffee extract, kashk, rheological propertiesPlease cite this article in press as: Rahpeyma E, Sekhavatizadeh SS. Effects of encapsulated green coffee extract and canola oilon liquid kashk quality. Foods and Raw Materials. 2020;8(1):40–51. DOI: http://doi.org/10.21603/2308-4057-2020-1-40-51.Copyright © 2020, Rahpeyma et al. This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 InternationalLicense (http://creativecommons.org/licenses/by/4.0/), allowing third parties to copy and redistribute the material in any medium or format and to remix,transform, and build upon the material for any purpose, even commercially, provided the original work is properly cited and states its license.Foods and Raw Materials, 2020, vol. 8, no. 1E-ISSN 2310-9599ISSN 2308-405741Rahpeyma E. et al. Foods and Raw Materials, 2020, vol. 8, no. 1, pp. 40–51and increased the shelf life of bioactive componentsand extracts [8–11]. There are numerous studies aboutencapsulation of coffee extract and its antioxidantcompounds [12–14].The incorporation of green coffee extract (GCE)in kashk has not been reported so far. Since kashk isprepared using hot filling, and heating causes a loss ofantioxidants, GCE was encapsulated using a waterin-oil (W/O) emulsion technique. Thus, our aimwas to investigate the physicochemical and sensorycharacteristics of kashk incorporated with free andmicroencapsulated GCE during the storage period.STUDY OBJECTS AND METHODSMaterials. Green coffee (Robusta coffee) wassupplied from the local market of Shiraz (Fars, Iran).Starter culture (-CH1-DVS-50U) was purchased fromChristian Hansen (Denmark), and all chemical reagentswere obtained from Merck Co. (Germany).The research was conducted at the Fars Agriculturaland Natural Resources Research and Education Center.Green coffee extraction. According to the slightlymodified method of Upadhyay and Ramalakshmi, 10 gof ground coffee was added to 100 mL of distilled waterand held in a hot water bath for 30 min [2]. Then theslurry was cooled at room temperature and filtered toobtain a clear extract for analysis.Microencapsulation. GCE was microencapsulatedusing a W/O emulsion technique based on the Tranet al. method with slight modifications [15]. Glycerolmonostearate (GMS) with HLB 3.8 (1.5 wt%) was addedto canola oil and shaken at 4000 rpm at 70°C. Then,the aqueous solution containing the GCE was heatedto 40°C. The W/O emulsion (10:90) was prepared byblending the GCE-containing aqueous phase and theGMS-containing canola oil phase at 27000 rpm and70°C for 2 min. Then, the suspension was cooled whilestirring with a magnetic at 1000 rpm for 2 h and leftfor 30 min for microcapsules to precipitate. Finally,the suspension was centrifuged at 350 g and 4°C for10 min. The precipitate was washed twice with salineand filtered. The obtained microcapsules were stored ina refrigerator until usage.GCE kashk production. Liquid kashk was preparedaccording to the method described in [1]. Then, freeand encapsulated green coffee extracts in differentamounts (0.25‒5%) were added to kashk. The sampleswere named GCE 1%, GCE 0.5%, GCE 0.25%, EGCE5%, EGCE 2.5%, and EGCE 1.25%. The sample withoutGCE was used as a control.Total phenolic content. The amount of totalphenol in different concentrations of the extractswas determined according to Folin-Ciocalteuas described by Ballesteros et al., with somemodifications [12]. Briefly, the samples (0.1 mL)were introduced into test tubes containing 0.75 mLof Folin–Ciocalteu’s reagent and 0.75 mLof 2% sodium carbonate. The tube was mixed andkept for 1 h in the darkness at room temperature. Theabsorbance was measured at 765 nm using a UNICO2100 UV–vis spectrophotometer. Phenolic compoundswere measured in triplicate, and the results wereaveraged. A calibration curve of gallic acid (rangingfrom 25 to 100 mg mL–1) was prepared in methanol.The results, which were determined by the regressionequation of the calibration curve (y = 0.000245x –0.0377; correlation coefficient r = 0.998), were expressedas gallic acid (GA) mg equivalents g–1 sample.Antioxidant activity. The free radical scavengingactivity of GCE was measured by using the1,1-diphemyl-2-picryl-hydrazyl (DPPH) following themethod of Ribeiro et al. with a slight modification [16].A DPPH solution was added to the extract and mixed.Then, the mixture was kept at room temperature in thedarkness for 1 h. The absorbance of resulting solutionswas measured at 515 nm. The blank sample wasprepared in the same manner except that methanol wasused instead of the DPPH solution. A standard curve wasprepared using TBHQ (tertiary butylhydroquinone) atdifferent concentrations. The percentage of scavengingactivity was calculated as below:% of scavenging = [(A0–A1) (A0)–1] × 100 (1)where A0 is the absorbance of the control and A1 is theabsorbance of the sample turbidity factor. Finally, IC50(an absorbance value of 0.5 in the reducing power assay)was calculated.Oxidative stability. The 892 Professional Rancimat(Metrohm, Herisau, Switzerland) was used to determinethe oxidation stability of the samples. Three grams of thesample was heated at 110°C under a purified air flow rateof 20 L h–1.pH value. The pH of kashk enriched with GCEwas measured using a pH meter (Greisinger electronic,Germany).Particle size distribution. The mean particle sizeof the microcapsules was measured using a dynamiclight scattering technique at ambient temperature (NanoParticle Analyzer SZ-100, Horiba, Germany).Fourier transform infrared spectroscopy (FTIR).The FTIR analysis of GCE and EGCE was recorded bya Perkin-Elmer Spectrum RXI spectrometer (USA) inthe transmit mode in the range of 400–4000 cm–1 in KBrpellets at a resolution of 4 cm–1. A DLaTGS (DeuteratedTriglycine Sulphate Doped with L-Alanine) detector wasused to perform the measurements at room temperature(25 ± 0.5°C) at 24 scan/min to find possible functionalgroups.Fatty acid composition. The fatty acid compositionwas determined according to the Golmakani et al.method [17].Rheological measurements. Viscosity wasmeasured using a Brookfield rotational viscometer(Model LVDV I+, Version 3.0, Stoughton, MN, USA)42Rahpeyma E. et al. Foods and Raw Materials, 2020, vol. 8, no. 1, pp. 40–51with a spindle C30 and a heating circulator. The kashksamples were mixed for 5 min at room temperature at60 rpm. Flow behavior was described using the Powerlaw, Bingham, and Casson models according toequations:τ = kγn (Power law model) (2)σ–σ0 = ηγ (Bingham model) (3)σ0.5 = k0c + kc(γ)0.5 (Casson model) (4)where τ is the shear stress, Pa; γ is the shear rate, s–1; k isthe consistency coefficients, Pa·sn; η is Bingham plasticviscosity; kc is Casson plastic viscosity; n is the flowindex, and σ0 and k0c are yield stress of Bingham andCasson models, respectively.Color analysis. Changes in kashk color weremeasured using a Choroma CR-400 meter (Japan). L, aand b values were expressed as L* (black to white), a*(green to red), and b* (blue to yellow) [18].Sensory analysis. Sensory properties of the sampleswere determined by 30 panelists. Sensory analysis includedaroma, color, taste, and overall acceptability. It used afive-point hedonic scale, with 5 indicating “like extremely”and 0 “dislike extremely”, compared to the control sample.The analysis lasted 7 consecutive days [19].Statistical analysis. A one-way analysis of variance(ANOVA) was performed at a confidence level of 0.05(SPSS version 16.0). The means were compared usingthe Duncan’s multiple range at a significance level of0.05. All experiments were performed in triplicate.RESULTS AND DISCUSSIONTotal phenolic content. The total phenolic content ofGCE is presented in Table 1.Тhe content of phenols in a 1200 ppm concentrationof GCE was 39.08 mg GA g–1, confirming its antioxidantactivity due to the polyphenolic compound [12]. Theseresults were close to those found by Siva, Rajikin [20].They reported total phenolic contents of GCEs obtainedby isopropanol and methanol to be 30.65 mg GA g–1and 16.26 mg GA g–1, respectively. Similarly, Naiduet al. reported 32.19% and 31.71% for arabica androbusta isopropanol/water extracts, respectively [21].However, the total phenolic content of coffee dependson its variety [6]. Bidchol et al. and Ballesteros et al.also found that extracts of spent coffee grounds had19.99 ± 3.56 mg 100 mL–1 chlorogenic acid and 350.28 ±11.71 mg GAE 100 mL–1 [7, 12].The most common polyphenols in coffee arephenolic acids, mainly caffeic acid, a type of transcinnamicacid, and its derivative, chlorogenicacid [22]. Chlorogenic acid is able to directly interactwith reactive oxygen species (ROS), making it aneffective OH• scavenger [12]. However, the content ofchlorogenic acid in green coffee beans varies dependingon genes, species, climate, nutrient state of soil,processing techniques such as decaffeination, degreeof ripeness, and also roasting. Since phenolic acid isheat sensitive, green coffee beans have a higher contentof chlorogenic acid [8]. In addition, it was found thatcoffee extract exhibited an antioxidant activity similar tograpes and pomegranates [12].Antioxidant activity. According to Table 1, the IC50of GCE is 1.95 mg mL–1, which is higher than that ofTable 1 Total phenolic content and IC50 in green coffee extract(GCE) and TBHQSample Total phenol, mg GA g–1 IC50GCE 39.08 1.95TBHQ – 1.04Figure 1 Oxidative stability of: (a) control, (b) GCE 1%and (c) EGCE 5%Time, h(a)Time, h(b)Time, h(c)μS/cm μS/cm μS/cm43Rahpeyma E. et al. Foods and Raw Materials, 2020, vol. 8, no. 1, pp. 40–51TBHQ (1.04 mg mL–1). Based on the data, we calculatedthe DPPH radical scavenging activity of GCE (74%) [23].The high total phenolic content of GCE found in ourstudy confirmed its high radical scavenging activityand antioxidant potential. In addition, the technique weused and the coating material had a great impact on theretention of phenolic compounds and antioxidant activityof encapsulated samples [12].Similar results were revealed by Naidu et al. whoreported 92, 87, and 76% antioxidant activity forarabica and 88, 82, and 78% for robusta at 60:40, 70:30,and 80:20 isopropanol/water ratios, respectively [21].Jeszka-Skowron et al. also found that the antioxidantcapacities of arabica and robusta green coffee averaged56.3% [24]. Interacting with DPPH, antioxidantstransfer an electron or a hydrogen atom to DPPH,thus neutralizing its free radical character [25]. Thedegree of discoloration indicates the scavengingpotential of the antioxidant extract. This radicalscavenging ability is mostly related to the types andamounts of antioxidative components in the extractand their ability to donate a hydrogen group [7].The antioxidant constituents in green coffee arechlorogenic, ferulic, caffeic, and coumaric acids [26].Chlorogenic and caffeic acids are considered the mostrelevant markers in coffee samples [27]. However,other compounds such as caffeine, trigonelline, andphenylalanines (formed during coffee roasting) haveantioxidant properties as well.Oxidative stability. The oxidative stability curveand the induction time for different treatments of kashksamples are shown in Fig. 1.As we can see, the longer the induction time, thegreater the oxidative stability. The induction time of thecontrol, GCE, and EGCE were 0.35, 0.58, and 0.43 h,respectively. The kashk samples containing GCE (freeor encapsulated) showed a longer induction time than thecontrol due to GCE’s antioxidant activity. Noteworthily,food products supplemented with extracts rich inpolyphenols have an increased antioxidant potential[2, 4, 28]. However, encapsulation shortened inductiontime, which was in contrast to [2, 29, 30]. Many studiesthat used microencapsulation managed to avoid thedeterioration of unsaturated fatty acids by oxidation.Indeed, the wall materials surrounded droplets andprotected them from environmental conditions [31–33].As expected, canola oil is sensible to oxidation due to ahigh amount of unsaturated fatty acids [34].pH. According to Table 2, the initial pH of thecontrol was significantly higher than that of othersamples (P < 0.05). It decreased from 4.04 to 3.81 duringstorage. However, the GCE kashk showed the reversetrend for 30 days, followed by a pH decrease (P < 0.05).We also found that higher GCE concentrations led tohigher pH of the samples, even at the end of the storageperiod. According to Carvalho et al., the pH of GCE isapproximately 5.77–5.95 [31].The acidity of all kashk samples containingGCE microencapsules also increased throughout theTable 2 pH of kashk samples supplemented with GCE during storageStorage period, daysSample 1 15 30 45 60Control 4.04 ± 0.06aA 4.02 ± 0.04aB 3.98 ± 0.03aC 3.86 ± 0.07aD 3.81 ± 0.01aEGCE 1% 4.02 ± 0.01bA 4.04 ± 0.02bB 4.06 ± 0.05bC 3.99 ± 0.06bD 3.94 ± 0.06bDGCE 0.5% 3.98 ± 0.02cA 4.02 ± 0.03cB 4.04 ± 0.06cC 3.96 ± 0.05bD 3.91 ± 0.01bEGCE 0.25% 3.95 ± 0.03dA 3.98 ± 0.07dA 4.01 ± 0.03dA 3.94 ± 0.01bB 3.88 ± 0.03cCEGCE 5% 4.03 ± 0.06bA 4.05 ± 0.01bA 4.08 ± 0.01bA 4.01 ± 0.03bA 3.97 ± 0.01bBEGCE 2.5% 4.00 ± 0.02cA 4.02 ± 0.02cA 4.05 ± 0.01cC 3.98 ± 0.01bA 3.92 ± 0.05bBEGCE 1.25% 3.97 ± 0.08dA 4.00 ± 0.05dB 4.03 ± 0.06cC 3.95 ± 0.06bD 3.89 ± 0.07cESmall letters in each column and capital letters in each row show a significant difference between samples (P < 0.05)Results are reported as means ± SE, for three replicates of each sampleFigure 2 (a) Optical microscopy of EGCE (100×, 400×)and (b) Droplet size distribution in initial emulsion(a)Diameter, μm(b)Frequency, μm44Rahpeyma E. et al. Foods and Raw Materials, 2020, vol. 8, no. 1, pp. 40–51and 1.2% (w/v) had a positive effect on the fermentationand survival of probiotic bacteria in milk and yoghurt.According to Lee et al., the reduction of pH in all kashksamples containing GCE microencapsules was related tothe production of lactic acid during storage [35].Particle size distribution. The average particle sizeof the initial emulsion droplets is presented in Fig. 2.According to the curves, the droplets size was in therange of 3.20 to 8.51 μm.These results confirm a microscopic size ofencapsulated emulsion droplets. Particle size has a greatinfluence on such features as the surface oil and the finalcontent of the ultimate encapsulated powder [39]. Forinstance, particles larger than 30 μm may create a sandyFigure 3 Fourier transform infrared spectra of: (a) GCE 1% and (b) EGCE 5%whole storage period, which was in agreement withLee et al. [35]. This phenomenon was due to theproduction of lactic acid during storage. The activityof lactic acid bacteria, hydrolysis, and lipid oxidationresulted in lactic acid accumulation. Thus, the reductionof pH occured [36].On the other hand, increased acidity during storage,called post-acidification, is attributed to the activity ofkashk starter cultures at refrigerated temperature. Theyinclude Streptococcus thermophilus and Lactobacillusdelbruekii subsp. bulgaricus that produce small amountsof lactic acid by fermenting lactose [37]. In addition, thepositive effect of GCE on probiotic bacteria found inour study was confirmed by Marhamatizadeh et al. [38].They reported that adding coffee extract at 0.4%, 0.8%,(a)(b)CM-1CM-1%T %T1009080706050403020101009080706050403020104000.0 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 400.04000.0 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 400.03371.812927.882557.141613.701390.921269.92 1050.961120.081157.98991.36616.80764.15814.57853.28610.25673.95762.48745.80815.65852.92991.881051.871119.571157.811391.53 1270.181651.53 1606.111698.142931.723329.0245Rahpeyma E. et al. Foods and Raw Materials, 2020, vol. 8, no. 1, pp. 40–51feel in the mouth. The average diameter of vegetable oildroplets was 32 μm in the initial emulsion and 1–4 μm inD-limonene primary emulsion [40, 41]. Our results werein agreement with Karim et al. and Silva et al. [29, 31].FTIR analysis. The average values of typical FTIRspectra for GCE and EGCE are depicted in Fig. 3.There, we can see characteristic broad low-frequencyabsorption bands at 3371 and 3329 cm−1 respectively,in the hydroxyl region (4000–3000 cm−1). These bandsrepresented the stretching vibrations of O–H in theconstituent sugar residues and adsorbed water [36]. Wealso observed three sharp peaks in the range of 3000 to2800 cm–1 (2927, 2931, and 2852 cm–1) for both arabicaand robusta roasted coffee samples, but no identificationwas attempted. Nonetheless, FTIR analysis of caffeinein soft drinks revealed two sharp peaks at 2882 and2829 cm–1, which were attributed to the asymmetricstretching of C-H bonds of methyl (-CH3) group in thecaffeine molecule and the peak region.These findings were successfully used by Chematet al. to develop predictive models for quantitativeanalysis of caffeine [42]. In their study, the wave numberin the region between 1400–900 cm–1 was identifiedby vibrations of several types of bonds, includingC-H, C-O, C-N, and P-O. The stretching of the cis= C-H and cis -C=C- at 1651 cm–1 was an indicator ofunsaturated fatty acids in vegetable oils [1]. Alcohols,saturated aldehydes and α, β-unsaturated aldehydesare major secondary oxidation products. In the regionbetween 1730–1680 cm–1, all aldehydes exhibit C=Ostretching bands [2]. It is evident that carbonyl stretchingaround 1698 cm–1 may represent acetone [27]. Thedisappearance of peaks at 2857 cm–1 and the appearanceof two peaks at 1698 and 1651 cm−1 in the EGCEtreatment was due to the presence of canola oil andindicated well-encapsulated GCE.Fatty acid composition. As can be seen inTable 3, the samples containing encapsulated GCEhad a lower content of fatty acids than the others. Theresults indicated the protective role of encapsulationagainst the oxidation reaction. This finding was inagreement with the results of Fantoni et al. and Sun-Waterhouse et al. [19, 43]. In addition, EGCE treatmentprovided higher amounts of linoleic, oleic, and elaidicacids. The significant difference found in the contentsof oleic and linoleic acids was due to the presence ofcanola oil in the encapsulated samples. Indeed, oleicacid and linoleic acid are two main fatty acids foundin canola oil [44]. Dubois et al. claimed that oleic acidis the principal ingredient of various vegetable oils,including olive and rapeseed oils, and is a major dietarymonoenoic acid [45]. Among PUFAs (polyunsaturatedfatty acids), linoleic acid is the only one that reducesLDL-cholesterol. Consequently, encapsulation of canolaoil has health benefits, as well as protective properties.According to the fatty acid profile, the control hada slightly higher content of total saturated fatty acidsthan the GCE kashks, while the EGCE samples had thelowest. The EGCE had the largest content of PUFAs andMUFAs due to the presence of canola oil.Rheological characteristics. The rheological parametersof the three kashk samples (control, GCE 1%,Table 3 Fatty acid composition of control, GCE 1%and EGCE 5% kashks, %Fatty acid Control GCE 1% EGCE 5%Butyric acid 0.93 2.25 0.77Caproic acid 0.63 3.97 1.78Caprylic acid 1.51 4.26 1.65Capric acid 4.37 3.67 3.83Lauric acid 4.27 5.56 2.63Myrisric acid 15.23 12.31 5.89Myristoleic acid 1.34 1.24 0.54Pentadecanoic acid 0.72 0.37 0.20Palmitic acid 39.21 28.01 17.99Palmitoleic acid 1.02 1.10 0.97Margaric acid 0.68 0.50 0.35Heptadecenoic acid 0.67 0.78 0.15Stearic acid 3.24 2.26 2.02Oleic acid (cis) 21.95 14.67 43.93Elaidic acid (trans) 2.09 10.79 13.99Linoleic acid (cis) 0.06 0.51 2.59Behenic acid 0.06 0.42 0.05100Table 4 Effects of storage on some model parameters of control, GCE 1%, and EGCE 5% samplesPower Law Bingham CassonDay Sample n k, Pa·sn r2 σ0 η, Pa·sn r2 k0C kc, Pa·sn r21 control 0.07 ± 0.00a 15.04 ± 0.15a 97.33 ± 0.15a 18.41 ± 0.07a 0.017 ± 0.00a 94.27 ± 0.40a 4.25 ± 0.01a 0.04 ± 0.00a 95.83 ± 0.25aGCE 1% 0.07 ± 0.00a 14.94 ± 0.27a 97.10 ± 0.20a 18.36 ± 0.12a 0.017 ± 0.00a 93.80 ± 0.40a 4.24 ± 0.02a 0.04 ± 0.00a 95.50 ± 0.30aEGCE5%0.07 ± 0.01a 14.72 ± 0.97a 96.90 ± 0.72a 18.25 ± 0.44a 0.018 ± 0.00a 93.33 ± 1.55a 4.22 ± 0.07a 0.04 ± 0.01a 95.16 ± 1.14a60 control 0.06 ± 0.01a 15.29 ± 1.06a 97.30 ± 0.69a 18.51 ± 0.45a 0.016 ± 0.00a 94.23 ± 1.44a 4.26 ± 0.06a 0.04 ± 0.01a 95.83 ± 1.09aGCE 1% 0.07 ± 0.03a 15.30 ± 2.36a 97.23 ± 1.49a 18.47 ± 0.99a 0.016 ± 0.01a 93.96 ± 3.28a 4.25 ± 0.14a 0.04 ± 0.02a 95.60 ± 2.36aEGCE5%0.09 ± 0.01a 13.53 ± 0.71a 96.03 ± 0.60a 17.71 ± 0.34a 0.022 ± 0.00a 91.33 ± 1.35a 4.14 ± 0.05a 0.05 ± 0.01a 93.67 ± 1.01aResults are reported as means ± SE, for three replicates of each samplea P < 0.0546Rahpeyma E. et al. Foods and Raw Materials, 2020, vol. 8, no. 1, pp. 40–51Figure 4 Apparent viscosity of the kashk samples duringstorage: (a) control; (b) GCE 1%; and (c) EGCE 5%(b)(c), ,and EGCE 5%) on days 1 and 60 of the storage periodare presented in Table 4. We found no significantdifferences in the viscosity of the treatments. As shownin Fig. 4, the samples’ viscosity increased after 60 daysof storage. Noteworthily, post-acidification caused adecline in the negative electric charge of casein micellesby dissolving calcium and inorganic phosphate. Itattenuates colloidal stability and subsequently caseinbecomes insoluble near its isoelectric pH (about 4.6).This phenomenon, strengthening protein-proteincomplexes and protein-polyphenol interaction, enhancesserum released from the gel matrix and, at the sametime, increases viscosity [46]. To explain the consistencycoefficient, flow index, and yield stress of the samples,we used different rheological models, namely Power law,Bingham, and Casson models. According to R2 of thesamples (higher than 0.97), the data fitted the Power Lawmodel more than the others (Table 4). As can be seen,there were no significant differences in flow index valuesbetween all the samples at the beginning and at the endof storage (P > 0.05). The same trend was observed inthe other model parameters throughout storage.According to the Power law equation, the kashkscontaining GCE 1% and EGCE 5% had the highest andthe lowest consistency coefficient on the first day ofstorage, respectively. At the end of storage, however, thecontrol and the EGCE 5% kashk had the lowest (0.064)and the highest (0.090) flow index.In Bingham’s equation, the control had the lowest(0.016) and the EGCE 5% sample had the highest (0.018)consistency coefficient at the beginning of storage. Thesame trend was revealed on day 60. The highest yieldstress (18.51) was observed in the control and the lowest(17.71) in the EGCE 5% kashk on day 60. Further, nosignificant differences were seen among the samples.The Casson equation showed the same trend atthe end of storage. However, we recorded the lowestyield stress (4.14) in EGCE 5% and the highest (4.26)in the control on day 60. Values within this range werefound in kashk with and without the addition of gumtragacanth [1]. A higher consistency coefficient anda lower flow index can be considered appropriate toachieve a high viscosity and a clean mouthfeel [1].Overall, all the samples showed a plastic-shear thinningbehavior because of lower flow index (n) values (< 1) anda higher consistency coefficient.Color. Table 5 summarizes the color characteristicsof the kashk samples supplemented with GCE on days 1and 60.Increasing the concentration of GCE (microencapsulatedor free) led to higher b* and L* values. We foundthat the highest b* value was related to the brownishyellow color of coffee extract. Among all the samples,those incorporated with microencapsulated GCE hadhigher b* and L* values compared with the others(P < 0.05). The higher L* value might be dueto the reflection properties of lipid droplets ormicroencapsulated particles.In general, we noticed no differences in thea* value of the kashk samples incorporated with GCE(free or microencapsulated). The L* and a* values of allthe treatments increased throughout storage (60 days)(P < 0.05). However, the b* value did not changesignificantly in the microencapsulated GCE samples,while increasing considerably in the control (P < 0.05)due to the brownish yellow color of coffee extract. Thisresult was in agreement with the studies of Lee et al. andAlavi et al. who reported enhanced L* and a* values inyoghurt enriched with powder peanut sprout extractmicrocapsules during 16 days [35, 36]. However, theb* value remained almost unchanged.Sensory analysis. Sensory attributes of the sampleswere evaluated on the first and last days of storage(Fig. 5).We found significant differences between the color ofthe kashk samples at the beginning (P < 0.05). The GCE1% sample had a higher score, whereas the EGCE 0.25%Days 1 Days 60(a)Shear rate, s–1Apparent viscosity, Pa·sShear rate, s–1Apparent viscosity, Pa·sShear rate, s–1Apparent viscosity, Pa·s280240200160120804003.84 23.04 42.24 61.44 80.64 99.84 119.04 138.24 157.44 176.643603002401801206003.84 23.04 42.24 61.44 80.64 99.84 119.04 138.24 157.44 176.643.84 23.04 42.24 61.44 80.64 99.84 119.04 138.24 157.44 176.642802402001601208040047Rahpeyma E. et al. Foods and Raw Materials, 2020, vol. 8, no. 1, pp. 40–51Table 5 Color characteristics of kashk samples with GCE on day 1 and day 60Day 1 Day 60Sample L* a* b* L* a* b*Control 56.50 ± 0.50Bb –6.20 ± 0.21Ba 11.37 ± 0.70Bc 63.55 ± 0.09Ab –3.55 ± 0.19Ab 14.34 ± 0.95AaGCE 1% 55.06 ± 0.90Bc –6.18 ± 0.30Ba 12.30 ± 0.02Bb 69.68 ± 0.42Aa –2.22 ± 0.08Aa 13.62 ± 0.43AbGCE 0.5% 56.27 ± 0.85Bb –6.13 ± 0.54Ba 12.60 ± 0.32Bb 62.33 ± 0.14Ab –3.24 ± 0.11Ab 10.26 ± 0.15BcGCE 0.25% 54.41 ± 0.56Bb –6.08 ± 0.26Ba 11.44 ± 0.84Ac 62.37 ± 0.74Ab –4.85 ± 0.42Ac 11.65 ± 0.71AcEGCE 5% 57.44 ± 0.19Ba –6.71 ± 0.40Ba 13.57 ± 0.42Aa 62.24 ± 0.06Ab –2.24 ± 0.11Aa 13.22 ± 0.48AbEGCE 2.5% 56.72 ± 0.85Bb –6.51 ± 0.99Ba 12.37 ± 0.59Ab 60.34 ± 0.08Ac –3.12 ± 0.36Ab 11.97 ± 0.35AcEGCE 1.25% 55.42 ± 0.80Bc –6.95 ± 0.09Ba 11.61 ± 0.64Ac 60.55 ± 0.29Ac –4.34 ± 0.65Ac 11.25 ± 0.31AcDifferent small letters in each column and capital letters in each row show a significant difference between samples (P < 0.05). Results arereported as means ± SE, for three replicates of each sampleFigure 5 Sensory attributes of kashk samples supplemented with GCE on day 1 and day 60: (a) Color; (b) Odor; (c) Taste;(d) Consistency; (e) Total acceptabilityhad a lower score (P < 0.05). The color of all the samplesimproved throughout storage. The GCE 1% had a betterscore, which was attributed to the light greenish yellowcolor of the coffee extract. However, it seems that GCEencapsulation protected the color of the kashk samples.As expected, the product with GCE 1% (free form) hadbetter odor, taste, consistency, and overall acceptancythan the others.Among all the samples, the EGCE 1.25% kashkobtained a lower score (P < 0.05). As predicted, the odorof kashk containing a free form of GCE improved withthe concentration increased. This finding confirms apositive effect of GCE on kashk products. Nevertheless,a decline in odor was observed during storage due tothe loss of some volatile compounds. We found thatthe samples’ consistency improved with the extractDay 1 Day 60Sensory score4.94.74.54.34.13.93.73.5Day 1 Day 60Sensory score4.94.74.54.34.13.93.73.5(b)Day 1 Day 60Sensory score4.94.74.54.34.13.93.73.5Day 1 Day 60Sensory score4.94.74.54.34.13.93.73.5(d)(e)Day 1 Day 60Sensory score4.94.74.54.34.13.93.73.51324 5671 Control2 GCE 1%3 GCE 0.5%4 GCE 0.25%5 EGGE 5%6 EGGE 2.5%7 EGGE 1.25%(a)(c)48Rahpeyma E. et al. Foods and Raw Materials, 2020, vol. 8, no. 1, pp. 40–51concentration (free form) increased. Overall, the kashkfortified with GCE 1% was favored by the assessors,possibly due to its oily appearance.Nonetheless, the type of microencapsulationtechnique influences the protective effects of odor/flavor components. For instance, Rodrigues et al.used cashew gum and Arabic gum as a wall materialfor microencapsulation of coffee extract duringspray drying [47]. Both Arabic gum and cashew gumhad similar aroma protective effect and showed nodifferences between the control and experimentalsamples. Furthermore, there is some dependencebetween the product’s sensory properties and the typeof herbal extract. Ribeiro et al. evaluated the colorparameter of cottage cheese incorporated with free andencapsulated mushroom extract and revealed no changesin cottage cheese color [16]. However, the samples’ colorimproved after 7 days (P < 0.05). Also, Gurkan et al.analyzed the taste and flavor characteristics of yogurtenriched with basil (Ocimum basilicum L.) powder orextract during three weeks of storage at 4°C [48]. Theyidentified 49 volatile compounds which enhanced thesensory score of yogurts. The presence of some volatilecarboxylic acids gave yogurt an acceptable acidic taste.CONCLUSIONWe employed a W/O emulsion to encapsulate GCEin kashk at different concentrations. Among all thesamples, the kashk incorporated with a high amount ofGCE had a higher pH. However, pH reduced throughoutstorage due to the production of a small amount oflactic acid. Furthermore, GCE possessed antioxidantactivity, mainly due to its high phenolic content, and thesamples containing GCE and EGCE had high oxidativestability. In addition, we confirmed a protective effect ofencapsulation against the oxidation reaction. The EGCEsample had higher contents of only three free fattyacids (linoleic, oleic, and elaidic) due to the presence ofcanola oil in its composition. The control had the highestcontent of total saturated fatty acids.The FTIR spectra indicated that the GCE was wellencapsulated. The rheological behavior of the samples,a plastic-shear thinning behavior, fitted the PowerLaw model more than the others. Since particles largerthan 30 μm may create a sandy feel in the mouth andthe emulsion droplets in our study had a micrometersize lower than 30 μm, we could claim that the EGCEkashk had the best texture. Adding GCE (free andmicroencapsulated) to kashk affected its sensoryproperties. The kashk with 1% of GCE was preferredby the assessors. In addition, the coffee aroma was feltat the end of storage, which is a positive effect of usingGCE in dairy products. Finally, GCE encapsulationprotected the color of the samples and the b* valueremained unchanged, while lightness (L*) increased.Overall, using a W/O emulsion can be successfullyemployed as a technique for GCE encapsulation in kashkand the resulting product can be offered to consumers asan alternative type of flavored dairy product.ACKNOWLEDGEMENTThe authors thank the Department of Food Scienceat Sarvestan Islamic Azad University for their greatassistance during the research. We did not receive anyspecific grants from funding agencies in the public,commercial or not-for-profit sectors.