Q.Zhangetal./FoodChemistry179(2015)15–25Table1

Box–Behnkendesignmatrixandtheresponsevaluesfortheextractionrateofresveratrol.No.Extractiontemperature(°C)X112345678910111213141516171819202122232425262728294050405045454545405040504545454540504050454545454545454545Codex1à11à110000à11à110000à11à11000000000CaOtorawmaterialratio(g/g,m:m)X25:1005:1007:1007:1006:1006:1006:1006:1006:1006:1006:1006:1005:1007:1005:1007:1006:1006:1006:1006:1005:1007:1005:1007:1006:1006:1006:1006:1006:100Codex2à1à11100000000à11à110000à11à1100000Watertorawmaterialratio(ml/g,V:m)X39:19:19:19:17:111:17:111:19:19:19:19:17:17:111:111:17:17:111:111:19:19:19:19:19:19:19:19:19:1Codex30000à11à110000à1à111à1à111000000000Extractiontime(min)X45050505040406060404060605050505050505050404060605050505050Codex40000à1à111à1à11100000000à1à111000000.5770.7620.6070.7680.7320.6720.7620.7600.5670.7230.6230.7570.7500.7620.7470.6820.6220.8030.5650.7680.6680.6750.7680.7620.8820.8770.8780.8880.902Extractionrate(%)19whereYisthepredictedresponseextractionratio(%)ofresveratrol,a0isaconstant,ai,aiiandaijaretheregressioncoef?cientsforintercept,linear,quadraticandinteractionterms,respectively.XiandXjrepresenttheleveloftheindependentvariables(i–j).2.6.StatisticalanalysisThestatisticalanalysiswasperformedbyusingthereportedprocedure(Jiang,Wang,Liu,Gan,&Zeng,2011).AnalysisoftheexperimentaldesignanddatawascarriedoutusingDesignExpertsoftwareofversion7.0(Stat-EaseInc.,Minneapolis,USA).Analysisofvariance(ANOVA)wascarriedoutandthe?tnessofthepolyno-mialmodelequationwasexpressedasthecoef?cientofdetermi-nationR2.Thesigni?cancesoftheregressioncoef?cientsweretestedbytheF-test.One-wayANOVAwasperformedusingtheSPSS13.0forwindows(SPSS,Chicago,IL,USA).Multiplecompari-sonsofmeansweredonebytheleastsigni?cancedifference(LSD)test.P-valuesoflessthan0.05wereregardedassigni?cant.3.Resultsanddiscussion3.1.EffectofextractiontemperatureonextractionratioofresveratrolToinvestigatetheeffectofextractiontemperatureontheextractionratioofresveratrol,theextractionprocesswascarriedoutusingdifferentextractiontemperaturesof25,30,35,40,45,50,55,60or65°C,whileotherextractingparameterswere?ttedasfollowing:CaOtorawmaterialratioof5:100,watertorawmaterialratioof7:1,andextractiontimeof40min.AsshowninFig.2A,theresveratrolextractionratiowasrapidlyincreasedasthetemperatureincreased,butwhenthetemperaturewasincreasedmorethan45°C,theresveratrolextractionratiowasslowlydecreased.Theseresultscanbeexplainedasextractionatthelowertemperaturesisnotconducivetodissolutionandmass-transferofresveratrolbetweenthesolventandthesample.Inapreviousstudy,itwasfoundthatthereactionrateoftheextractiontemperatureisgenerallyincreasedby3–4timesasthetemperatureincreased10°C(Xiong,Zhang,etal.,2014).Andinacertaintemperaturerange,themovement-activeforceofmolecu-larmatterwasstrengthenedduetothetemperaturerise.Thetem-peratureincreasemaychangethephysicalpropertiesofthematerial,whilereducingitsabsorptionintensityandmakingresve-ratroleasiertobeextracted(Alexandre&Dubois,2000).Therefore,hightemperaturesenhancetheextractionef?ciency.However,res-veratrolwasaf?liatedwithpolyphenols.Excessivelyhightemper-aturewillresultintheoxygenolysisofresveratrol,changethemolecularstructureofresveratrolandadverselyaffecttheextrac-tionef?ciency(Liu&Liu,2004;Yang,Ou-Yang,Wu,&Hu,2008).Therefore,45°Cwasselectedasthecenterpointofextractingtem-peratureintheBBDexperiments.3.2.EffectofCaOtorawmaterialratioonextractionratioofresveratrolResveratrolisahydrophobicorganiccompound.Inordertobeextractedeffectivelybywater,resveratrolmustbetransformedintosaltstoimproveitssolubilityandpolaritybyaneutralizationreactionwithCaO.Therefore,CaOdosageisthemostcriticalfactoraffectingtheextractionratioofresveratrol.Inthepresentstudy,theeffectofCaOdosageontheextractionrateofresveratrolwasinvestigatedusingdifferentCaOtorawmaterialratios(2:100,3:100,4:100,5:100,6:100,7:100,8:100and9:100)whileotherextractionvariablesweresetasfollows:extractiontemperatureof45°C,watertorawmaterialratioof7:1,extractiontimeof40min.AsshowninFig.2B,theextractionratioimprovedastheratioofCaOtorawmaterialincreasedfrom2:100to6:100,buttherewerenosigni?cantdifferencesbetween6:100and9:100.Therefore,CaOtorawmaterialratioof6:100wasusedinthefol-lowingexperimentsasanexcessofCaOdosagewouldnotonly20Q.Zhangetal./FoodChemistry179(2015)15–25Fig.2.Effectofextractiontemperature(A),CaOtorawmaterialratio(B),watertorawmaterialratio(C)andextractiontime(D)ontheyieldofresveratrol.Q.Zhangetal./FoodChemistry179(2015)15–25Table2

Estimateparameterofregressionmodel.VariablesModelX1X2X3X4X1X2X1X3X1X4X2X3X2X4X3X4X21X22X23X24ResidualLackof?tPureerrorCorrelationtotalR2=0.9854,adjustedR2=0.9711.Sumofsquares0.2606470.08670.00002130.0046810.0130020.0001440.0001210.0001210.0014820.0000420.0008410.1068840.0416350.0299860.0496530.0039620.0035430.0004190.264609DOF14111111111111111410428Meansquare0.0186180.08670.00002130.0046810.0130020.0001440.0001210.0001210.0014820.0000420.0008410.1068840.0416350.0299860.0496530.0002830.0003540.000105F-value65.7847306.35140.07538116.5392745.94240.5088190.4275490.4275495.2374780.1492892.971644377.672147.1147105.9551175.44613.380646P-valueProb.>F<0.0001<0.00010.78770.0012<0.00010.48740.52380.52380.03820.70500.1067<0.0001<0.0001<0.0001<0.00010.125921Fig.3.Responsesurfaceplots(A,C,E,G,IandK)andcontourplots(B,D,F,H,JandL)showingtheeffectsofextractiontemperature,CaOtorawmaterialratio,watertorawmaterialratioandextractiontimeandtheirmutualeffectsonextractionyieldofresveratrol.22Q.Zhangetal./FoodChemistry179(2015)15–25Fig.3(continued)costmore,butalsoproducemorehydrochloricacid(subsequentacidprecipitationprocess)waste.3.3.EffectofwatertorawmaterialratioonextractionratioofresveratrolRatiosofwatertorawmaterialsigni?cantlyaffectstheextrac-tionrate(Bezerra,Santelli,Oliveira,Villar,&Escaleira,2008).Res-veratrolcanbediffusedmorequicklybyalargerratioofwatertorawmaterial,whichmakesagreaterconcentrationdifferencebetweentheinteriorplantcellsandtheexteriorsolvent.Inaddi-tion,moreresveratrolmoleculescandissolveinwater,resultinginanenhancementoftheextractionratio.Nevertheless,iftheratioofsolventtorawmaterialwastoohigh,theconcentrationofresve-ratroldissolvedinwaterwouldbecomelow,leadingtogreaterenergyconsumptionforsolventvolatilization.Therefore,asuitableratioofwatertorawmaterialshouldbeselectedforextractionofresveratrolinMSCE.Tostudytheeffectofdifferentratiosofwatertorawmaterialonextractionrateofresveratrol,watertorawmaterialratiosweresetat3:1,5:1,7:1,9:1,11:1,13:1and15:1.Otherextractioncon-ditionsweresetasfollows:extractiontemperatureof45°C,CaOtorawmaterialratioof6:100,extractiontimeof40min.Theresultsshowedthattherewasasigni?cantincreasingtrendintheextrac-tionratioofresveratrolasthewatertorawmaterialratiowasincreasedfrom3:1to9:1,butinsigni?cantdifferencebetween9:1and15:1(Fig.2C).Thus,watertorawmaterialratioof9:1wasadoptedinthefollowingBBDexperiment.3.4.EffectofextractiontimeonextractionratioofresveratrolExtractiontimeisanotherimportantfactorthataffectstheextractionef?ciencyandselectivityofMSCE.Ithadbeenreportedthatalongextractiontimefavoredtheproductionofbioactivecompounds(Soria&Villamiel,2010).However,excessiveexten-sionofextractiontimemayalsocausethechangesinmolecularstructureofresveratrol.Therefore,theeffectsofdifferentextrac-tiontimesontheextractionratioofresveratrolwereinvestigated.Extractiontimesof20,30,40,50,60,70or80minweretestedwithotherextractionparametersweresetasfollowings:extractiontemperatureof45°C,CaOtorawmaterialratioof6:100andwatertorawmaterialratioof9:1.AsshowninFig.2D,theextractionratioofresveratrolincreasedasextractiontimeincreasedfrom20to50min,reachingamaximumat50min.Afterthispoint,theextractionratioofresveratrolstartedtodecreasewithincreas-ingextractiontime.Thechangesoftheextractionratiomightbeduetothedifferentchemicalstabilityofresveratrol.Thistendencywasinagreementwithotherreports(He,Gongke,&Zhan,2003).