Determining the optimal shaking rate of a reciprocal agitation sterilization system for liquid foods: A computational approach with experimental validation

Please note that changes made in the online proofing system will be added to the article before publication but are not reflected in this PDF.

We also ask that this file not be used for submitting corrections.

Pleasecitethisarticleinpressas:Erdogdu,F.,etal., Determiningtheoptimalshakingrateofareciprocalagitationsterilizationsystemfor ContentslistsavailableatScienceDirect

Food and Bioproducts Processing

jo u r n al h om ep a g e :w w w . e l s e v i e r . c o m / l o c a t e / f b p

Determining the optimal shaking rate of a

reciprocal agitation sterilization system for liquid foods: A computational approach with

experimental validation

Ferruh Erdogdu

, Mustafa Tutar

, Sigurd Oines

, Igor Barreno

,

Q1

Dagbjorn Skipnes

aDepartmentofFoodEngineering,AnkaraUniversity,Ankara,Turkey

bMechanicalandManufacturingDepartment,MGEPMondragonGoiEskolaPoliteknikoa,Spain

cIKERBASQUE,BasqueFoundationforScience,Spain

dDepartmentofProcessTechnology,Nofima,AS,Norway

eCSCentroStirlingS.Coop,Aretxabaleta,Spain

a r t i c l e i n f o

Articlehistory:

Received22February2016

Receivedinrevisedform19July2016 Accepted22July2016

Availableonlinexxx

Keywords:

Cannedfoods Optimization Modelling

Reciprocalagitation-shaking

a bs t r a c t

Anewcanningprocesswhereareciprocatingagitationiscarriedoutinhorizontallyoriented containershasbeenrecentlydemonstratedtoreduceprocessingtimesandenableenergy savingswithlessdegradationinthequalityofprocessedfoodproducts.Reciprocalagita- tionbyimposingadditionalforcesenhancesconvectivemixingwithincreasedproduction efficiency.Thereciprocalagitationusesthehorizontalaccelerationinadditiontogravity andsumoftheseforcesleadtoaconsiderableincreaseintheheattransferrates.Inthelit- erature,therehavebeenexperimentalapproachestoevaluateheattransferenhancement.

However,duetothebalanceamongtheseforces,theremightbeanoptimumreciprocal agitationratefortheincreasedheattransferdependinguponthephysicalpropertiesofthe liquidprocessed.Therefore,theobjectivesofthisstudyweretodeterminetheoptimum agitationratesbydevelopingacomputationalmodelforheattransfer.Forthispurpose,a multi-phasemodelsimulationwasperformedusingafinitevolumemethodbasedondis- cretizationofgoverningflowequationsforliquidandgasphaseinanon-inertialreference frameofmovingmesh.Experimentalstudiesformodelvalidationwerecarriedoutina reciprocallyagitatedretortusing98.2mm×115mmcanscontainingdistilledwaterwith 2%headspaceasamodelcase.Themodelresultswereinagreementwiththeexperimen- tal data,andtheoptimumreciprocalagitationratewasdetermined.Theresultsofthis studyaretobeusedtooptimizetheprocesswithrespecttoimprovethehealth-promoting compoundsofprocessedfoods.

©2016InstitutionofChemicalEngineers.PublishedbyElsevierB.V.Allrightsreserved.

1. Introduction

Traditional canning has been a convenient way and pro- Q2

videdageneralistandeconomicmethodforprocessingand preservationoffoodproducts.Consumerdemands forhigh

∗ Correspondingauthor.Tel.:+905338120686;fax:+903123178711.

E-mailaddresses:[email protected],[email protected](F.Erdogdu).

qualityfoods,however,forcethefoodprocessorstoimprove and innovate their processing.It isawell-known fact that the shorter the process time ata givenprocess condition, while still achieving the required safety for consumption, theless thedamagetothesensoryand nutritivequalityof thefoodproducts.Basedonthisconcept,followingtheuse

http://dx.doi.org/10.1016/j.fbp.2016.07.012

0960-3085/©2016InstitutionofChemicalEngineers.PublishedbyElsevierB.V.Allrightsreserved.

1

2

3

4

5 6

7 8 9 10 11 12

13 14 15 16 17 18 19 20 21 22 23 24 25

26 27 28

29 30 31 32 33

Pleasecitethisarticleinpressas:Erdogdu,F.,etal., Determiningtheoptimalshakingrateofareciprocalagitationsterilizationsystemfor ofretortsforcanning,theagitationretortswereintroduced

in1920swiththeagitationmechanismbasedonhorizontal axialrotation(Atesetal.,2014).Verticalrotationofthecans waslaterintroducedwiththeend-over-endrotationprinciple (Clifcorn et al., 1950). Introductionofagitation mechanism incanningforliquidorliquid–solidparticlescontainingfood productswastheresultofacertaindisadvantageofthestatic retortsystems(Rosnesetal.,2011).Theprimarychallengein thestatic processingistheslowheatpenetrationresulting inalackofconsistencyinsensoryand nutritiveproperties (Ohlsson,1980).

Consideringtheeffectiveheattransferratesobtainedwith agitation,areciprocatinghorizontalagitationwithrapidback andforthmotionofthehorizontalcontainersinanoscillat- ingwayhasbeenproposedtoincreasetheheattransferrate further,andanagitatingretortwithhighfrequencylongitu- dinalmechanismwasdevelopedin2006(Atesetal., 2014).

Bothhorizontalandend-over-endbasedagitationretortssuf- ferfromthelimitationofthattheappliedforcestoenablethe motionwithinthe containerwere abalance betweengrav- ityandcentrifugalforces(Waldenand Emanuel,2010).Due tothisbalance,theagitationincreasestheheattransferrate uptoanoptimumwhilefurtheragitationmightnotaffector mightinfluencetheprocessinanegativewaydependingespe- ciallyupontheviscosityoffoodproduct.Adetailedanalysis andcomparisonamongthegravityandcentrifugalforcesfor thecaseofaxialrotationeffectsinhorizontalaxialrotation ofcanswerereportedbyErdogduandTutar(2012)andTutar andErdogdu(2012).Thereciprocalagitation,however, used thehorizontalaccelerationinadditiontogravity,andthesum oftheseforcesenabledaconsiderable increaseintheheat transferrateswithreductionsintheprocesstime(Waldenand Emanuel,2010).

Thefirststudiesinthefoodengineeringliteratureusing thereciprocatingagitationsystemswereexperimentalbased to demonstrate the possible process time reductions and improvementsintheheattransferrates.Bermudez-Aguirre etal.(2013a)demonstratedtheimprovementinheattransfer coefficientunderstaticandhorizontalgentle-rockingmodes.

Atesetal.(2014),forexample,comparedthenovelagitating retort and static retortprocesses forbacterial inactivation, anditwasconcludedthatagitatingretortprocesssignificantly loweredtherequiredprocesstime.ThestudybySinghetal.

(2015a)focusedonevaluatingtheheattransferenhancement underreciprocal agitation whileSingh et al. (2015b) devel- oped anexperimental methodology todetermine the heat transfercoefficientincannedparticulatefluidsunderrecipro- catingfrequenciesupto3Hz.SinghandRamaswamy(2015a) focusedontheeffectofproductrelatedparametersonheat transferwhileSinghandRamaswamy(2015b)determinedthe effectoftheorientationofcans duringreciprocating agita- tion thermal processing. Singh et al.(2016) introduced the conceptofreciprocalagitationprocesstoimprovethequal- ityofcannedgreenbeansduringthermalprocessing.Singh andRamaswamy(2016)carriedoutanoptimizationstudyfor theheattransferrateandreciprocationintensityforthermal processingofliquidparticulatemixtures.Thesestudieswere basedonexperimentalapproacheswhileasimilarsituation wasexploredbyLiffmanetal.(1997)andPeschetal.(2008)ina computational–theoreticalwayforconvectionduetohorizon- talshakingandheatedfluidlayerssubjectedtotime-periodic horizontalaccelerations,respectively.

Eventhoughtherewerecertainfindingsreportedforthe effectofreciprocalagitationonthetemperatureincreaseand

enhancedheattransferrate(Bermudez-Aguirreetal.,2013a;

Atesetal.,2014;SinghandRamaswamy,2015a,b,2016;Singh etal.,2016),developmentofacomputationalmodel(withone exceptionwheretheheattransfercoefficientbasedlumped modelwithoutconsideringthetemperaturedistributionwas introducedbyBermudez-Aguirreetal.(2013b)anddetermin- ing the optimalagitation rates were not focusedindetail.

For determiningtheoptimalconditions, oneexceptionwas reported by Singh and Ramaswamy (2016) wherethe opti- malconditionsofreciprocationintensityforliquidparticulate mixtureswereexperimentallydetermined.Theoptimization studiesbasedonacomputationalmodelaresignificantsince the computational model mightalso beused alsofor pro- cessdevelopmentpurposes.Therefore,theprimaryobjective ofthisstudywastodeterminetheoptimalagitationratein areciprocalagitation processusinganexperimentallyvali- datedcomputational model.Thesecondaryobjectives were firsttodevelopacomputationalnumericalmodelforheatand momentumtransferinsidethereciprocallyagitatedcansto determinethetemperaturedistributionandvelocitychanges andthenexperimentallyvalidatethemodel.

2. Materials and methods

Forthegivenobjectives,thestudyconsistedofexperimental and computationalparts.Intheexperimentalpart, awater filledcanwasprocessedinboilingwaterandagitatingcondi- tions.Inbothcases,thehorizontallyorientedcancontained waterasatestliquidtorepresentalowviscosityNewtonian liquid. Thetime–temperaturedataobtainedatthegeomet- riccenterinthefirstexperimentswereusedtodevelopand validatethecomputationalmodel,todecideuponthecom- putationalparameterswiththemeshindependencystudies.

Followingthemeshindependencystudy,thecomputational model was validated with the temperature data obtained under horizontal agitating conditions, and the model was appliedto horizontalagitation ratesfrom 20 to140rpmto obtain the agitationrateinthe directionofthe axisofthe horizontal canresulting in maximumheat transfer. In the reciprocalagitation systems,the crankshaft,usedtoderive the horizontalmotion,angularvelocityisrelatedtoengine revolutionspermin(rpm).

2.1. Experimentalmethodology

Thefirststepinthisstudywastodecideuponthecomputa- tionalparametersandtesttheaccuracyofthecomputational method. For this objective, an experimental study with a cannedwatersample(98.2mm×115mmcansfilledwithdis- tilledwaterwith2%headspace)wascarriedoutinaMicroflow 911 EAT Shakaretort (Steriflow, Roanne, France)in boiling water under stationaryconditions. The retort system was heatedbydirectsteaminjection,equippedwithapreheating tanktoprocesswater.Thisprocessedwaterwasthencircu- latedthroughaheatexchanger(onlyusedforcooling)tothe retortandspreadbyaperforatedplatetoobtainwaterraining overthecans.Thecanwasfixedinahorizontalpositioninthe boilingwater.Type-Tthermocoupleconnectedtoadatalog- gerE-ValFlez(Ellab,Copenhagen,Denmark)waslocatedatthe geometricalcenterusingringgasketsandlocking-receptacles.

Theexperimentalset-upwasshowninFig.1.

The canmaterial wasa steelsheet witha thicknessof 0.19mmandthermalconductivityvalueof15–16W/m²K.This enabledtheassumptionofthenegligibleconductioneffectof

34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97

98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118

119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136

137

138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156

Pleasecitethisarticleinpressas:Erdogdu,F.,etal., Determiningtheoptimalshakingrateofareciprocalagitationsterilizationsystemfor Fig.1–Experimentalset-upwherethecanwasfixedina

horizontalpositionwithinstalledtype-Tthermocouples.

canwallontheheattransfer,andthemediumtemperature wasacceptedtobethecansurfacetemperature.

Inthesecondgroupoftheexperiments,thesamecanwas usedunderhorizontallyagitationconditionswheretheshak- ingratechangedfrom0rpm(inthefirst26softheprocess) to 80rpm (at the 35s of the process). Since the horizon- tal-acceleratedagitation rates were included inthis model

validationpartofthestudy,theagitationrateswerefirstcon- vertedtothetangentialvelocityvalues.Forthispurpose,since a horizontalagitatedsystem used aslider –crank derived mechanism(Fig.2a–modifiedfromReaderandHooper,1982), displacement of the can during the agitation process was definedwiththefollowingequation:

x_p=

r−r·(1−sin(ω·t))+n·

1−

1−cos² (ω·t) n²

(1) wheren=L/r,xisthedisplacement(m),ωistheshakingrate (rpm), t isthetime (s),risthe crankradiusofthe system (0.075m),andListhelengthoftheconnectivityrod(0.5m).

Since (n²>>cos²(ω·t)), this equation was then simplified with:

xp=[r·sin(ω·t)] (2)

ThecomparisonofEqs.(1)and(2)forthechangeofdis- placementwithrespecttothe(ω·t)valuesdidnotshowany significantdifferenceresultinginverysamedisplacementval- ues.Therefore,thesecondequationwithitssimplifiedform

Fig.2–(a)Ahorizontalagitatedsystemwithaslider–crankderivedmechanism(modifiedfromReaderandHooper,1982);

(b)reciprocalagitationrateusedinthesecondsetofmodelvalidationexperiments;(c)tangentialvelocitychangeinthe transitionperidfrom0ro80rpmreciprocalagitationrate.

157 158 159 160 161 162 163

164 165 166 167 168 169

170

171

172 173 174 175 176

177

178 179 180 181

Pleasecitethisarticleinpressas:Erdogdu,F.,etal., Determiningtheoptimalshakingrateofareciprocalagitationsterilizationsystemfor waschosentoderivethetangentialvelocityequationforthe

reciprocalagitationsystem:

v=r·ω·cos (ω·t) (3)

However,thisequationbroughttheissueofnon-zeroveloc- ityatthebeginningofthehorizontalagitationprocess.This resultedintheinstabilitiesandnon-convergenceproblemsin thenumericalcomputations.Topreventthisandtomakesure thattheprocessstartwitha‘0’velocityinitially,themovement was displaced with oneperiod, and the following velocity equationwaspreferredtostartwith:

v=r·ω·sin (ω·t)

limt→0v=0 (4)

Asexplainedintheexperimentalset-upformodelvali- dation,followingtheinitialsteady26s,reciprocalagitation transitionofthesystemwasfrom0toahigheragitationrate of80rpm[0.623m/s].Thistransitionperiodtook9sandwas assumedtofollowanexponentialincrease.Theexponential increasefrom0velocityto80rpmreciprocalagitationratewas preferredtoenablethesmoothtransitionbetweentheserates.

Theexponentialincreasewastheonlywayforthissmooth transitionandtopreventtheovershootafter9softhetran- sition period. Amongvarious trials, the linear increase for exampleresultedinasharptransitiontothe80rpm,which might be rather difficult to control physically. To conform thetransitionperiodwith80rpmofreciprocalagitationrate after9ssmoothly, thegivenvaluesbelowforthefollowing transitionstageequation,wherethe velocitychangeinthe transitionperiodwasshown,enabledthis:

v= 80[rpm]

60[s/min]·[1−exp(−k·t)] (5) where(k=1)isthe constant(s⁻¹)and t=(t−26)(s).Fig. 2b showsthehorizontalagitationratethroughtheexperiments whileFig.2cdemonstratethevelocityprofilefrom0to80rpm inthetransitionperiodof9s(from26to35s).Thechangein thehorizontalagitationrate,asreportedinFig.2b,c,andthe variable–experimentallyrecordedmediumtemperaturewere usedinthemodelvalidationcasetocomparethenumerical resultswiththeexperimentaloneobtainedatthecentreof thecan.

Forbothcases,3-experimentswerecarriedout,theaver- agevalueswiththestandarddeviationwereusedinthemodel validation. Sincethe standard deviationsofthe average of thetemperaturechangebasedon thesethreeexperiments, additionalexperimentswereavoided.

2.2. Governingequationsandthecomputational model

The numerical methodology and full scale model experi- mentaltestingverificationsproposedausefulcomputational algorithmfordynamicmonitoringofheadspace(air)andliq- uid(water)interactionsthroughtheagitationandsolvedthe fluid-thermal energy interactions in order to optimize the reciprocalagitationprocess.Thetwo-phasevolumeoffluid (VOF)approachaccompaniedwiththefinitevolumemethod (FVM)basednumericaldiscretizationschemewasutilizedin thesimulationoftwo-phaseflowundervaryingphysicalcon- ditionsthrough unsteady, three-dimensionaland turbulent

flowsimulationsforthegivenRayleighnumberrangeover1E9 attheinitialphaseoftheheatingasexplainedbelow.

The basic mathematical model for the discretization processincludedthesolutionoffundamentalgoverningequa- tionsoffluidflowmotion,knownascontinuityequationand momentumconservationequations,i.e.,Navier-Stokes(N-S) equationsforincompressiblefluidinanon-inertialframe:

2.3. Continuityequation

∂

∂t +∇v_r = 0 (6)

2.4. Momentumequation

∂

∂t(vr)+∇(vrvr)=−∇P+∇¯¯r+ F (7) wherewasthedensity(kgm⁻³),twasthetime(s),vrwas the relativevelocityvector ofafluidparticle (ms⁻¹),Pwas the staticpressure (Pa), ¯¯_r wasthe stresstensor (described below),F wastheexternalbody force(N)includinggravita- tionaleffectsandaccelerationduetothenon-inertialframe motion.Thestresstensor, ¯¯_rwas:

¯¯

=[(∇v_r+∇v^T_r)−2

3∇v៭_rI] (8)

wherewasthedynamicviscosity(Pas).Itwasdefinedtobe atemperaturedependentpolynomialfunction.Iwastheunit tensor,andthesecondtermontherighthandsidewasthe effectofvolumedilation.Thevolumedilationwasneglected inthesolutionssincetherewasnoeffectintheprocess.Energy conservationequation,alsosolvedforthepresentflow,was writtenintermsofrelativeinternalenergy(Er)andrelative totalenthalpy(Hr):

2.5. Energyequation

∂

∂tE_r+∇(v_rH_r)=∇(k∇T+¯¯_r)+S_h (9) where

E_r=h−P +1

2(v²_r−u²_r) (10)

Hr = Er+ P

(11)

Velocityevolutionswerethentransformedfromstationary torotatingframeusing:

vr = v− ur (12)

where vr wasthe relative velocity(ms⁻¹)arising from the meshmotion(velocityviewedfromthemovingmeshofthe oscillatory reciprocatingmotion),v wasthe absoluteveloc- ity (ms⁻¹)(velocity viewedfrom thestationaryframe),and

ur wasthelongitudinalvelocity(ms⁻¹)(velocityduetothe movingmesh).Theabovegoverningequationsweredirectly discretizedwithafinitevolumemethod(FVM)inconjunction withaninterfacetrackingmodel(asdescribedbelow)forthe air–liquid system. Reynolds-averaged Navier-Stokes (RANS) basedformoftheseequationswerediscretizedtogetherwith thetransportequationsofturbulencekineticenergyandits dissipationwithinthefinitevolumeschemebyusingtheRANS

182 183

184

185 186 187 188 189 190 191

192

193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208

209

210 211 212 213 214 215 216 217 218 219 220 221 222 223

224 225

226 227 228 229 230 231 232 233 234 235

236 237 238 239 240 241 242

243

244

245

246

247 248 249 250 251 252

253

254 255 256 257 258 259 260 261

262

263

264

265

266

267 268

269

270 271 272 273 274 275 276 277 278 279 280 281

Pleasecitethisarticleinpressas:Erdogdu,F.,etal., Determiningtheoptimalshakingrateofareciprocalagitationsterilizationsystemfor basedk-εturbulenceclosuremodel(YakhotandOrszag,1986)

accompaniedwith the utilized interface tracking model at higheragitationrates.

Julienetal.(1996)reportedthe‘soft’and‘hard’turbulence conditionsledbythehigherRayleighnumbers,andthelower rangeswheretheRayleighnumber(Ra)wassmallerthan1E7, wascharacterizedbysoftturbulenceconditionsbyKooijetal.

(2015).TheinitialRayleighnumber forthefirstexperimen- talcasewas 2.66E9(thisvaluewas obtainedatthe 0.5sof themodelvalidationsimulationwheretherewasnomove- mentofthecan)whereitwasbeyondthesoftturbulencecase.

Eventhough theRayleighnumber, asaproductofGrashof andPrandtlnumber,determinedtheturbulenceinducement inthenaturalconvectionflowinthecylindricalcavity,local cellReynolds(ReL)numberchangewasalsocontrolled.Itwas upto30alongthesurfaceofthecanattheinitialphaseofthe stationarymodelvalidationcase(0.5s)whileitincreasedupto 100towardstheend.TheturbulenceReynoldsnumber(Re_T) was,ontheotherhand,highenoughtoresolvethepresent flowwitha turbulencemodel, withits instantaneous local valueupto18,400initiallyatthecentreofthehorizontalcylin- dricalcavity.TheRayleighandlocalReynoldsandturbulence ReynoldsnumberweredeterminedusingEq.(13):

Ra= g·ˇ·T·D³ (/)² ·Pr Re_L=Vc1/3·v·

ReT=k²·

·ε

(13)

where g was the gravitational acceleration (m/s²), ˇ was thermalexpansioncoefficientforwater(1/K), Twasthemax- imumtemperaturedifferencebetweentheheatingmedium andtheinitialtemperatureofthesystem(K),Dwasthechar- acteristicdimension(diameterofthecylindricalcavity,m),Pr wasPrandtlnumber,andwasthedynamicviscosity(Pas), was the density (kg/m³), was the velocityencountered inagivencell(m/s),andV^1/3_c wasthecharacteristiclength ofthelocalcell,andkεwereturbulencekineticenergyand dissipationrate,respectively.BesideshighRayleighnumber encounteredattheinitialphaseoftheprocess,thelaminar flowconditionwasstilltestedforconvergenceduringtheini- tialtestsimulations,butthesetrialsresultedinconvergence problems.Therefore,basedontheRayleighnumberinforma- tionforturbulenceconditionsandconsideringtheresultsof theinitialsimulationsformodelvalidationpurposes,thetur- bulencemodelwasactivatedinthesimulations.Inaddition, forthesimulationstudycarriedoutundersteadyconditions formodelvalidation–meshindependencystudy,theturbu- lenceReynoldsnumberwasaround80towardtheendofthe simulation.

Fortheturbulencemodel,thefollowingturbulenceparam- eterswereapplied:

- Initial turbulence intensity (I) was assumed to be 5%, Basedonthemaximumtangentialvelocityvalueof1.1m/s (obtainedbyEq.(4)),theturbulencekineticenergyvalue(k) was0.00453m²/s²:

k= 3

2·(v_max·I)²=0.00453 (14)

- Turbulencedissipationrate(ε)was0.025m²/a:

ε=C^3/4 ·k^3/2

L =0.025 (15)

whereLwastheturbulentlengthscale(0.002m),andCwas turbulencemodelconstant(0.09).

Thetrackingofinterface betweenair–water phaseswas accomplishedthroughthevolumeoffluid(VOF)methodpro- posedbyHirtandNichols(1981).Inthismodel,asinglesetof momentumequationswassharedbythefluidsandthevol- umefractionsofeachofthefluidsineachcomputationalcell weretrackedthroughthedomain.Thefieldsforallvariables andpropertiesaresharedbyphasesandrepresentsvolume- averagedvaluesaslongasthevolumefractionofeachofthe phases isknown ateach location. Thus, thevariables and propertiesinanygivencell areeitherpurely representative ofoneofthephasesorrepresentativeofmixtureofphases dependingonthevolumefractionvalues.Thevolumefrac- tionsofwaterandairinthecomputationalcellsumtounity.

Interfacetrackingwascarriedoutbysolvingcontinuityequa- tionforvolumefractionofoneofthephaseswhereairwas specifiedasprimaryphaseandthusthevolumefractionofthe liquidphasewassolved.InadditiontoVOFmethod,Ubbink’s compressive interfacecapturingscheme (Ubbink andIssaa, 1999)forarbitrarymeshes(CICSAM)wasalsoapplied.

For the numerical solution procedure, a finite volume method (FVM)basedsolver (Ansys Fluent V15, Ansys,Inc., Canonsburg,PA,USA)wasusedtosolvetheprecedingpartial differentialgoverningequationsofthepresenttwo-phaseflow problem.Intheproposedcomputationalmodel,thecollocated FVMwasemployedtodiscretizethegoverning3Dflow-energy equations.Alltherequiredthermaland physicalproperties forairand waterphaseswere temperaturedependent and reportedinErdogduandTutar(2012).Initially,waterinthecan inbothexperimentalconditionwasatrestandhadtheinitial temperatureof300.92and301.38K inthesteadyandagita- tioncases,respectively.Whiletheboilingwatertemperature andvariablemediumtemperatureswereusedinthemodel validationsimulation,auniformconstantwalltemperatureof (Tw=373.15K)wasusedtodeterminetheeffectofreciprocal agitationratesonthetemperatureevolution.Forthecaseof agitationprocess,theheatingmediumtemperaturewasvari- able,butitwasstillusedtobeasaconstantwalltemperature overthecansurfaceduetotherapidmovementofthecandur- ingtheprocess.Overtheinitialperiodoftheagitationprocess wheretherewasnoreciprocalmovementofthecan(Fig.2c), theheatingmediumtemperatureandtheinitialtemperature ofthecanweresimilar.Therefore,thegivenassumptionwas assumedtoholdtrueduringtheinitialperiod.Thesurfaceten- sionvaluealongtheinterfaceofairandwaterwasassigned tobe0.72N/m,andthetimestepsizeusedinallsimulations was1E−4s.

3. Results and discussion

Usingtheresultsofthefirstexperimentaldataset,thesimu- lationschemesweredecided:

- apressurebasedsolverwiththeabsolutevelocityformula- tion,

282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304

305

306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328

329 330 331 332

333

334

335

336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383

384

385 386

387 388

Pleasecitethisarticleinpressas:Erdogdu,F.,etal., Determiningtheoptimalshakingrateofareciprocalagitationsterilizationsystemfor Fig.3–Initialphasecontoursofthegeometrywithair(atthetop)andwaterphases(a)andthemeshstructuresadoptedfor (b)theminimum–199,728and(c)themaximum–337,800sizedmeshes.

- the pressure-velocitycouplingwas carriedout withPISO (pressureimplicit withsplittingofoperator)schemewith skewness-neighbourcoupling,

- transientformulationwasfirstorderimplicit,and

- spatial discretization forgradient was Green-Gaussnode based; for pressure PRESTO; for momentum first order upwind; forvolume fractionCICSAM;and forturbulence kineticenergyfirstorderupwindschemeswereused.Even though the first order upwind scheme is toodissipative to stabilizethe computation, the initialsimulation stud- iesconfirmedthatthegivensolutionschemerssuitbetter for straight convergence and stabilized computation for the chosen timestep sizeand meshresolution. Besides, regardingtheorderofthediscretizationscheme,thesys- temuncertaintyaswellastheturbulencemodeluncertainty mightbelargerthantheerrorcausedbynumericaldissipa- tion.

Usingtheseschemes,thecomputationalmodelwasfirst appliedtostudymeshindependencyandhencetodetermine thefinalmeshconfigurationbasedonthefirstsetofexperi- mentalresults.Then,themodelvalidationstudywascarried outunderareciprocalagitationcondition,andthereciprocally agitatingspeedsfrom20to140rpmwerethentestedfortem- peraturechangeduringtheagitationtodeterminetheeffect ofagitationandoptimumagitationrate.

Fig.3showstheinitialphasecontoursofthegeometrywith headspace–air(atthetop)andwaterphasesandthemesh structuresadoptedfortheminimum(199,728cells)andmax- imum(337,800cells)numberedmeshconfigurations.Fig.4a showstheresultsofmeshindependencystudywithrespect toexperimentaldataobtainedatthecentreofthehorizontal canlocatedinboilingwater.Therewasnotasignificantdiffer- encebetweenthe199,728and286,080cellswhilethe337,800 cellstructuredmeshover-predictedthetemperature(Fig.4a).

Thisdifferencemightbeduetothecontextofmesh-density –round-off errorrelation.Though round-off errorsmay be accumulatedmorewithhighernumberofmeshcells,further

round-offanalysismightberequiredtoidentifytheireffect onthemeshstructure andresolutionfordifferenttime-step sizes.However,itshouldbeemphasizedthattheflowsystem uncertaintyaswellastheturbulencemodeluncertaintyfor

280 290 300 310 320 330 340 350 360

0 30 60 90 120

Temperature (K)

Time (s)

Experimental 199728 cells 286080 337800

(b) (a)

290 300 310 320 330 340 350 360

0 15 30 45 60 75

Temperature (K)

Process me (s)

Experiment Series1 Medium

Fig.4–Comparisonofexperimentaldatawiththe simulationresults(a)experimentaldataobtainedatthe centreofahorizontallyplacedstaticcaninboilingwater with;(b)experimentaldataobtainedatthecentreofa horizontallyplacedstaticcanunderreciprocallyagitating conditions(themeshstructureusedinbothcomputational modelshad199,728cells).

389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405

406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425

426 427 428 429

Pleasecitethisarticleinpressas:Erdogdu,F.,etal., Determiningtheoptimalshakingrateofareciprocalagitationsterilizationsystemfor -1.5

-1 -0.5 0 0.5 1 1.5

0 1 2 3 4

Tangenal Velocity (m/s)

TIme (s)

20 rpm 80 rpm 140 rpm

Fig.5–Comparisonoftangentialvelocitychangeversus reciprocalagitationrates.

thechoiceofmeshresolutionandtimestepsizeaccompa- niedcould besignificant onthe flowresultsinaddition to theorderofthespatialdiscretizationschemeandrelaxation parameters.Therecould/maybenostraightforwardsolution forminimumnumericaldiffusionandhigheraccuracywith useofveryhighmeshresolutionaccompaniedwithsmaller timestepsizeandhigherorderspatialdiscretizationscheme.

Themeshindependencysimulationsandtheinitialsimula- tions havedemonstratethat the selectedtimestep sizeof 1E−4sforthegeneratedmeshresolutionof199,728meshcells were accurateenough toobtain resultswhich wouldbe in goodcorrespondencewiththeexperimentaldata.Therefore, basedonthemeshindependencyresults,themeshstructure with199,728cellswasusedinthesecondpartofthemodel

290 300 310 320 330 340 350 360

0 20 40 60 80 100

Tavg (K)

Time (s)

0 rpm 20 40 60 80 140

330 335 340 345 350

40 50 60 70 80

Tavg (K)

Time (s)

60 rpm 80 140

Fig.6–Effectofreciprocalagitationrateonthevolume averageincreaseoftemperature.

Fig.7–Phasecontoursinthevariousx-andz-planesofthe computationalgeometryatthe(a)beginning(1s);(b)30s;

and(c)90softhe20rpmreciprocalagitationcase.

430 431 432 433 434 435 436 437 438 439 440 441 442 443

Pleasecitethisarticleinpressas:Erdogdu,F.,etal., Determiningtheoptimalshakingrateofareciprocalagitationsterilizationsystemfor Fig.8–Phasecontoursinthevariousx-andz-planesofthe

computationalgeometryatthe(a)beginning(1s);(b)30s;

and(c)90softhe80rpmreciprocalagitationcase.

validationandsimulationstodeterminetheeffectofrecipro- calagitationrate.

Followingthis,themodelwasvalidatedcomparedtothe experimentaldata obtained under different reciprocal agi- tationconditions(summarized inFig.2).Fig. 4b showsthe comparisonofthecancentretemperaturedatawithrespectto

Fig.9–Temperature(K)contoursinthecentralx-and z-planeofthecomputationalgeometryatthe(a)beginning (1s);(b)30s;and(c)90softhe20rpmreciprocalagitation case.

themodelresultsforthefirst75softheprocess.Asobserved inthisfigure,themodelresultsdemonstratedthevalidityof the developedcomputationalmodel. Even thoughthesim- ulation resultscompared well withthe experimentaldata, there wasa differencebetween the simulationresults and experimentaldata.However,consideringthecomplexnature

444 445 446 447 448 449

450 451 452 453 454 455

Pleasecitethisarticleinpressas:Erdogdu,F.,etal., Determiningtheoptimalshakingrateofareciprocalagitationsterilizationsystemfor Fig.10–Temperature(K)contourswithinthecentralx-

andz-planeofthecomputationalgeometryatthe(a) beginning(1s);(b)30s;and(c)90softhe80rpmreciprocal agitationcase.

oftheprocessandexperimentalconditions,themodelpre- dictionscaught thetrend oftheexperimentaltemperature data.After75softheprocessing,themodelvalidationcase studydidnotcontinuetorunduetoveryhighrequirementof

computational time. Ittook 39.3hto complete a1s ofthe simulationforvalidationpurposeunder80rpmreciprocalagi- tationinanIntelZeon4-Core,3.7GHz–32GBRAMsystem.

3.2. Effectofreciprocalagitationrate

Aftervalidatingthemodel,effectofthereciprocalagitation ratesfrom20to140rpmonthetemperatureevolutioninthe cansweretested.Fig.5showsthecomparisonoftangential velocity values for 20, 80 and 140rpm reciprocal agitation ratesversustimewhileFig.6showstheeffectofreciprocal agitation rates on the volume average temperature (T_avg) increaseofthecanforthefirst100softheprocess.Inthese simulations,theboundarytemperaturewassettobeboiling conditions as reported above in the first model validation case. As demonstrated in Fig. 6, the effect of reciprocal agitation rate was noticeable, and the increased agitation rates increased the temperature evolution especially until 80rpmreciprocalagitation.Theeffectoffurtherincreasing the reciprocal agitation over 80rpm did not result in any significantdifference.Thiseffectofreciprocalagitationuntil amaximumagitationvaluemightbeexplainedbythebalance

Fig.11–Temperature(K)contoursinthecentralx-and z-planeofthecomputationalgeometryatthe(a)beginning (1s);(b)90softhe0rpm(naturalconvectionheatingcase).

456 457 458 459

460 461 462

463

464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479

Pleasecitethisarticleinpressas:Erdogdu,F.,etal., Determiningtheoptimalshakingrateofareciprocalagitationsterilizationsystemfor between the agitation,gravitational buoyancy and viscous

forcesgoverningthereciprocalagitationprocess.

AsindicatedbyBoonpongmaeeandMakotani(2009)and TutarandErdogdu(2012),theflowfield duringanagitation processinvolves complexinteractionsofcentrifugal–rota- tional,gravitational buoyancyand viscousforces.Basedon thisconcept,the followingforceanalysiswasperformedto betterunderstandthetemperatureevolutionduringtherecip- rocalagitationprocessasafunctionofgravitational,agitation andviscousforces:

f_cbf f_gbf =ω²r

g (15)

f_inertial f_viscous= ω²r⁴

² (16)

whererwasthecrankradiusofthesystem(0.075m),gwas thegravitationalacceleration(9.81m/s²),andwaskinematic viscosity(m²/s), f_cbf,f_gbf,f_inertial andf_viscouswere centrifugal buoyancy,gravitationalbuoyancy,agitationrelatedforcesand viscousforces,respectively,andωwasthedimensionalspeed ofrotation(1/s).ωwasshownby(ω=2f/60)wherefwasthe horizontalagitationrate(rpm).Eqs.(15)and(16)represented FroudeandTaylornumbers,respectively.

The(ω²r)value, infact, showedthe effect ofhorizontal agitationrateoverthegravitationalacceleration,anditwas definedtobethereciprocationintensity(g0):

g₀=ω²·r·

1+ r

L (17)

whereLwasthelengthoftheconnectivityrod(0.5m)of theslidercranktypesystemusedintheexperimentalstudies.

Consideringthat[(r/L=0.15)<1],thereciprocationintensity wasapproximatedby:

g₀=ω²·r (18)

This was used in the definition of Froude number to demonstratethereciprocationeffectovergravitationalforce.

Froudenumberincreasedfrom0.03to1.64withtheincrease ofthehorizontalagitationratefrom20to140rpm.Simulta- neously,Taylornumberincreasedfrom2.26E8to1.18E10,from 3.65E9to2.29E10andfrom1.13E10to6.87E10atthe100sof 20,80and140rpmshaking,respectively.Theincreaseratio ofTaylornumberwas5.22,6.27 and6.08(theseratios were betweentheTaylor numbervaluesobtainedatthe100and 1softheprocess)forthese3-reciprocalagitationrates.This change in the Taylor number indicated that the reciprocal agitationforcesstartedshowingtheirimpactevenatlowagi- tationrates,butthiseffectincreasedtoacertainhighestvalue at80rpm.Theeffectofinertialforcesobtainedbythehori- zontalagitationoverviscousforcesasaninternalresistance oftheprocessedliquidwasnotsignificantbeyond80rpm.The furtherincreaseafter80rpmagitationratedidnotmakeany significanteffectontheTaylornumberincreaseinthegiven process,anddidnotleadtoanyfurthertemperatureincrease.

In fact, the volumeaverage temperature increaseobtained with80rpmagitationwasoverthecaseof140rpmtowards theendoftheprocess(Fig.6)asalsoindicatedbytheincrease

Fig.12–Instantaneousvelocityvectors(m/s)onx–zandy–zmid-planesoftheflowdomainat(a)1s(verybeginningofthe agitation)and(b)90s(towardstheendoftheagitation)at20rpm.

480 481 482 483 484 485 486 487 488 489

490

491

492 493 494 495 496 497 498 499 500 501 502

503

504 505 506 507

508

509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530

Pleasecitethisarticleinpressas:Erdogdu,F.,etal., Determiningtheoptimalshakingrateofareciprocalagitationsterilizationsystemfor ratioinTaylornumber.Froudenumberwas0.54at80rpm,and

thisvaluewas50%ofthegravitationalaccelerationeffect.

AsindicatedbyWaldenandEmanuel(2010),thereciprocal agitationapparentlyusedthehorizontalaccelerationinaddi- tiontogravityforces,andthesumoftheseforcesenableda considerabletemperatureincreasewhilethefurtherincrease after80rpmwas preventedviatheeffect ofviscousforces.

Eventhoughthereciprocationintensitywas1.64timeshigher thantheregularnaturalconvectioncasegovernedbythegrav- itationalaccelerationat140rpm,thetemperatureincreasedid notgain anymorebenefitfrom this highrate ofagitation.

Thelowerviscosityoftheliquid(water)alsoplayedasignifi- cantroleinthisconcept,anditseffectisexpectedtobemore pronouncedforthecaseofnon-Newtonianhigherviscosity liquids.Trevino(2009)alsonotedthattheoscillatingtechnol- ogywithrackingmovement(reciprocallyagitation)mightnot bebeneficialforprocessinglow(1%starch–watermixture)and high (5% starch–watermixture)viscosity productswhere a comparisonoftheeffectsofoscillatingandstaticretortther- malprocessingwascarriedout.However,itwasalsostated thataneffectiveheattransferratemightbepossibletoobtain formediumviscosity(3%starch–watermixture)products.

Tutar and Erdogdu (2012) explained that, in agitation relatedprocesses,headspace–airbubblemovesthroughvia thegiveneffectofagitationandviscosityforces.Thisleadsto themixingtoincreasetheheattransferrate.Forthecaseof lowviscosityNewtonianliquids,however,thismixingeffectis

hardlyseen,andtheairbubblegenerallymightmovethrough thetopovertheprocessunderacertaineffectofagitationrate.

Figs.7and8showthephase(Phase2–headspace)contours (headspaceshownwithredcontouratthebeginningofthe process)atthebeginning(1s),30and90softhe20and80rpm cases inthe various x- and z-planesof the computational geometry,respectively.AsobservedinFig.7,headspacemoved atthetopofthegeometrycontinuouslywiththegivenmove- mentofthecanat20rpmreciprocalagitationrate.However, the80rpmagitationledtoanabruptchangeoftheheadspace distributionduetothesuddenstartofthehigheragitation.

Theconsiderabledifferencebetween20and80rpmagitation rateswasalsoshowninFig.5.Fig.8ashowsthesuddendisrup- tionoftheheadspaceatthebeginningcomparedtothecase of20rpmandacertaininhomogeneousdistributionthrough theprocess(Figs.8b,c).Atthe90softheprocess,basedon thephasecontours,itmightbeassumedthatalowamount ofheadspacewasmixedinthewaterat20rpmagitationrate whilethiswas higherat80rpmagitation rate.Thismixing alsobroughtaconsiderableandhomogeneoustemperature increasecomparedtothecaseof20rpm.

Figs.9and10showthetemperaturedistributionthrough the canfor20and 80rpm,respectively.Themorehomoge- neous distribution natureof the temperature contoursare observedat80rpmagitationratewhilethe20rpmagitation rate resembled more like a natural convection case with the distinctly stratified temperature contours. The natural

Fig.13–Instantaneousvelocityvectors(m/s)onx–zandy–zmid-planesoftheflowdomainat(a)1s(verybeginningofthe agitation)and(b)90s(towardstheendoftheagitation)at80rpm.

531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557

558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584

Pleasecitethisarticleinpressas:Erdogdu,F.,etal., Determiningtheoptimalshakingrateofareciprocalagitationsterilizationsystemfor convectioncasewassummarizedinFig.11.Figs.12and13,on

theotherhand,comparativelyrepresentedtheinstantaneous velocityvectorsobtainedonthexzandyzmid-planesofthe flow domain at1 (very beginning of the agitation motion) and90s(towardstheendoftheagitationmotion)at20and 80rpm,respectivelytofurthercomparetheagitationeffects forthefieldofflowmotion.Whentheagitationstarted,the headspacewasimmediatelyaffected,anditcouldnotmain- tainitsshapeandorientationwithrespecttothewaterflow domain,whichwasacceleratedlikeasolidbodymotiondue tothehorizontalaccelerationasseeninFigs.12aand13a.As theagitationmotioncontinued,dynamicpressuredeveloped intheflowdomain,anddifferentspatialandtemporalevolu- tionofthefluidflow,dependinguponthesimulationtimeand agitationrate,wereclearlyidentifiedfromtheinstantaneous velocity vectors in the vicinity of the air–water interface whoseshapeandpositionbecamehighlyunstablewiththe increasingagitation rate (Figs. 12b and 13b). Superposition ofinertial,agitation andnaturalforcesduetogravitational forces(mixingconvectiveforces)became moreevidentand effectiveontheair–waterflowdomainathigheragitationrate of80rpm,leadingtohigherinertialandconvectiveinstabili- tiesofthepresenttwo-phaseflowsystemwithmorecomplex, three-dimensionalflowbehaviourinthewholedomainsys- tem.Thisbehavioureventuallywouldmakeapositiveeffect onthetemperatureevolutionfor80rpmcomparedto20rpm aspreviouslyobservedinthetemperaturecontoursinFig.10.

With this positive effect, forced convective heat transfer mechanism on the temperature evolution became more dominantastheagitationrateincreasedupto80rpmanddid notchangeathigheragitationratesasalsoexplainedabove.

4. Conclusions

This study introduced determining the optimal reciprocal agitation rate based on the experimentally validated com- putational model. The computational model was used to determinethetemperaturechangeinacanfilledwithwater with 2% headspace undergoing a reciprocal agitation pro- cess,andthetemperatureevolutionduringtheprocesswas determinedtobeundercontrolofreciprocationintensityand theratioofagitationandviscousforces.Thecomputational resultsindicateda certainlimitofreciprocal agitationrate intheviewoftemperatureincreaseindicatingthesignificant effectofviscosityand inertialforcesobtainedbytheagita- tion.ForaNewtonianlowviscosityliquidcase,represented bywater,the80rpmreciprocalagitationratewasdetermined tobeanoptimumrate.

Itwouldbevaluabletodeterminetheoptimumagitation conditionsforhighviscositynon-Newtonianliquidsconsid- eringthatasignificantportionofthefoodproductsprocessed incanslieinthiscategory.Besides,afurtherstudytodemon- stratetheeffectofheadspacevolumetoincreasetheagitation ratesforliquidandparticulatefoodproductswouldalsobe required.

Acknowledgement

This study was developed within the framework of the ERA-net SUSFOOD Sunniva project, “Sustainable food pro- ductionthroughqualityoptimizedrawmaterialproduction andprocessingtechnologies forpremiumqualityvegetable productsandgeneratedby-products”.Thefollowingacknowl- edgementsarerecognizedbytheauthors:

FerruhErdogduacknowledgestheMinistryofFood,Agri- cultureandLivestock(GDAR)ofTurkeyforthetravelsupport toattendtheprojectmeetings.

Mustafa Tutar acknowledges the support from the Daniel and Nina Carasso Foundation, France Q3 (www.fondationcarasso.org)andtheDepartmentofEconomic DevelopmentandCompetitiveness(ELIKA).

SigurdOinesandDagbjornSkipnesacknowledgethesup- portfromtheResearchCouncilofNorwaythroughgrantno.

NO10829.

References

Ates,M.B.,Skipnes,D.,Rode,T.M.,Lekang,O-I.,2014.Comparison ofbacterialinactivationwithnovelagitatingretortandstatic retortaftermildtreatments.FoodControl43,150–154.

Bermudez-Aguirre,D.,Lima,F.,Reitzel,J.,Garcia-Prez,M., Barbosa-Canovas,G.V.,2013a.Evaluationoftotalheattransfer coefficient(hT)duringinnovativeretortprocessing:static, gentlemotionandrockingmode.In:IFTAnnualMeeting, Abstractnumber:031-02.

Bermudez-Aguirre,D.,Lima,F.,Reitzel,J.,Garcia-Prez,M., Barbosa-Canovas,G.V.,2013b.Developmentofa

mathematicalmodeltodescribeheattransferusingthree differentprocessingmodes.In:IFTAnnualMeeting,Abstract Number:031-138.

Boonpongmaee,T.,Makotani,Y.,2009.Heattransferinrotating cylindricalcellswithpartitions.Int.J.HeatFluidFlow30, 211–217.

Clifcorn,I.E.,Peterson,G.T.,BoydJ.m.O‘Neil,J.H.,1950.Anew principleforagitatinginprocessedcannedfoods.Food Technol.4,450–460.

Erdogdu,F.,Tutar,M.,2012.Acomputationalstudyforaxial rotationeffectsonheattransferinrotatingcanscontaining liquidwater,semi-fluidsystemandheadspace.Int.J.Heat MassTransf.55,3774–3788.

Hirt,C.W.,Nichols,B.D.,1981.Acomputationalmethodfor pressure0surfacehydrodynamics.J.Press.VesselTechnol.– Trans.ASME103,136–141.

Julien,K.,Legg,S.,McWilliams,J.,Werne,J.,1996.Hardturbulence inrotationRayleigh-Benardconvection.Phys.Rev.E53,R5557.

Kooij,G.L.,Botchev,M.A.,Geurts,B.J.,2015.Directnumerical simulationofNusseltnumberscalinginrotating

Rayligh-Benardconvection.Int.J.HeatFluidFlow55,26–33.

Liffman,K.,Metcalfe,G.,Cleary,P.,1997.Convectiondueto horizontalshaking.In:CSIRO-1997:InternationalConference onCFDinMineralandMetalProcessingandPower

Generation,pp.165–168.

Ohlsson,T.,1980.Optimalsterilizationtemperaturesforsensory qualityincylindricalcontainers.J.FoodSci.45,1517–1521.

Pesch,W.,Palaniappan,D.,Tao,J.,Busse,F.H.,2008.Convectionin heatedfluidlayerssubjectedtotime-periodichorizontal accelerations.J.FluidMech.596,313–332.

Reader,G.T.,Hooper,L.C.,1982.StirlingEngines.SponPress, Oxfordshire,UK.

Rosnes,J.T.,Skara,T.,Skipnes,D.,2011.Recentadvancesin minimalheatprocessingoffish:effectsonmicrobialactivity andsafety.FoodBioprocessTechnol.4,833–848.

Singh,A.P.,Singh,A.,Ramaswamy,H.S.,2015a.Modificationofa staticsteamretortforevaluationheattransferunder reciprocationagitationthermalprocessing.J.FoodEng.153, 63–72.

Singh,A.,Singh,A.P.,Ramaswamy,H.S.,2015b.Arefined methodologyforevaluationofheattransfercoefficientsin cannedparticulatefluidsunderrapidheatingconditions.

FoodBioprod.Process.94,169–179.

Singh,A.,Ramaswamy,H.S.,2015a.Effectofproduct-related parametersonheat-transferratestocannedparticulate non-Newtonianfluids(CMC)duringreciprocatingagitation thermalprocessing.J.FoodEng.165,1–12.

585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615

616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636

637 638 639 640 641 642

643 644 645 646 647 648 649 650 651 652

653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709

Pleasecitethisarticleinpressas:Erdogdu,F.,etal., Determiningtheoptimalshakingrateofareciprocalagitationsterilizationsystemfor Singh,A.P.,Ramaswamy,H.S.,2015b.EffectofCanorientationon

heattransfercoefficientsassociatedwithliquidparticulate mixturesduringreciprocationagitationthermalprocessing.

FoodBioprocessTechnol.8,1405–1418.

Singh,A.,Singh,A.P.,Ramaswamy,H.S.,2016.Acontrolled Q4

agitationprocessforimprovingqualityofcannedgreenbeans duringagitationthermalprocessing.J.FoodSci.(inpress).

Singh,A.P.,Ramaswamy,H.S.,2016.Simultaneousoptimization ofheattransferandreciprocationintensityforthermal processingofliquidparticulatemixturesundergoing reciprocatingagitation.Innov.FoodSci.Emerg.Technol.33, 405–415.

Trevino,J.,(masterofsciencethesis)2009.EffectofOscillating andStaticRetortThermalProcessingTechnologyUsingand InstitutionalSizePouch.ClemsonUniversity,Clemson,SC, USA.

Tutar,M.,Erdogdu,F.,2012.Numericalsimulationforheat transferandvelocityfieldcharacteristicsoftwo-phaseflow systemsinaxiallyrotatinghorizontalcans.J.FoodEng.111, 366–385.

Ubbink,O.,Issaa,R.J.,1999.Amethodforcapturingsharpfluid interfacesonarbitrarymeshes.J.Comput.Phys.153,26–50.

Walden,R.,Emanuel,J.,2010.Developmentsinin-container retorttechnology:theZinetecShakaprocess.In:Doona,C.J., Kustin,K.,Feery,F.E.(Eds.),CaseStudiesinNovelFood ProcessingTechnologies:InnovationinProcessing,Packaging andPredictiveModelling.WoodheadPublishingLtd., Cambridge,UK.

Yakhot,H.Q.,Orszag,S.A.,1986.Renormalizationgroupanalysis ofturbulence.I.Basictheory.J.Sci.Comput.1,1–51.

710 711 712 713 714 715 716 717 718 719 720 721 722 723 724

725 726 727 728 729 730 731 732 733 734 735 736 737 738