development of a millimanipulation device to study the removal of soft solid fouling layers from solid substrates and its application to cooked lard deposits

With this device, a horizontal bar 30×6×1mm is movedthrough the materialat a set height from the substrate and the force on the bar mea-sured.Itisacontrolledstrainordeformationtest,where

foodandbioproductsprocessing 93 (2015)256–268

ContentslistsavailableatScienceDirect

jo u r n al ho m e p 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

deposits

aDepartment of Chemical Engineering and Biotechnology, New Museums Site, Pembroke Street, Cambridge CB2

3RA, United Kingdom

bProctor & Gamble Technical Centres Ltd., Whitley Road, Longbenton, Newcastle-upon-Tyne NE12 9TS, UK

a b s t r a c t

Amm-scalescrapingdevicewasdevelopedtostudytheremovalbehaviourofsoftsolidfoulinglayers(thickness

0.5–10mm)fromsolidsubstrates.Abladeisdraggedthroughthecircularorrectangularsamplesatcontrolledspeed

andtheresistanceforcesmeasured.Testswithaviscousliquid(honey)andviscoplasticmaterial(aVaseline®-carbon

blackpaste)indicatedthatcohesivedeformationdominatedthemeasuredforce.Twomodelfoodsoilswere:(i)

unbakedlardand(ii)lardbakedfordifferenttimeswithandwithoutaddedovalbumin.Thecohesivestrengthof

thebakedlard,anditsremovalbehaviour,changednoticeablyfollowingautoxidativepolymerisation.Ovalbumin

delayedtheonsetofpolymerisation

©2014TheAuthors.PublishedbyElsevierB.V.onbehalfofTheInstitutionofChemicalEngineers.Thisisanopen

accessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/3.0/)

Keywords: Adhesion;Cohesivestrength;Cleaning;Fouling;Fats;Rheology

1 Introduction

The formation of fouling layers from baked fats and oils

isawidespreadprobleminfoodprocessing,causing

equip-mentdegradation,unhygienicconditionsand,onhousehold

appliances, an undesirable appearance Food fats and oils

aremixturesofmono,di-andtri-glyceridesaswellasother

hydrophobiccomponents.Duringfryingandbaking,proteins

andstarchescanbeaddedtothemelt.Foulingdepositsare

formedbyoxidativepolymerisationoftheunsaturated

com-ponentstogivesolidorsemi-solidlayers,whichcanadhere

stronglytotheprocesssurfaces.Extended(orrepeated)heat

treatmentpromotesfurtherpolymerisation,colourchanges

andultimately, athigher temperatures,degradation to

car-bonaceous(coke)layers

∗ Corresponding author.Tel.:+447833466898

E-mailaddress:aa620@cam.ac.uk(A.Ali)

Received2May2014;Receivedinrevisedform31August2014;Accepted2September2014

Availableonline22October2014

Theselayersareamongstthemostchallengingtoremove and cleaning often requires strong chemical treatments and/or large mechanical forces Both of these can lead to degradation ofthe underlyingsubstrate andanincreasein surface roughness (i.e scratches), which can increase the propensityforfurtherfoulingandmicrobialgrowth.Cleaning baked-onfatsoilsofteninvolvesacombinationofchemical and mechanical actions, wherein a chemical reagent pro-motes softeningofthe soillayersothatit canberemoved

byfluidshear, impactingliquid jetsormechanical friction Beingabletomeasuretheforces involved,andthereby the rheologyandevolutionofmicrostructure,ofthesematerials duringcleaningishighlydesirableforunderstandingcleaning mechanismsand developingcleaningagents.Asstiff semi-solidmaterials,thesesoilsdonotlendthemselvestostudyby

http://dx.doi.org/10.1016/j.fbp.2014.09.001

0960-3085/©2014TheAuthors.PublishedbyElsevierB.V.onbehalfofTheInstitutionofChemicalEngineers.Thisisanopenaccessarticle undertheCCBYlicense(http://creativecommons.org/licenses/by/3.0/)

Roman

a Carreau–Yasudomodelparameter,–

fI forcecomponenttodisplacematerialaheadof

theblade,N

fII forcecomponenttoraisedisplacedmaterial,N

fIII forcecomponenttoresistmaterialunderneath

blade,N

Fw forceperunitwidth,Nm−1

Fw,0 forceperunitwidthatr/R=0,Nm−1

F0

w plasticcomponentofFw,Nm−1

FI

w forceperunitwidthtodisplacematerialahead

oftheblade,Nm−1

FII

w forceperunitwidthtoraisedisplacedmaterial,

Nm−1

FIII

w forceperunitwidthtoresistmaterialunderthe

blade,Nm−1

Fw averageforceperunitwidth,Nm−1

g accelerationduetogravity,m−2

G elasticmodulus,Pa

G viscousmodulus,Pa

m Carreau–Yasudopower-lawindex,–

r distancefromsamplecenterand

millimanipu-lationtool,m

R2

fit coefficientofdetermination,–

t o scrapingtime,s

V scrapespeed,ms−1

Vdisplace samplevelocityatthesubstrate,ms−1

Vsubstrate samplevelocityatthesubstrate,ms−1

w widthofthebladeincontactwithsample,m

wt widthofthemillimanipulationblade,m

x distancescrapedthroughsamplerelativetothe

pointoffirstcontact,m

Greek

ı clearanceunderneathmillimanipulationblade,

m

ıo initialthicknessofsoil,m

˙

 shearrate,s−1

 apparentviscosity,Pas

0 viscosityat0shearrate,Pas

∞ viscosityatinfiniteshearrate,Pas

 characteristicviscoelastictime,s

w wallshearstress,Pa

Acronyms

BHT butylatedhydroxytoluene

DSC differentialscanningcalorimetry

FT-IR Fouriertransforminfraredspectroscopy

GPC gelpermeationchromatography

SS stainlesssteel

Vas Vaseline®

VCB Vaseline®blackpaste

existinghydrodynamicdevicessuchastheparallelplateflow cell(Bakkeretal.,2003),impingingjet(Bayoudhetal.,2005), radialflowcell(Detryetal.,2007),rotatingdisc(Garcaetal.,

1997),theplynometer(Zoritaetal.,2010)andfluiddynamic gauging (Chew et al., 2004) These all require the user to observethemomentandnatureofremovalatalengthscale

ofseveralmicronsandabove

Atsmallerlengthscales,atomicforcemicroscopyhasbeen usedtocharacteriseadhesionandcohesionoffoodstuffssuch

ascaramel, sweetenedcondensedmilkand Turkishdelight (Akhtaretal.,2010)andbiofilms(Garrettetal.,2008).Ofthe techniquesmentioned,however,noneprovideadirect mea-sureoftheforceswithinasoilorthestrengthofattachment

tothefouledsurface.Withtheexceptionoftheplynometer, theyallgenerallyrequirelargeliquidvolumesand/oraretime intensive.Wepresentthecommissioningofadevicethatis easytouse,requireslittleornocleaningliquidandcanmake measurementsquickly(≤5min)

1.1 From micro- to milli-manipulation

Zhang and co-workers developed the micromanipulation techniqueattheUniversityofBirminghamforstudyingthe deformationofcells(Zhangetal.,1991)andtheirmechanical properties (Zhang et al., 1992; Yap et al., 2006) A modi-fied‘micromanipulation’devicewasdevelopedbyZhangand Fryer (Liu et al., 2002) tostudy the adhesive and cohesive forces withinsoiling layers With this device, a horizontal bar (30×6×1mm) is movedthrough the materialat a set height from the substrate and the force on the bar mea-sured.Itisacontrolledstrain(ordeformation)test,whereas the hydrodynamic devices mentioned above employ con-trolled (or estimated) shear stress conditions By adjusting the level at which the tool is pulled through the layer, the technique can investigate cohesive (soil–soil) or adhe-sive(soil–substrate)interactions.Table1summarisesstudies wheremicromanipulationhasbeenusedtostudyfoodstuffs, including proteinaceous(Liuet al.,2006a)andstarchlayers (Liuetal.,2006b).Thesestudieshaveshownhowsample rhe-ologyandsoil–substrateinterfacialpropertiesaffectboththe strengthofattachmentandthetypeofremovalobserved Theheterogeneityandthehighcohesivestrengthoffood fatsoilspromptedourdevelopmentofa‘millimanipulation’ tooldesignedtomeasurelargerforcesand workwithdeep layers,ofthickness0.2–10mm.Bakedfatlayers,forexample, havegreaterstrengthsthanthematerialreportedinTable1 Thestrengthoflayersintheaforementionedstudiesranges from∼0.3to80Jm−2(asdefinedbelow).The millimanipula-tiondevicepresentedhere,isshowntoworkonmaterialswith

awider rangeofstrength,includinghoney (∼0.4Jm−2)and bakedlard(∼420Jm−2)

TheschematicinFig.1(a)illustratestheworkingaction:

averticalstainlesssteelbladeismovedhorizontallythrough thelayeratvelocityV.Thelayerisscrapedfrominitialdepth,

ıo,tofinaldepth,ı.Theforcerequiredtoimposethestrain

ismeasured and videomicroscopes are usedtorecord the deformationbehaviour.Theforcemeasured,f,includes com-ponentsrequiredto(seeFig.1

(I) Deformmaterialinthelayerofthicknesss,aheadofthe blade,fI;

(II) Displacethedeformedmaterial, usuallyupwardsalong thefaceoftheblade,f ;

(mm)

Apparent cohesivestrength (Jm−2)

Apparent adhesivestrength (Jm−2)

Experimental variables investigated

hydration

Liuetal.(2002)

temperature

Increasingtemperatureand NaOHconcentration (0.1–5wt%)

Liuetal.(2006a)

exposuretoair

Liuetal.(2006b)

Tomatopaste Ni-P-PTFE

composites

substratesurface energy,sample thickness

Minimumsampleadhesion occursonsurfaceswith surfaceenergiesfrom20to

25mJm−2

Liuetal.(2006b)

substrate

Alayerofcornoilbetween substrateanddeposit reducesadhesion

Liuetal.(2006)

temperature

Higherbakingtemperatures increaseadhesivestrength

Liuetal.(2007)

o,

FI

FII

FIII

fcanbecomparedwiththeadhesivework.Itisnotedthat

settingthebladealongthesurface,i.e.ı=0,islikelytoinvite

contributionsfromsurfacefriction.Forthesoilsstudiedhere,

thecohesivestrength,fIisnon-zero.TheschematicinFig.1(a)

pertainstocohesiveremovalofaviscousliquidorsemi-solid

soil.Fig.1(b)showsanexampleofaVaseline®carbonblack

(VCB)pastebilletundergoingtesting.Iftheremovedsoildoes

notadheretotheblade,fIIisnegligible.Theforcetransducer

onlymeasuresforcesinthehorizontalplane;anyshearcaused

bymaterialclimbingupthebladeisnotmeasureddirectlybut

theworkrequiredtoelevatesuchmaterialisreflectedinFw

2 Materials and methods

2.1 Device

Themillimanipulationdevicewasdesignedtoworkwithsoil

samplespreparedoncircularmetaldiscsofdiameter50mm

Theunitwasbuiltaroundaforcetransducer(SauterGmbH,

FH5) witha measurementlimit of5000mN and resolution

±1mN.Thetransducerismountedonasteelx–ytabledriven

bytwosteppermotorsallowingindependentmotioninthe

xandydirectionsatspeedsrangingfrom0.5to20mms−1

Theforceismeasuredatasamplingrateof20Hz.Thestepper

motorsare controlledviaadesktopPCanddataarelogged

usingaLabVIEWTMapplication.Thedeformationisrecorded

byavideomicroscope(400XUSBMicroscope,MaplinGadgets)

Twosampleshapeswereusedforthestudiesreportedhere

Softersamples,suchashoney,Vaseline®andlard,were

pre-paredoncirculardiscs.Thetoothwidth,wt,wasselectedto

bewiderthanthesample,suchthatthelengthincontactwith

theblade,w,varieswithscrapedistance,x(seeFig.2(a)).For

harderlayers,suchasbakedlard,anarrowrectangular

sec-tionisprepared,suchthatwstaysconstantduringscraping

(seeFig.2(b)).Thetoolisheldverticalbylubricatedguiderails

(seeFig.2(c)and(d))connectedtotheforcetransducerand

ispulledhorizontallythroughthesampleatconstantspeed,

typicallyfor38mm

2.2 Sample preparation

Commissioning tests were conducted using honey

(Sains-bury’sBasicTM,aviscousliquid),Vaseline®(aviscoplasticsoft

solid)andlard(amodelfoodsoil).Honeysampleswere

pre-paredby pipetting 14g of materialon to 50mm diameter

(1mmthick)stainlesssteel(SS)316testdiscsatroom

tem-perature.Thisprocedurewasfoundtogivereproduciblelayers

approximately3.5mmthick,measuredusingthevideo

micro-scope

Vaseline® and lard samples were spread evenly across

50mmdiameterdiscswithaspatula,usingsetsofhalf-ring

formersto providesamplesofknown thickness(≤0.9mm)

AsVaseline®istranslucent,samplesweremixedwith5wt%

graphitepowder (FisherScientific,UK) togive aVCB paste

suitableforobservationwiththevideomicroscope

Model food fat layers were prepared using lard

(Sains-bury’sBasicTM),bakedfor1–5hattemperaturesof50–250◦C

Themeltingpointrangeofthelard,asmeasuredby

differ-entialscanning calorimetry(DSC,datanotpresentedhere),

was27–38◦C.Lardwasheatedto39◦Candeggwhiteprotein

(ovalbumin:Sigma–Aldrich,gradeII)wasaddedtogivea

liq-uidmixturewithproteincontentsrangingfrom0to9.5wt%

AliquotsofthemixturewerepipettedontothecentreofaSS

discandallowedtospreadout.Thediscswereplacedinan ovenforbakinginairattherequiredtemperatureand dura-tion.Sampleswereallowedtocooltoroomtemperaturebefore testing

This method of preparation was suitable for preparing layersupto200␮mthick.Someofthesethinlayerswere, how-ever,sohardthattheforcesrequiredtoremovethemexceeded theinstrumentlimit,orcausedthebladetodeflect.Thicker andsomewhatsoftersampleswere thenpreparedusingSS discswith0.6mmhighraisededges.Thelardwasloadedon

tothediscandcookedasabove.Rectangularsamples,12mm wideandapproximately40mmlong,werecreatedwhenthe samplehadcooledandbyclearingtheoutermaterialaway usingastencilandspatula(seeFig.2(c)featureD).Thetooth toolwasusedinthesetests.Theraisededgerestrictedthe blademovementto38mm

TheSSdiscsurfaceswerecleanedaftertestingby immer-sionin1Msodiumhydroxidesolutionovernight,sonicationin reverseosmosiswaterat40◦Cfor30min,followedbydrying

inairatambientconditions

Attemptstocharacterisethechemicalnatureofbakedlard weremadeusingFouriertransforminfraredspectroscopy (FT-IR)andgelpermeationchromatography(GPC).FT-IRspectra didnotshowanymarkeddifferencebetweenlardandbaked lard(datanotpresented).GPCtestingprovedinconclusive,as baked larddidnotdissolve inany ofthesolventsavailable fortesting (including acetone, tetrahydrofuran,chloroform, chlorobenzeneandtoluene)

2.3 Material rheology

Rheological measurements were conducted using a Bohlin CVO120HRcontrolledstressrheometer(MalvernInstruments, London,UK)usingparallelplateswhichwereeither(i)smooth,

20mmdiameter,or(ii)sandblasted,25mmdiameter,to pre-ventwallslip.Oscillatorytests wereconductedat20◦Cfor Vaseline®,VCBpasteandlard.Frequencysweepswere per-formedfrom0.01to50Hzwithastrainamplitude<1%

3 Results and discussion

3.1 Material rheology

3.1.1 Honey

Fig 3 shows that the honey exhibited noticeable shear thinning, with a low shear rate behaviour that is consid-eredpseudo-Newtonian.Thenon-Newtonianbehaviourwas attributedtothepresenceofcrystallisedsugarparticlesinthe honey.ThiscanbedescribedbytheCarreau–Yasudamodel:

−∞

0−∞ =(1+()˙a)(m−1)/a (4) where0 and ∞ arethe lowandhigh shear-rateviscosity,

respectively,˙istheshearrate,isacharacteristictime con-stantandparametersaandmdescribethebehaviourbetween thelowandhigh-shearrateplateaus.Forhoney, regression

of the experimental data gave 0=13.4Pas, ∞=4Pas,

=0.0025s,m=0,a=1

3.1.2 Vaseline®and VCB

The steady shear results in Fig 4(a) show that both the Vaseline® and VCBpastesexhibitalowshear, high viscos-ityplateau.Aboveacriticalshearrate,ofaround5×10−4s−1,

260 foodandbioproductsprocessing 93 (2015) 256–268

Fig 2 – (a) Schematic of circular samples, with w = 2

R 2 − ( R − x) 2

The shaded area denotes the location of the sample.xis the distance scraped through the layer.ris the minimum displacement between the blade and disc centre.rvaries between

−RandR,such thatr/Ralways lies between −1 and 1 (b) Schematic of rectangular samples The width of the blade, w t , is larger than the sample For the rectangular samples, w =/f(x) (c) Photograph of the millimanipulation rig showing: (A) motorisedx–ytable; (B) force transducer; (C) scraper blade held in place with guide rails; (D) adjustable height mount, with baked lard strip; (E) video microscope (d) Millimanipulation tooth in (b); dimensions in mm.

(shear stress of ∼50Pa) the apparent viscosity decreases

stronglywith increasing shearrate Thevalues ofthe low

shearapparentviscosity,ofaround105Pa,aresimilartothose

reported foradifferent Vaseline® materialbyChang et al

(2003),aswellassomeothercommonfoodspreadsthatare

noteasilycleanedbywaterrinsing(see(Yangetal.,2014))

Whenthedataareplottedagainstshearstress,asinFig.4(b),

theysuggestthatthematerialexhibitsacriticalshearstress

ofaround50Pa,abovewhichthematerialisshearthinning

Theeffectofslip,givinglowerapparentviscosity,isnoticeable

withsmoothtools

TheoscillatorysheardatafortheVaseline®basedmaterial

inFig.5showdifferencesintheirsmallstrainbehaviour,i.e

beforeyielding.Theviscousmodulus,G,fortheVaseline®is

consistentlyhigherthantheelasticmodulus,G,untilhigher

frequencies,indicatingliquid-likebehaviour,whereastheVCB

modulivaluesaremoresimilarinmagnitude,indicatingthat

agel-likestructureiscreatedbytheadditionofcarbonblack

particles

3.1.3 Lard

Thelard proved tobe too stiff for steady shear testing at

20◦C The oscillatory shear data in Fig 6, however, show

thatthelardwaspredominantlyelasticatthistemperature, withG∼15kPa,overthefrequencyrangetested.Theviscous modulus was ∼10kPa, so someviscoelasticcontribution is expected

3.2 Millimanipulation data

3.2.1 Honey

Tests withhoney employeda 1mm sheargap The forces required to deform the honey were small so these tests employedcircularshapesanda60mmwideblade.Fig.7(a) shows measuredforceversus scrapeddistanceforahoney layerofinitialthickness3.5mmsubjecttoascrapedepthof 2.5mm.Thelengthofthebladeincontactwiththefilm,w, variedwithpositionandthevariationisplottedalongsidethe measuredforce Theforceinitiallyincreasesaswincreases anddecreasesafterxreachedthepointofmaximumwidth (at x=R=31.75mm).Separatetestsestablishedthat compli-anceinthefittingsgaveanuncertaintyofaround±2mNin theforcemeasurements

The force data are plotted as normalised force versus dimensionlessscrapingdistance,r/R,inFig.7(b).Thereisan

Fig 3 – Apparent viscosity of honey measured at 20 ◦ C.

1 mm gap, 20 mm radius smooth plates Solid symbols

indicate increasing shear rate, open symbols return ramp

(decreasing shear rate) Dashed line shows fit of data to

Carreau–Yasuda model, Eq (4) , with parameters

 0 = 13.4 Pa s,  ∞ = 4 Pa s,  = 0.0025 s,m= 0,a= 1, R 2

fit = 0 986.

initialstepinFwoverthefirstfewmmoftravel,associated

withthesheargapbeingfilled,andasteadyincrease with

x until the disc midpoint is reachedthen decreasing as x

increases.TheasymmetryevidentinFig.7(a)isagainevident

andFwincreasessharplyasxapproachesandpassesx=2Ras

wapproacheszero.Theforceprofileisnotsymmetricalabout

x=R,andthereisanon-zeroforcewhenthebladereachesthe

edgeofthedisc(x=63.5mm)

Thelinearshearrateinthegapwasestimatedusing:

˙

=V−Vsubstrate

where Vsubstrate is the velocity at the substrate A no-slip

boundaryconditionwasassumed,giving˙=V/ı.Thelargest

shear rate expected under the blade for the velocities in

Fig 5 – Oscillatory testing of Vaseline®(Vas) and Vaseline®-black paste (VCB) at 20 ◦ C 25 mm diameter, parallel, roughened plates, 1% strain amplitude, 1.5 mm gap.

Fig.7(b),1–4mms−1,isthen4s−1.Atthisshearrate,Fig.3 indi-catesthatthehoneyisNewtonianwithaviscosityof∼14Pas Thecontributionofshear,processIII,canbeestimatedby treatingthegapunderthebladeasaone-dimensionalslit,so that:

FIII

wherewisthewallshearstressandL,thebladethickness,is

2mm.FortheconditionsinFig.7(b),theapparentshearstress ratesare1–4s−1andFIII

w=25–100mNm−1.Thesevaluesareall smallerthantheinitialFw(measuredatr/R∼−1),indicating thatFwisdominatedbydeformationanddisplacement TheFwprofilesinFig.7(b)showaproportionaldependency

on V, whichis expectedfor aNewtonian fluidundergoing

Fig 4 – Plots of apparent viscosity versus (a) shear rate and (b) shear stress for Vaseline®(labelled ‘Vas’) and

Vaseline®-carbon black (labelled ‘VCB’) paste at 20 ◦ C.

(b) is reproduced from Yang et al (2014) , with permission.

262 foodandbioproductsprocessing 93 (2015) 256–268

Fig 6 – Oscillatory testing of lard at 20 ◦ C Rough parallel

plates, 1% strain, 1 mm gap.

shearand deformationinthe laminarregion.Inspectionof

theFwatr/R=0,Fw,0,showedalinearcorrelationwithV,viz

Fw,0(r/R=0)=133V+0.15 (R2

whereR2

fitiscoefficientofdetermination

Videoimagesindicatedthatthedislodgedmaterial

accu-mulatedatthefront oftheblade(asshown inFig.1 and

beyondthemid-point,i.e.r/R>0,somedrippedofftheedge

ofthedisc.Thecontributionfrom displacement(processII

inFig.1)withcircular samplesisthereforecomplexand is

believedtoberesponsibleforthechangeinFwwithx.Tests

onbakedlardweresubsequentlyperformedwithrectangular

stripsofmaterialsothattheareaofsampledisplacedbythe

bladeisalmostconstant

3.2.2 VCB paste

TheresultsforremovingVCBpastelayers,atascrapedepth

of1mm,inFig.8(a)showtheincreaseinfwithscrapewidth observedwithhoneyandasimilarasymmetryintheprofiles Theforcesarenoticeably larger,reaching250mNcompared

tothe18mNobservedwithhoneyforscrapedepth,s=1mm,

V=1mms−1 The measured force is insensitiveto the gap height,ı, whichvariesfrom 2to8mminthesetests, indi-catingthatFIII

wisnegligible(estimatedas2mNm−1usingEq (6))andthatthedeviceismeasuringthecohesivestrengthof thematerial

ThenormalisedprofileinFig.8(b)shows3regionspresent

InregionA,thereisasteepinitialriseinFwfor−1≤r/R−0.85,

toabout2Nm−1.Theinsensitivitytoıisagainevident.The initialriseinAisattributedtosomeinitialcreepand compli-anceinthematerial

InregionB,thegradualincreaseinFwby1–2Nm−1with displacementislikelyduetoaccumulationofdislodged mate-rialinfrontoftheblade.Thecontributionofhydrostaticstress

toFw,arisingfromraisingthemassofdislodgedmaterial,was estimatedbygVdisplaceas∼5×10−5N(whereVdisplace isthe

volumeofdisplacedmaterial).ProcessII,wasnottherefore consideredtoaffectmillimanipulationdata.InregionC,Fw

risesmarkedly,aswapproacheszero.Thiswasalsoobserved withhoney(seeFig.7(b))andthesedataarediscarded Testson12mmwiderectangularVCBsamplesareshown

inFig.9.TheforcevaluesinFig.9(a)arelowerthanthedisc tests(Fig.8 asthewidthofthetoothisgenerallysmallerthan thechordincontactwiththesampleinthedisctests.The sim-plerdeformationzoneaheadoftheblademaybethereason whythereisnogradualriseinFw.RegionCinFig.8(b)isnot evident.Aswiththedisctests,initialsamplethicknessdoes notaffecttheresultsandthemagnitudesofFw(∼2–3Nm−1) aresimilar

Fig.10showstheeffectofscrapedepth,wherelayersof differentinitialthicknesswerescrapedtoacommonresidual thickness,of5mm.Theapparentshearrateundertheblade

inthesecasesis0.2s−1andtheFIII

w contributionsestimated fromEq.(6)isaround0.4Nm−1.Theindividualprofilesdonot

Fig 7 – Millimanipulation of honey on 63.5 mm diameter circular SS discs at 20 ◦ C ıo= 3.5 mm,s= 2.5 mm,R= 31.75 mm (a) Force,f,and blade contact length, w, forV= 1 mm s −1 Error bars on selected points show measurement uncertainty Gaps between data points are due to the limits of the force transducer resolution (b) Effect of velocity onFw , with normalised scrape force versus distance The horizontal dashed lines indicate estimated values of F III from Eq (6)

Fig 8 – Removal of VCB paste from 50 mm diameter discs.V= 1 mm s −1 ,s= 1 mm,R= 25 mm, 20 ◦ C (a) raw data, including blade contact length, w, displayed with the solid line (b) normalised force versus normalised radial distance Legend entries indicate initial sample thickness ıo.

exhibitthetrendsinFig.8,exceptfors=1mm.Fors=2mm

and4mm,thereisariseinFwuntilr/R∼−0.5afterwhichpoint

thereisagradualdecreaseinFw.InadditiontheFwvaluesin

regionBdonotincreaseproportionatelywiths,whichwould

beexpectedifFI

wandFII

wwererelatedtothevolumeofmaterial deformedanddisplaced.ThedeformationoftheVCBmaterial

iscomplexandisthesubjectofongoinginvestigation

3.2.3 Lard

Theprofiles obtained forscraping circular lard samples in

Fig.11(a)showsimilarfeaturestotheVCBpasteinFig.8(b)

Sampleswerealsopreparedasrectangularstrips.Fwvalues

fortherectangularsectionsinFig.11(b)were,ingeneral,lower

thatonthecirculardisc.Thereasonforthedifferencesisnot

currentlyunderstood.Thereisalsomorevariationbetween

samplesonthediscs.Thetrendsobservedremainthesame

betweentestingontherectangularsectionanddiscs;in

par-ticular,scrapespeeddidnotaffectFw

The measured forces for lard are about 3 times larger than the VCBpaste.There is,however, nosimple relation-shipbetweenfandtherheometrydata.Thescrapingspeed has no effect on Fw, indicating that FIII

w is small For soft solids that fail in acohesive manner, a flat surface isleft behind (see schematicin Fig.1(a)) Therefore, byreturning thebladeoverthescrapedsample,separatemeasurements

ofFIII

w canbemade.Forlard,thesemeasurementsconfirmed that FIII

w<0.3mN Theinsensitivity to V indicates that the deformation isprimarilydue toelasticand plastic compo-nents

3.2.4 Baked lard

The behaviour observed for baked lard different markedly fromlard.TheresultsforbakedlardreportedinFig.12were obtained with 12mm wide strip samples tested using the toothtool(Fig.2(b)–(d)).Fig.12presentsatypicalplotofFw

versustime,ratherthandisplacement,obtainedforasample

Fig 9 – Removal of VCB paste from 12 mm wide rectangular sections.V= 1 mm s −1 ,s= 1 mm, 20 ◦ C (a) Scrape profiles (b) Normalised force versusx.Legend entries indicate initial sample thickness ı

264 foodandbioproductsprocessing 93 (2015) 256–268

Fig 10 – Effect of scrape depth,s,onFw for VCB paste on

circular discs 20 ◦ C,V= 1 mm s −1 , ı = 5 mm,R= 25 mm, ıo

varies (=5 mm +s).

ofprotein-free lard baked at 250◦C for 4h and scraped at

V=2.0mms−1.Theblademovedthroughthesample,leaving

a0.2mmresidualfilm,untilx=38mm,atwhichpointithalted

(att=19s).Asbefore,threeregionsareevident.Intheinitial

region,markedA,thereisasteep,initialrise.Thepointat

whichFwbeginstoriseissetastheorigin.Thisbehaviouris

attributedtocomplianceand initialcreep(elasticity)inthe

material.Theexistenceofcreepwasconfirmedbyhaltingthe

motionofthebladeduringatraverse.Theforcedecayedtoa

smallvalue,F0

w,inanexponentialmanner.Videosconfirmed

thatthisdecaywasnotduetomovementofthesoil,

indicat-ingvisco-elasticbehaviourandconsistentwiththematerial

developingstrongerandnewcohesiveinteractionsasaresults

ofoxidativepolymerisation(WilsonandWatkinson,1996)

InregionB,Fwfluctuatesaroundanaveragevalue,denoted

Fw This is interpreted as arising from steady shear and

Fig 12 – Plot ofFw and displacement,x,versus time for lard baked for 4 h at 250 ◦ C Test conditions:T= 20 ◦ C,

V= 2 mm s −1 , ıo= 0.6 mm,s= 0.4 mm F 0

w marked on as the residual force remaining after scraping.

yieldingwithinthematerial,andthemagnitudeofFwistaken heretorepresentthecohesivestrengthofthelayer.The fluctu-ationsariseduetobrittlefractureandductiledeformationof thissample(i.e.crackingandcreationofnewsurfaces) Sam-plescookedatlowertemperaturesandshortertimestended

toexhibitductileyieldingandgavemoreuniformFwprofiles,

aswellasevenresiduallayers

InregionC,thebladeisstationary,neartheendofthe sam-ple.ThebladeisstillincontactwiththelayerandFwdecays withtime.Thiswasalsoobservedinthecreeptestsreferred

toabove.ItisnoticeablethatFwdoesnotdecaytozero,but approachesaresidualvalueofaround37Nm−1,suggestinga plasticcomponenttothedeformation

ThetimedependentdecayinFwwasinvestigatedby halt-ing a traverse before approaching the end of the sample,

as shown in Fig 13(a) Fw had reached a steady value of

∼115Nm−1 before motion was halted at ∼2.5s The force decaystoameasurableresidualF0

w,ofaround20Nm−1.The

Fig 11 – Effect ofVonF for (a) circular(R= 25 mm) and (b) rectangular strip lard samples 20 ◦ C, ı = 6.7 mm,s= 1 mm.

Fig 13 – Decay in measured force after scraping is halted at timet0 Strip sample of lard baked for 4 h at 250 ◦ C The blade is moved 5 mm into a 40 mm long sample at 2 mm s −1 then stopped att0 = 2.5 s 20 ◦ C, ıo= 0.6 mm,s= 0.4 mm (a)Fw −tprofile, with residual force F o

w (b) Data in (a) plotted in the form of exponential decay.

transientdataareplottedintheformofanexponentialdecay

modelinFig.13(b)andexhibitaroughlylineartrendon

log-linear axes This suggests that the stress relaxation could

beapproximatedbyasingle-elementMaxwellmodel,but a

multi-modemodelmaybemoreappropriate

3.2.4.1 Types of removal. Thereare three typesofremoval

observedfortheselayers.Ifthelayersfailadhesively,aclean

substrateisleftbehindafterscraping(seeFig.14(a)).Adhesive

strengthwouldbeassessedbyeither(i)scrapingoffcloseto

thediscsurface(i.e.settingı∼0)or(ii)iftheadhesivestrength

issufficientlylowerthanthecohesivestrengthsuchthatthe

soil–substratebondyieldsbeforesoil–soilbondsdo.If they

failcohesively,aresiduallayerwillremainafterscraping.For

theharder,bakedlardlayers,thiswillleaveajagged,uneven

wake,ofvaryingthickness(seeFig.14(b)).Ifthelayerfailsboth

adhesively and cohesively, a mixture of the patterns in Fig.14(a)and(b)isobserved

3.2.4.2 Effect of testing conditions. Fig.15showstheeffectof scrapevelocityontheremovalbehaviourofpolymerisedlard layerspreparedbybakingat250◦Cfor4h.Eachdatum repre-sentsaseparateexperimentandthereisnoticeablevariation betweentests,whichispartlyduetotheextentof polymeri-sationnotbeingcontrollablewiththismaterial.Thevariation betweensamplesissignificantlygreaterthanthefluctuation

inFwforindividualsamples,markedbytheerrorbars There is noticeable scatter at the lowest scrape speed

(V=0.5mms−1)whichisaccompaniedbyamixtureof cohe-sive and adhesive failure At higher speeds removal is dominatedbythelattermode Theweakdependencyon V

confirms that the materials exhibit plastic behaviour The

Fig 14 – Photographs of wake region after millimanipulation scraping of lard baked at 250 ◦ C.V= 2 mm s −1 , ıo= 0.6 mm,

s= 0.4 mm Arrow indicates direction of scraping Dashed lines indicate regions where the sample was scraped away (a) Adhesive failure and (b) cohesive failure.