withdrawn是什么意思hdrawn在线翻译读音

Studyofgrowthmechanismofzinchotdip

galvanisingcoatings

idis,as*

Zinchotdipgalvanisingisoneofthemosteffectivemethodsforthecorrosionprotectionofferrous

r,althoughitsefficiencyisundisputable,theexactgrowthmechanismofthese

resentwork,thediffusivityofliquidzincincarbonsteels

rmore,thechangeofthefreeGibbsenergywas

heless,thecoatingstructurewasmicroscopically

studiedforanextremely送元二使安西 shortimmersiontime(about1s).Inthisway,itwasdeducedthatthe

coatinggrowthbeginswiththenucleationofthefphaseoftheFe–ddphases

aseismechanicallydriftedduetothe

surfacetensionoftheliquidmetal.

Keywords:Zinccoatings,Corrosionprotec高适的《别董大》 tion,Metalsandalloys,Electronmicroscopy

Introduction

Hotdipgalvanisingiswidelyusedforthecorrosion

protectionofironandsteelproductsfortheautomotive,

industrialandhomeapplianceindustry,asitiscon-

sideredthemosteffectivemethodforthiskindof

sonforthisistheirtwofoldpro-

oatingsprotect

thesubstratebyactingasabarrier,whichisolatesthe

substratefromtheaggressiveenvironment,alongwith

thefactthatzincisanodictoironandsteelandasa

resultitbehavesasasacrificialanode.1

Hotdipgalvanisingisaccomplishedthroughthe

immersionofacleanandoxidefreeferroussubstrate

ult,itiscoatedwithazinclayer

withanaveragethicknessofafewtensofmicrometres.

Thiscoatingismetallurgicallybondedtothesurfaceof

thesubstrate,becauseunderthesecircumstanceszinc

interactswithironandanumberofdifferentphasesare

formedbetweenthetwometals,whichintheliterature

arereportedasc,d,fandgphases,1,2startingfromthe

Fe/Zninterface.

However,theexactmechanismfortheseinteractions

isnottotallyunderstood,althoughliteratureprovides

anelevatednumberofrelativestudies.1–13Inanycase,it

iswellestablishedthatthefphaseisthefirstonethat

nucleatesonthesurfaceofthesubstrate.1Nevertheless,

apartfromthisfact,thereareonlyspeculationsregard-

ingthesequenceformationoftherestofthephases.

Furthermore,transitionalandout-of-equilibriumphases

sereasons,thisworkisfocused

onthephenomenathattakeplaceduringtheinitialfew

secondsofthedepositionprocess,alongwiththe

calculationoftheactualdiffusioncoefficientsofthe

Fe–Znsystem,astheyappearduringthegrowthofthe

y,Gibbsfreeenergycalculationsare

usedtoverifytheconclusionsdrawnwiththeprevious

methods.

Experimental

ThesubstratesusedweremadeofcarbonsteelUSt37–1

(C(0?17%,Mn:0?20–0?50%,S(0?05%,P(0?08%-154

VHN)andtheywereeitherorthogonalorcylindrical.

Thecoatinggrowthtookplaceinabathofpurezinc

(industrialgrade)at450uCafterthestandardpretreat-

mentofthesubstrates.1Thedippingtimerangedfrom1

to2400s(1,30,60,180,600,1200,1800and2400s).

Withdrawalofthesamplesfromthemeltwasperformed

withasteadyspeedofabout80cmmin21.

Fortheexaminationofthemicrostructureoftheas-

castspecimens,cross-sectionshavebeencutfromthe

coupons,mountedinbakelite,polisheddownto5mm

aluminaemulsion,etchedina2%Nitalsolution

(1%HNO

3

inCH

3

CH

2

OH)andobservedwithan

OlympusBX60opticalmicroscopeconnectedwitha

cameraCCDJVCTK-C1381anda20kVJEOL840A

scanningelectronmicroscope(SEM)equippedwithan

OXFORDISIS300energydispersivespectrometer

(EDS).TheEDSresultswereusedforthecalculation

ofthediffusioncoefficient.

Results

Structureofgalvanisedcoatingsforveryshort

immersiontime(,5s)

Typicalmicrographsoftheas-depositedcoatingsare

tingofthismicro-

graphwasdepositedwithadippingtimeequalto1sina

stratesufferedthe

usualtreatmentbeforehotdipping(degreasinginan

PhysicsDepartment,AristotleUniversityofThessaloniki,54124

Thessaloniki,Greece

*Correspondingauthor,emailgvourlia@

2009InstituteofMaterials,MineralsandMining

PublishedbyManeyonbehalfoftheInstitute

Received21January2008;accepted12May2008

594SurfaceEngineering2009VOL25NO8DOI10.1179/174329408X326461

aqueoussolutionwith20%NaOH,picklinginan

aqueoussolutionwith16%HClandfluxinginan

aqueoussolutionwith50%ZnCl

2

.2NH

4

Cl).Fromthese

micrographs,itisobviousthatthecoatingthicknessis

veryhigh(,1mm,whichis郭沫若作者简介及代表作 almost10timestheaverage

thickness1).F大风起兮云飞扬安得猛士兮守四方 urthermore,thecoatingstructureis

differentfromthetypicalhotdipgalvanisedstructure.1

Longcolumnarcrystalsareobservedgrownperpendi-

utersurfaceofthe

coating,smallerequiaxedcrystalsaredistinguished

alongwithinclusions(Fig.1b)referringtozincimpu-

rities(especiallyFe–ZnphasesalongwithZnandFe

oxides)asEDSshowed.

Thisstructureimpliesthatthiscoatingisduetofast

coolingoftheliquidzinconthesurfaceofthesubstrate.

Indeed,itisverysimilartothestructureofmetalcastsin

,asthesubstrateisimmersed,the

liquidphaseimmediatelysolidifi

largetemperaturedifferencejustifiesthehighthickness.

Howevertheas-formedsolidphaseisredissolvedinthe

meltasthetemperatureofthesystembecomesuniform.

Consequently,fromtheaboveanalysisitcouldbe

deducedthatthefirststepofthesequencethatleadsto

thegrowthofthecoatingisthefastsolidificationand

dissolutionofliquidzinconthesubstrateduetothe

temperaturedifference.

CalculationofdiffusioncoefficientsatFe–Zn

systemduringhotdipgalvanising

Forthecalculationofthediffusioncoefficients,the

integratedformofthesecondFick’slawwasused

C~C

o

1{erf

x

2Dt1=2

\"#()

(1)

whereDstandsforthediffusioncoefficientandtforthe

tofthesymbolsaredefinedin

calculationsbasedonequation(1),itis

presumedthatthediffusioncoefficientisnotafunction

ofzincconcentrationinthesubstrateandthatdiffusion

stopsasthesubstrateiswit城阙辅三秦的下一句 hdrawnfromthemelt.

Furthermore,forthemeasurementofx,thethicknessof

thegphasewasnottakenintoaccount,asobviouslythis

phaseisnotformedthroughdiffusionbutthroughthe

driftingoftheliquidphaseduetosurfacetensionwhen

y

C

o

51,sincetheliquidispurezinc,asitwasalready

mentioned.

TheresultofthisprocessissummarisedinTable1

eircompar-

ison,itcouldbededucedthattheexperimentalvaluefor

DisveryclosetothevalueforthediffusionofZninthe

fphase,whileDforthediffusionina-Feisalmostten

ordersofmagnitudelower.

Hencethepredominantphenomenonduringthe

growthofthecoatingisnotthediffusionofzincin

iron,

wasalreadymentioned,thefphaseisthefirstonethat

ecanddphasesare

formedwiththediffusionofzincthroughthelayerof

thefphase,whichcouldalsoexplainthefactthatthese

phasesaremuchthinnercomparedtothefphase.1

Consequently,theseobservationsverifythefactthatthe

secondstepinthesequenceofthecoatingformationis

er,theyrevealthat

thethirdstepisthegrowthofthecanddphaseswith

se,thethickness

andtheshapeofthefphase(Fig.3)showsthatits

rast,thisphasecontinues

togrowdentricallyaslongasthesubstrateisimmersed

intheliquidzinc.

Gibbsfreeenergycalculations

FortheGibbsfreeenergycalculations,theexcessGibbs

freeenergywasused,ascanbeseeninequation(2)16博学而笃志 切问而近思

Table1DiffusioncoefficientsforZnatFe–Znsystemat

450uC

Diffusioncoefficient,cm2s21

Experimental1

.

8861026

Znina-Fe141

.

20610216

Zninfphase151

.

4361027

1Opticalmicrographsofcross-sectionsofcoatingsformedwithdippingtimeof1s:aatlowmagnification;bathigher

magnification

2Schematicrepresentationofdiffusionphenomenadur-

inghotdipgalvanising

fgrowthmechanismofzinchotdipgalvanisingcoatings

SurfaceEngineering2009VOL25NO8595

Gm

~

X

i

x

i

G

i

zRT

X

i

x

i

lnx

i

zG

E(2)

whereGmistheGibbsfreeenergyofeachFe–Znphase,

x

i

themolecularfractionsofeachcomponent(Feand

Zn),G

i

theGibbsfreeenergyofeachcomponent,

R58?314Jmol21K21,TthetemperatureandGEthe

excessfreeenergy.

G

i

couldbecalculatedfollowingequation(2)foreach

oneofthetwocomponents

G~H{TS(3)

whereHistheenthalpyofeachcomponentattem-

peratureT,SitsentropyandTthesystemtemperature.

Theexcessfreeenergyforabinarysystemcouldbe

calculatedwiththeRedlich-Kisterpolyonyme16

GE

x

1

x

2

~L

0

zL

1

(x

1

{x

2

)zL

2

(x

1

{x

2

)2(4)

wherex

1

andx

2

arethemolecularfractionsofthetwo

componentsandL

0

,L

1

andL

2

theRedlich-Kister

coefficients.9,13Thesecoefficientsarelinearfunctionsof

temperature,asequation(4)shows

L

k

~A

k

zB

k

|T(5)

wherekistheorderofthecoefficientandAandBare

constants.

Theresultswhichoccurredfromtheapplicationof

theseequationsintheFe–Znsystemaresummarisedin

hphase,twovaluesarepresented

becausethecompositionofeachphaseinthecoatingis

notstable,butvariesbetweenaminimumanda

ase,thefactthatallthevalues

inTable2arenegativeverifiesthefactthateachphaseis

rmore,it

impliesthatthegrowthofthecanddphasesare

lythisisthereason

whythesephasesareformedalthoughtheirgrowth

takesplaceafterthefphasewhichtotallycoversthe

substrate.

Conclusions

Themicroscopicstudyofthestructureofzinchotdip

galvanisingcoatingsforanextremelyshortimmersion

time(about1s)revealedthatthefirststepofthe

sequencethatleadstothegrowthofthegalvanised

coatingsisthefastsolidificationofliquidzinconthe

r

theas-formedsolidphaseisquicklyredissolvedinthe

meltasthetemperatureofthesystembecomesuniform.

Afterwards,thefphaseisnucleatedonthesubstrate,

whilethecanddphasesgrowwithzincdiffusion

aseismechanicallydrifted

esults

werealsoverifiedwithcalculationsofthechangeofthe

freeGibbsenergyforeachphaseofthecoating.

References

:.,2000,45,191–271.

ann:‘Reactionbetweenironandzinc’;1978,London,

ZincDevelopmentAssociation.

:IntMetRev,1979,24,1–19.

i,be:ISIJInt.,1995,35,1388–

1393.

:.,1997,32,5593–

5602.

:.,1997,32,5603–

5610.

k,:.,1997,

358,135–140.

i,,:.

Techn.,1999,122,21–23.

,N.-:.,2001,325,129–

136.

ais,lez:.,2002,

346,211–216.

:.,2003,35,943–947.

,ello,eandJ.-t:.

Sci.,2004,39,5803–5808.

,:Calphad,2005,29,

276–288.

v,va:de,

1973,64,652–654.

gs,n:de,

1979,70,315–317.

,s:‘Introductiontochemicalengineering

thermodynamics’;1987,NewYork,McGrawHill.

3Micrograph(SE曹丕的诗 M)ofcross-sectionsofcoatingformed

withdippingtimeof180s

Table2Gibbsfreeenergychanges(Jmol21)forhotdip

galvanising

PhaseMinimumx

Fe

Maximumx

Fe

c2

d21

f2136421525

fgrowthmechanismofzinchotdipgalvanisingcoatings

596SurfaceEngineering2009VOL25NO8