www.mejournal.

orgJournalofMetallurgicalEngineering(ME)Volume2Issue3,July2013

PhysicalandNumericalSimulationsStudyof
HeatInputDependenceofHAZCrackingin
NickelBaseSuperalloyIN718
L.O.Osoba1,Z.Gao2andO.A.Ojo2
DepartmentofMetallurgicalandMaterialsEngineering,UniversityofLagos,Nigeria
DepartmentofMechanicalandManufacturingEngineering,UniversityofManitoba,Winnipeg,Manitoba,Canada
R3T5V6
losoba@unilag.edu.ng;2gao2@cc.umanitoba.ca;2ojo@cc.umanitoba.ca

Abstract
Physical and numerical simulations study of heat input
dependenceofcrackingintheweldheataffectedzone(HAZ)
inanickelbasesuperalloy, IN 718, was performed. Gleeble
thermomechanical physical simulation showed that
increase in resistance to cracking with heat input is
associated with enhanced hot ductility of the alloy.
Numericalcalculations by finiteelement analysis show that
in addition to the improved hot ductility, reduction in the
HAZ cracking with increase in heat input is also attributed
to a decrease in the gradient of the thermally generated
strainrateduringweldcooling.
Keywords
Superalloy;HAZ;Ductility;Welding

Introduction
Alloy IN 718 is the most widely used precipitation
strengthening nickel base superalloy for the
fabrication of aero and landbased turbine engine
components [McCay 1991, Radavich 2005] due to its
excellent oxidation and hot corrosion resistance and
high strength at elevated temperature up to 650oC
[Shonozaki1999,Luo2002,Kang2001].Inaddition,IN
718possesrelativelygoodresistancetopostweldheat
treatment (PWHT) cracking because of the sluggish
precipitation kinetics of its primary strengthening
phase Ni3Nb, [Lingenfelter 1989, Boech 1969]. The
everincreasingdemandforefficiencyoftheaeroand
landbased turbine engines components necessitates
that superalloys are fabricated in intricate shape and
design either by casting, forging or by joining. In
addition, owing to their high cost, it is economically
more viable to repair damage or servicedegraded
components than to replace them with new parts.
Traditionally, during fabrication of new parts and/or
repair of servicedamaged components of turbine

88

engines, fusion welding process has proved to be an

economical and reliable means of joining. A recent
trend in the aerospace, automotive and power
generationindustriesisthedrivetoutilizehighpower
beamdensitysourcestorapidlyproducedeep,narrow
and low heat input welds during components
fabrication and repairs. Laser beam welding (solid
state, fibre or gas) is an attractive and promising
joining technique with respect to the achievement of
these goals with high reliability and productivity
necessary for industrial manufacturing system [Sehar
andReed2002].However,weldabilityofalloyIN718
can be undesirably limited by the occurrence of
cracking in the HAZ during welding [Vincent 1980,
Thompson1998,Krenz2009,Thompson1969]andthe
applicability of laser welding to IN 718 alloy has not
been fully studied and understood. In general, the
cause of cracking in the HAZ of welded joint of
superalloys has been attributed to both metallurgical
(grain boundary liquation) and mechanical factors
(magnitude of stresses and strains generated during
weldingthermalcycle).Previousstudieshaveshowed
that the occurrence of cracking in the HAZ can be
significantly influenced by the magnitude of heat
inputduringwelding[Richard1994,Idowu2007].The
lower the magnitude of heat input the higher is the
susceptibilitytoweldHAZcracking[Krenz2009].The
main objective of the present work is to perform
physical and numerical simulations study to better
understand heat input dependence of weld HAZ
crackinginIN718superalloy.Theresultsofthestudy
arepresentedanddiscussedinthisarticle.
Materials & Experimental
The chemical composition of wrought IN 718
superalloy material used in this work in weight % is

JournalofMetallurgicalEngineering(ME)Volume2Issue3,July2013www.mejournal.org

0.5Al, 0.9Ti, 5.4Nb, 17.9Cr, 2.9Mo, 17.6Fe, 0.05Mn,

0.05Si, 0.026C, 0.007P, 0.0014Mg, more than 0.007S,
0.04B, and balance nickel. Physical simulation test
samplesweremachinedfromtheasreceivedmaterial
by a numerically controlled wire electrodischarge
machine (EDM) and subjected to preweld solution
heattreatment(SHT)at1050oCfor1hour,followedby
aircooling. The heat treated samples were surface
groundandproperlycleanedtoremovesurfaceoxides
beforeuse.Inordertocarefullystudyandevaluatethe
influenceofheatinputonhotductilityofthematerial
during nonequilibrium rapid heating, physical
simulation was performed using Gleeble 1500D
ThermoMechanical Simulation System at a rapid
heating rate of 150oC/second to peak temperature
1160oC. The test temperature, 1160oC, is significantly
lower than the equilibrium solidus and solid state
dissolution temperature of the constituent phases in
alloy IN 718 [WeiDiCao 2005]. Specimens were
pulled to failure at the test peak temperature (stroke
rate of50 mm/ second) after heldforspecific holding
times ranging from 0.5 seconds to 3 seconds. The hot
ductility of the alloy at test temperatures was
determined based on the reduction in area by
measuring the initial and final gauge lengths in
regions around the fractured surface with Vernier
calipers. Mounted polished sections of the solution
heat treated sample were etched by swabbing in a
solution of 48g CuCl + 480 ml HCl + 40 ml H2O. The
microstructure of the heat treated sample was first
examined by optical microscopy, using a ZEISS
Axiovert 25 invertedreflected light microscope,
equipped with CLEMEX Vision 3.0 image analysis
software. Higher magnification and more detail
microstructural study including fractographic
examination was carried out using a JEOL 5900
scanningelectronmicroscope(SEM)equippedwithan
Oxford(OxfordInstruments,Oxford,UnitedKingdom)
ultrathin window energy dispersive spectrometer
(EDS). Meanwhile, numerical simulation by two
dimensional thermal elastoplastic finite element
analysis was performed using ANSYS code to study
and better understand the dependence of heat input
on susceptibility of the superalloy to weld HAZ
cracking.

in the alloy [Qian 2003, Hong 2008]. Needle like

secondarydelta(Ni3Nb)phaseprecipitateswerenot
observed in the SHTed, material because the 1050oC
solutionheattreatmentusedisabovetheprecipitation
and solvus temperature of the phase precipitates.
The phase precipitates would normally form in the
material by solidstate precipitation reaction during
heat treatment in the temperature range 860995oC
[Janaki 2006]. The MC carbide precipitates were
uniformly dispersed along grain boundaries and
within grain interior. Furthermore, the main
strengtheningphaseofthealloy,precipitates,were
notobservedintheSHTedalloybytheSEM.

Nbrichcarbide

Tirichcarbide

FIG.1SEMIMAGEOFWROUGHTIN718PREWELDHEAT
TREATEDAT1050OC/1HR/AC

As earlier stated, both the mechanical and

metallurgicalfactorsplayedvitalrolesinsusceptibility
of welds to HAZ cracking. One important
metallurgical factor widely accepted to influence
resistance to weld cracking is hot ductility, which is
primarily investigated in the current work. Previous
work have also showed that increasing weld heat
input leads to reduction in the overall cracking
susceptibility in superalloy welds [Vincent 1980,
Thompson1998,Richard1994,Idowu2007],however,
themechanismthroughwhichhighheatinputcauses
reductionintheextentofintergranularHAZcracking
has not been fully studied and comprehended.
Therefore,tofurtherunderstandtheinfluenceofweld
heat input on HAZ cracking in alloy IN 718, the hot
ductility of the alloy was evaluated by subjecting
Gleeble specimens to tensile loading, first after 0.5
secondsand3secondsattesttemperature1160oCand
the result is presented in Figure 2. In general, the
higher the weld heat input the longer the time the
HAZ material spent at the peak temperature, that is,
the highest temperature attain during welding. A
major consequence of intergranular melting in nickel

Results and Discussion

Figure 1 shows a SEM micrograph of solution heat
treated(SHTed)materialwithanaveragegrainsizeof
100 30 m. The microstructure consists of primary
MC carbides based on titanium and niobium rich
carbideswhichhavebeenpreviouslyreportedtoform

89

www.mejournal.orgJournalofMetallurgicalEngineering(ME)Volume2Issue3,July2013

base superalloys is loss of material ductility under

tensile loading [Lin 1993]. The ductility of IN 718
Gleeble sample is initially very low (2.4%) for the
holding time of 0.5 seconds, which is in agreement
with reported poor ductility of solution heat treated
IN 718 alloy at 1160oC by previous investigators
[Vincent 1980, Vishwakarma 2008]. In contrast,
however, upon increasing the holding time of the
Gleeble sample during the hot ductility test from 0.5
secondsto3seconds,whichissimilartoincreasingthe
magnitudeoftheheatinputontheGleeblespecimen,
an improvement in ductility from 2.4 % to 17 % was
observed(Figure2).

FIG.2VARIATIONINHOTDUCTILITYOFALLOYIN718WITH
INCREASEINHOLDINGTIMEATPEAKTESTTEMPERATURE
1160OC

Previousstudieshaveshownthatanimportantwayto
confirm grain boundary liquation is to perform
fractographystudyontheassociatedfissuredsurface.
Scanning electron microscopy study of the fracture
surface of the Gleeble hot ductility specimens
confirmed brittle intergranular failure in specimens
tested at a peak temperature of 1160oC with 0.5
(a)

seconds holding time (Figure 3a). Higher

magnification examination of the grain surfaces
revealedregularroundishsurfacefeaturesalloverthe
grain facets (insert in Figure 3a) which have been
reportedtobeindicationsofresolidifiedintergranular
liquid phase [Wang 1991, Qian 2003]. For clarity and
comparison,examinationofthefracturesurfaceofhot
ductility specimen tested at 1160oC for 3 seconds is
showninFigure3b.Thespecimenexhibitedaductile
transgranularfailuremodewithshearstretchmarking
in contrast to the intergranular brittle failure features
observedinspecimensheldfor0.5secondsattestpeak
temperature 1160oC (Figure 3a). Therefore, it is
reasonabletoinferthatthenoticeableimprovementin
ductilitywithanincreaseinholdingtimesatthepeak
temperature is one of the vital contributing factors
aiding the reduction in extent of weld cracking with
increaseinheatinputinsuperalloyIN718.Asidefrom
poor hot ductility induced HAZ cracking that occurs
during welding of superalloy, the magnitude of
stresses and strains generated during weld cooling is
another factor that could influence the extent of the
HAZ cracking. During welding, besides the effect of
mechanical restraint, generation of thermally induced
stressandassociatedstrainisunavoidable,becauseof
the differences in the rates of expansion and
contractionofdifferentregionsintheHAZcausedby
the thermal gradient induced by welding operation.
The magnitude of such thermally induced welding
stress is generally influenced by HAZ temperature
gradient.Highheatinputweldsareknowntoexhibit
shallow gradient of temperature in the HAZ during
weldingcomparedtothesteepHAZthermalgradient
thatcharacteriseslowheatinputwelds.
(b)

FIG.3SEMFRACTOGRAPHOFGLEEBLESPECIMENTESTEDA)AT1160OCHELDFOR0.5SECONDSANDB)AT1160OCHELDFOR3
SECONDS

In this study, an additional investigation, using

numerical simulation, was performed to enable a
better understanding of the dependence of the

90

magnitudeofHAZcrackingontheweldingheatinput.
In its basic form, heat input, Q1, is dependent on the
power of the welding heat source, P, and welding

JournalofMetallurgicalEngineering(ME)Volume2Issue3,July2013www.mejournal.org

speed,,throughtheequation;

where k is a constant. Accordingly, the magnitude of

welding heat input can be varied by either changing
the laser power or welding speed. In the present
numericalsimulationstudy,boththelaserpowerand
weldingspeedwerevaried,exclusiveofeachother,to
analyze the effects of heat input on weld
characteristics during laser beam welding. Also,
crackinggenerallyoccurswhensignificantamountsof
thermal strains are generated during weld cooling.
Hence, the rate at which thermal strain develops at
different weld locations, thus, become important in
evaluating the propensity to HAZ liquation cracking.
The numerical model, developed in this work, was
used to calculate the strain rates generated at various
locations along the weld fusion boundary and
determinehowthesevaluesareinfluencedbywelding
speed and laser power. Figures 4 and 5 show
numerically calculated variation of strain rates along
theweldprofilefordifferentvaluesofweldingspeed
and power, during the laser welding. The abscissa
shows the distance from the weld top surface to the
rootoftheweld,alongtheweldfusionboundary.The
left side ordinate shows the distance from weld
centerline, along the weld top surface (a sketch of a
typical welded jointand distancesis shown in Figure
6),whiletherightsideordinateshowsthestrainrate.
Detailed analyses of the results show that while
increasingweldingspeedcausedasignificantincrease
in strain rate, an opposite effect is produced by
increasing power. The two trends show the same
influence of heat input on the rate at which thermal
strains are generated along the weld profile, starting
fromtheweldsurfacetotheweldroot.

Distance from weld bead centerline(mm)

1.4

1.5m/min,Strain rate
3.5m/min,Strain rate
6m/min,Strain rate

1.0
6

0.8

0.6

0.4
2
0.2

0.0
0.0

0.5

1.0

1.5

2.0

2.5

3.0

0
4.0

3.5

Distance from the surface to root of weld bead along fusion boundary(mm)

FIG.4STRAINRATEVARIATIONALONGFUSIONBOUNDARY
OFLASERWELDSPRODUCEDATDIFFERENTVALUESOF
WELDINGSPEEDATCONSTANTPOWER.
5

Distance from weld bead centerline(mm)

1.2

1.0

6kW,weld pool shape

3.5kW,weld pool shape
1.5kW,weld pool shape

6kW,Strain rate
3.5kW,Strain rate
1.5kW,Strain rate
4

0.8
3
0.6
2
0.4

0.2

0.0
0.0

0.5

1.0

1.5

2.0

2.5

3.0

0
4.0

3.5

Distance from the surface to root of weld bead along fusion boundary(mm)

FIG.5STRAINRATEVARIATIONALONGFUSIONBOUNDARY
OFLASERWELDSPRODUCEDATDIFFERENTVALUESOF
LASERPOWERATCONSTANTSPEED.

Acarefulstudyofthecalculatedresults,presentedin
Figures 4 and 5 shows another interesting behaviour
intermsofvariationsofthestrainratealongtheweld
profile.Areverseinthestrainratebehaviouroccursat
theneckregionofthenailshapedweldprofiles.The
strainrateinitiallydecreasesfromtheweldsurfaceto
the neck region, however, starting from the neck
region,therateincreasestowardstherootoftheweld.
Experimentalobservationreportedinpreviousstudies
showed that the HAZ liquation cracks concentrate
aroundtheneckregionofwelds[Krenz2009,Idowu
2007]. The strain rate gradients within the neck
region in welds produced by different welding
conditionswerenumericallycalculatedtocomparethe
propensitytocracking.AsshowninFigures7and8,

1.2

10

1.5m/min,weld pool shape

3.5m/min,weld pool shape
6m/min,weld pool shape

Strain rate for crack susceptibility (1/s)

Strain rate for crack susceptibility (1/s)

Q1 k

increaseinheatinput,eitherbyreducingthewelding
speedorincreasingthepower,reducesthemagnitude
ofthestrainrategradientwhichisinagreementwith
previouslyreportedexperimentalfindinginIN718

FIG.6SKETCHOFATYPICALWELDEDJOINTSHOWING
WELDCENTRELINE,ABSCISSAANDTHEORDINATE
DISTANCES

91

www.mejournal.orgJournalofMetallurgicalEngineering(ME)Volume2Issue3,July2013

Also, numerical calculations show that thermally

generatedstrainrategradientwithintheweldneck
region, which appears to be a major mechanical
drivingfactorthatinfluencesHAZcrackingduringthe
laser beam welding, increases with reduction in heat
input. Therefore, decrease in weld HAZ liquation
cracking is attributed to improved material hot
ductility coupled with reduced thermally induced
strain rate gradient that is generated during weld
cooling.

4.0

Strain rate gradient(1/s mm)

3.5
3.0

Laser welding

2.5
2.0
1.5
1.0
0.5

ACKNOWLEDGMENT

0.0
1

Welding speed(m/min)

FIG.7STRAINRATEGRADIENTINTHENECKREGIONOF
LASERWELDUSINGDIFFERENTWELDINGSPEEDSAT
CONSTANTPOWER.

The authors would like to thank Standard Aero Ltd,

WinnipegandNSERCofCanadaforthefinancialsupport.
REFERENCES

Boech,W.JandCanada,H.B(1969).Precipitationreactions

3.0

and stability of Ni3Nb in alloy 718, Journal of Metals,

Strain rate gradient(1/s mm)

21(10).

Laser welding

2.5

HongJ.K,ParkJ.H,ParkN.K,EomIS,KimM.BandKang
2.0

C.Y(2008).Microstructureandmechanicalpropertiesof
Inconel 718 welds by CO2 laser welding. Mater. Proc.

1.5

Tech.201.
1.0

Idowu,O.A,Ojo,O.AandChartuverdi,M.C(2007).Effect
of Heat Input on HAZ cracking in laser welded ATI

0.5

Allvac 718Plus superalloy. Materials Science and

0.0
0

EngineeringA454455.

Laser power(kW)

FIG.8STRAINRATEGRADIENTINTHENECKREGIONOF
LASERWELDUSINGDIFFERENTLASERPOWERVALUESAT
CONSTANTWELDINGSPEED.

alloy[Krenz2009].Therefore,thestrainrategradient
within the weldneck region appears to be an
important mechanical driving factor that influences
the occurrence of the HAZ liquation cracking during
application of low heat input laser beam welding of
superalloys.
Conclusions
Physicalandnumericalsimulationswereperformedin
order to better understand how weld HAZ liquation
cracking in a nickelbase superalloy, IN 718, is
influenced by the magnitude of welding heat input.
Gleeble thermomechanical simulation indicates that
increaseinheatinputminimizestheextentofthehot
ductilitydamageinthematerialduetothelongertime
available for resolidification of nonequilibrium
intergranularliquidthatotherwiseembrittlesthealloy.

92

Janaki Ram G. D, Venugopal R. A, Rao P. K and

MadhusudhanR.G(2006).Hightemperaturemechanical
properties of Inconel 718 pulsed NdYAG laser welds.
Mater.HighTemp.23.
Krenz,D.Egbewande,A.T,Zang,H.RandOjoO.A(2009).
SinglepasslaserjoiningofIN718superalloywithfiller.
Mater.Sci.Tech.
Luo,X.Shonozaki,K.Kuroki,H.Yoshihara,S.Okumoto,Y
and Shirai, M (2002). Analysis of temperature and
elevated temperature plastic strain distributions in laser
welding HAZ study of laser weldability of Nibase
superalloy.WeldingInternational,vol.XVI.
Lingenfelter, A (1989). Welding of Inconel Alloy 718. A
Historical Overview, Superalloy 718 Metallurgy and
ApplicationsEditedbyE.A.Loria.TheMinerals,Metals
&MaterialsSociety.
LinW,LippoldJ.C,BeaslackW.A(1993).Analysisof
Weldability testing technique for heat affected zone

JournalofMetallurgicalEngineering(ME)Volume2Issue3,July2013www.mejournal.org

liquationcracking.Weld.J.,71(4).

Join.,7(2).

McCay,M.H.McCay,T.Sharp,CandDahotre,N.B(1991).

Thompson, R. G (1998). Microfissuring of Alloy 718 in the

Laser welding techniques for Alloy 718TMS.

WeldHeatAffectedZone,J.Metals40.

Superalloy718,625andvariousderivatives.

Thompson, E. G (1969). Hot Cracking studies of alloy 718

Qian M and Lippold J. C (2003). Effect of annealing twin

weldheataffectedzones,48(2).
Vincent,E.R(1980).TheMicrostructureofWeldsinInco718,

generated special grain boundaries on HAZ liquation

UniversityofBristol,UK.

crackingofnickelbasesuperalloys.ActaMater.51.
Qian, M (2001). An Investigation of repair weldability of

Vishwakarma, K and Chaturvedi, M. C (2008). A study of

Waspaloy and Alloy 718. PhD Dissertation. The Ohio

HAZ microfissuring in a newly developed Allvac 718

StateUniversity.

Plus Superalloy, Superalloy 2008, TMS (The Minerals,

Radavich, J. F and Carneiro T (2005). 6th International

metals&MaterialsSociety).

SymposiumonSuperalloys718,625,706andDerivatives.

Woo,C.KangandNishimoto,K(2001).Influenceofelectron

Richards,N.L,Nakkalil,RandChaturvedi,MC(1994).The

beam welding parameters and metallurgical factors on

Influence ofEBW parameters on HAZ microfissuring in

intergranular liquation cracking susceptibility of castIN

Incoloy903.Metall.Mater.Trans.A.vol25A.

718.MetalsandMaterialsInternational,vol.VII,(4).

Shonozaki,K.Kuroki,H.Luo,X.Ariyoshi,HandShirai,M

WeiDiCao (2005) Solidification and Solid State Phase

(1991).Effectsofweldingparametersonlaserweldability

TransformationofAllvac718PlusAlloy,Superalloy718,

of IN 718. Study of laser weldability of Nibase heat

625,706andDerivatives2005,EditedbyE.A.Loria,TMS

resistant superalloy (1st report). Welding International,

(TheMineral,MetalsandMaterialsSociety).

vol.XIII(12).

Wang, L.A (1991). A methodology for quatifying heat

Sehar, N. C and Reed, R. C (2002). Power beam welding of

affected zone liquation cracking susceptibility, Ph.D.

Thesis,TheOhioStateUniversity,Columbus,OH.

thicksectionnickelbasedsuperalloys,Sci.Technol.Weld.

93