Surface & Coatings Technology 326 (2017) 63–71

Contents lists available at ScienceDirect

Surface & Coatings Technology

journal homepage: www.elsevier.com/locate/surfcoat

Hardfacing of AISI H13 tool steel with Stellite 21 alloy using cold metal
transfer welding process
Rajeev G.P., Kamaraj M., Srinivasa R. Bakshi ⁎
Department of Metallurgical and Materials Engineering, Indian Institute of Technology Madras, Chennai 600036, India

a r t i c l e i n f o a b s t r a c t

Article history: Cold metal transfer (CMT) welding is a new gas metal arc welding process having several advantages such as low
Received 11 April 2017 heat input and spatter free welding. This makes it of great advantage in weld cladding applications. In this study,
Revised 30 June 2017 hardfacing of AISI H13 die steel with Stellite 21 alloy has been carried using CMT process. Coatings were depos-
Accepted in revised form 20 July 2017
ited on H13 substrate in annealed as well as quenched & tempered (Q&T) condition at room temperature as well
Available online 21 July 2017
as with a preheat of 400 °C. The Q&T substrate with and without preheat and the annealed substrate without pre-
Keywords:
heat were found to be susceptible to underbead cracking upon Stellite deposition. The cracking in the heat affect-
Stellite 21 ed zone (HAZ) was due to formation of brittle martensite upon rapid cooling which is associated with formation
Hardfacing of high tensile residual stresses at the bead toe. The annealed substrate with preheat of 400 °C showed the least
H13 cracking tendency. The cracking tendency was investigated by studying the variation of the microstructure and
Cold metal transfer microhardness along the depth of HAZ. The dilution levels based on Fe content was found to be 3–4%, which
Cracking was considerably lower than that of conventional arc welding deposits. The Stellite coated H13 plate (annealed
Dilution with preheat) could be successfully subjected to quenching & tempering heat treatment to restore the properties
of the substrate without introducing any defects.
© 2017 Elsevier B.V. All rights reserved.

1. Introduction techniques to deposit a wear resistant material on the substrate [6].

Hardfacing is a very good method for forging die applications because
In hot working operations such as forging, the die material is sub- of the metallurgical bonding of the coating with substrate and large
jected to repeated impact loads and high temperature for prolonged coating thicknesses that can be deposited compared to other coating
time. In such harsh environment, the die gets damaged and needs to processes such as thermal spraying.
be replaced which reduces the productivity and adds to the production Co-based Stellite® alloys were first developed by Elwood Haynes in
cost. The major causes of damage are wear, plastic deformation, thermal 1900s and are considered as a good candidate for wear resistance appli-
fatigue and mechanical fatigue [1]. Among the mentioned modes of fail- cations, especially at elevated temperatures [7]. Stellite alloys are classi-
ures above, it has been reported that 70% of the die damage is due to fied into two types based on their composition; 1) Co-Cr-W-C type and
wear [2,3]. Various surface treatment techniques and surface coatings 2) Co-Cr-W/Mo-Ni/Fe-C type with some Si and B additions. Stellite al-
have been applied on forging dies in order to reduce wear and thereby loys derive their high hardness from the carbides dispersed in a relative-
improve its service life. AISI H13 is a Cr based hot working tool steel ly soft Co-based alloy matrix strengthened by Cr and W/Mo [8]. The
commonly used for hot forging and die casting applications. The H abrasion resistance of these alloys is attributed to the high carbide vol-
grade steels possess good toughness and shock resistance. These steels ume fraction and morphology [9], whereas the matrix composition is
are air hardenable in nature and are used in quenched and tempered important in case of adhesive wear [8]. Although at low temperatures
condition with adequate toughness. The hot work steels generally con- (less than 417 °C), pure Co is stable as hexagonal close packed (hcp)
tain 0.3 to 0.5 wt% carbon and medium alloy addition [4]. The major structure (ε), in practice Stellite alloy matrix exist in a face centered
weldability problem of these steels is the formation of martensite in cubic (fcc) structure (γ) which is metastable [10]. The allotropic trans-
the HAZ due to rapid cooling which results in cracking. In general the formation in Co-alloys is martensitic in nature and can be induced by
hot hardness values of H grade steels are relatively low beyond 425 °C low temperature deformation depending on stacking fault energy
and this necessitates the use of a high temperature wear resistant coat- (SFE) of the alloy. If the SFE is relatively low, the strain induced martens-
ing for improved service life [5]. The hardfacing process uses welding itic transformation increases the work hardening rate and makes it dif-
ficult to deform plastically [11]. Also, the resultant hcp structure has
⁎ Corresponding author. lower number of slip systems than fcc and is difficult to deform. These
E-mail address: [email protected] (S.R. Bakshi). features of Co-based alloys contribute to its adhesive wear resistance.

http://dx.doi.org/10.1016/j.surfcoat.2017.07.050
0257-8972/© 2017 Elsevier B.V. All rights reserved.
64 R. G.P. et al. / Surface & Coatings Technology 326 (2017) 63–71

Fig. 1. Optical micrographs of H13 steel substrate in a) annealed, and b) quenched and tempered (Q&T) condition.

Mo and/or W stabilize the hcp phase whereas Fe and Ni stabilize the fcc and better deposition rate [25]. CMT has also been explored as a low di-
phase. It is known that Fe increases the SFE and reduces the wear prop- lution cladding [26] and low energy process for weld repair applications
erties [12]. During hardfacing of Stellite alloys on steel, dilution of the for Al-alloys [27]. It has also been shown that Inconel 718 alloy cladding
coating occurs due to mixing with the molten substrate, which increases of low dilution of less than 10% can be deposited by CMT [28].
the Fe content in coating and reduces its wear properties. It is well As per the authors' knowledge, there are no reports available on the
known that the dilution increases with increase in heat input of the use of CMT process for hardfacing of H13 tool steels with Stellite 21.
welding process [13]. Also, in the reported literature on weld cladding, the cracking issues re-
Various processes have been used for hardfacing of Stellite on hot lated to substrate condition have not been elucidated properly. In this
forging die materials such as Manual Metal Arc Welding (MMAW), work, CMT process has been explored and the effect of substrate condi-
Gas Tungsten Arc Welding (GTAW) [14], Gas Metal Arc Welding tion on the microstructure, cracking tendency of the substrate, dilution
(GMAW) [15], Laser cladding [16] and Plasma Transferred Arc (PTA) and hardness has been studied. A method is provided for successful de-
welding [17]. Even though these hardfacing processes were done at dif- position of the coatings on H13 tool steel.
ferent operating conditions and substrate materials, it can be qualita-
tively seen that the dilution of the coating decreases in the order 2. Materials and methods
MMAW, GMAW, GTAW, PTA and Laser cladding. PTA has been reported
to be suitable for hardfacing applications as it can produce coatings with The substrate material used in this study was AISI H13 tool steel
dilution levels much lower than conventional arc welding processes block (130 mm × 75 mm × 30 mm). The typical microstructures of
[18]. Sidhu et al. have used high velocity oxy-fuel (HVOF) thermal annealed and quenched & tempered (Q&T) H13 steel are shown in
spray process to deposit Stellite coatings on Superalloys. Compared to Fig. 1. The average value of microhardness was 240 HV0.5 and
welding based hardfacing processes, the thickness of the coatings pro- 550 HV0.5 for annealed and Q&T H13, respectively. Stellite® 21 wire
duced by thermal spray processes are restricted to few hundreds of mi- (1.2 mm in diameter) was used as the filler wire for CMT process. The
crons. Welding processes can deposit coatings of few millimeters composition of the substrate and coating material are given in Table 1.
thickness in a single pass [19]. Prasad Rao et al. have used friction surfac- The Stellite alloy was deposited on H13 steel using a Fronius Transplus
ing to deposit Stellite 6 alloy on low carbon steel and observed im- Synergic 4000 CMT welding machine (Fronius International GMBH,
proved chemical homogeneity of the coating [20]. Welding processes Wels, Austria). Stellite coating were deposited under four substrate con-
are also used for modifying the existing hardfacing alloy composition ditions; annealed H13 with and without preheat (named AWPH and
in order to improve the wear resistance. Singla et al. used shielded AWOPH respectively) and Q&T H13 with and without preheat condition
metal arc welding process (SMAW) for hardfacing and modifying the (named Q&TWPH and Q&TWOPH respectively). The preheat and inter-
iron based alloy with addition of various amount of vanadium to im- pass temperatures were maintained at 400 °C between successive beads
prove the high temperature wear property of the alloy [21]. The type and temperature was verified using a thermocouple connected to the
of hardfacing process and its process parameters have significant effect surface of the substrate. Heating was carried out by placing the sub-
on the microstructure, dilution and hardness and thereby the wear re- strate on an electric heater, which was powered based on the feedback
sistance of the deposited material [22]. from the thermocouple. Bead-on-plate trials were carried out at first to
Cold metal transfer (CMT) is a modified GMAW process developed find out the welding parameters resulting in visibly uniform beads
by Fronius International. The term ‘Cold’ in the name highlights the showing good wettability with the substrate. Using these welding pa-
very low heat input of CMT process relative to other arc welding pro- rameters, a monolayer coating consisting of 5–6 overlapping beads
cesses. Low heat input is achieved by reducing the current during the was deposited. The trial beads and coating are shown in Fig. 2. The
short circuit phase with the help of a microcontroller in the power sup- values of welding parameters used for coating are given in Table. 2.
ply unit and implementing a retraction motion of the filler wire for en-
abling metal transfer. Pickin et al. have reviewed the operating principle Table 1
and equipment features of CMT [23]. CMT process is also extensively Composition of the materials used.
used for dissimilar material joining applications particularly for mate- H13 substrate Stellite® 21 wire
rials with large difference in melting point where fusion of both mate-
Element wt% Element wt%
rials can result in intermetallic formation [24]. The low heat input of
the process reduces the intermetallic layer thickness and improves the Fe 90.92 Co 62.8
C 0.49 C 0.2
joint efficiency. CMT offers the key advantage of low heat input and
Cr 4.99 Cr 27
thereby lower dilution levels, which is very important for a sound coat- Mo 1.3 Mo 3
ing. However, there are only few reports on deposition of coatings with Mn 0.4 Mn 0.8
CMT process. Pure CMT mode was reported as suitable for depositing Si 0.97 Si 0.7
anti-corrosive Inconel 625 overlay on C\\Mn steel which resulted in V 0.93 Ni 2.7
Fe 2.8
lower dilution of 2.45% in terms of weight percentage of iron content
 R. G.P. et al. / Surface & Coatings Technology 326 (2017) 63–71 65

electron microscope (Inspect F, FEI Company, Hillsboro, OR, USA) hav-

ing an energy dispersive X-ray spectroscopy facility (EDAX Inc.,
Mahwah, NJ, USA). X-ray diffraction (XRD) study was carried out
using a Panalytical X-pert diffractometer on the coating surface which
was made flat by grinding using 120 grit SiC paper. The microhardness
variation along the depth of the coating was measured using a Vickers
microhardness tester (Wolpert Wilson, Germany) with a load of 500 g
and dwell time of 15 s.

3. Results and discussion

3.1. Microstructural characterization of coatings

Fig. 3 shows panoramic view of the cross section of the coating pre-
Fig. 2. Bead-on-plate trials and the deposited coating. pared by stitching together several continuous SEM micrographs. Fig. 3a
shows the coating on Q&T substrate without preheating. Three cracks
were observed near the interface region in the heat affected zone
Table 2 (HAZ) of the substrate. The cracks are found to be under bead cracks.
Final welding parameters chosen for deposition of coating. The longer crack which is below the starting bead is about 2 mm in
Parameters Values length. The cracking tendency of the HAZ is due to the formation of brit-
Current (ampere, A) 126
tle martensite due to rapid cooling and also the presence of tensile re-
Voltage (volts, V) 9.1 sidual stress from the restraint offered by Stellite coating and the
Wire feed speed (m/min) 3.7 unaffected thick base metal during the contraction of the bead. A similar
Welding speed (in./min) 12 cracking problem in the HAZ of tool steels during repair welding was
Nozzle to plate distance (mm) 9
also reported by other researchers [29]. However, no evidence in
Argon gas flow rate (l/min) 15
Offset distance between beads (mm) 2.0–3.0 terms of microstructure was provided and cracking was not discussed
Torch angle (°) 90 in detail. It is a general practice that Stellite hardfacing is usually done
on preheated substrates having high carbon equivalent. Preheating of-
fers several advantages such as reduction of cooling rates, reduction of
Before depositing the coating, the substrate surface was ground with temperature gradients and decrease in the tensile residual stresses.
120 grit SiC paper and cleaned with acetone. The coating and the sub- Since the yield strength of the material reduces at higher temperatures,
strate were cleaned using a wire brush before each subsequent the maximum residual stress build up is also lower. In the present study,
deposition. a preheating temperature of 400 °C was chosen which is above the mar-
The transverse section of the deposited coating was prepared using tensite start (Ms) temperature of H13 steel (~370 °C) [4]. The coating
wire electrical discharge machining (wire EDM) and metallographic was then attempted on Q&T substrate with 400 °C preheat. As observed
preparation was done by grinding and polishing. Samples were etched from Fig. 3b, deposition with preheat of 400 °C resulted surprisingly in
with a solution of 50 ml HCl, 10 ml HNO3, 10 g FeCl3 and 100 ml ethanol. extensive underbead cracking. Two long cracks were observed under
The HAZ regions of H13 were etched using Vilella's Reagent (1 g picric the coating which started from the toe. The crack was found to propa-
acid, 5 ml HCl and 100 ml ethanol). The average thickness of all coatings gate horizontally under the coating in the HAZ region leaving only
was found to be about 2.5 mm. Microstructural characterization of the about 5 mm long region without any crack. It has been reported that
transverse sections was carried out using a field emission gun scanning heat treatable alloy steels with higher carbon content (0.3 wt% to

Fig. 3. Stitched SEM images of the cross section of Stellite 21 coating on H13 substrate under different conditions a) Q&T without preheat b) Q&T with preheat c) annealed without preheat,
and d) annealed with preheat. The cracks are circled in the images.
66 R. G.P. et al. / Surface & Coatings Technology 326 (2017) 63–71

Fig. 4. Optical images of the microstructure of the HAZ region a) near the interface, and b) near the crack.

0.5 wt%) can be welded in the soft annealed condition and then heat- lower hardness. It is also noted that the crack does not pass through
treated to attain the desired strength and toughness [30]. So, the coating the over-tempered region which is much below and is expected to
was carried out on annealed substrate without preheat. As shown in Fig. have lower hardness but much higher toughness. It is also observed
3c, four cracks were observed starting from the interface and going into that the region just below the interface has a lower hardness, which
the HAZ of H13 substrate. Two cracks were observed starting from the could be due to the diffusion of cobalt from the coating. Based on
ends and two cracks were seen in the centre close to the interface. these observations, a schematic showing the reasons for cracking is pre-
Coatings were then deposited on the annealed substrate with preheat sented in Fig. 6. As shown in Fig. 6, the substrate close to interface is
of 400 °C. As shown in Fig. 3d, no cracks were found for this condition. heated above the inter-critical temperature and transforms to austenite
This shows that the substrate condition and preheat have strong effect during deposition of the bead. The bead contracts during solidification
on the cracking tendency (Fig. 3a–c), even for a low heat input process and subsequent cooling to the preheat temperature leads to thermal
such as CMT. contraction which is opposed by the restraint of the base metal resulting
in tensile residual stress in the transverse direction in the bead. After the
3.2. Mechanism of cracking coating deposition as the temperature decreases below the Ms., the
transformed austenite changes to martensite resulting in volume ex-
To understand the reasons for underbead cracking, the microstruc- pansion while the coating undergoes thermal contraction. The restraint
ture of the HAZ region was studied. Fig. 4 shows the optical images of offered by the coating as well as untransformed substrate to the volume
the HAZ region close to the interface and near the crack. It is clearly ob- expansion leads to generation of high tensile stresses at the toe of the
served that just below the interface, the prior-austenite grains are very bead leading to cracking. The brittle martensite of lower hardness pro-
large while they are smaller close to the crack. This is due to the fact that vides an easy path for the crack to propagate under the bead in the
the regions close to the interface are exposed to higher temperatures HAZ region. Thus, crack initiation seems to be more critical while
which results in austenitization followed by grain growth. However, crack propagation through martensite is easier. When preheating is
the regions far from the interface also get austenitized but do not coars- not employed, the depth/volume of austenite formation and subse-
en. Higher austenitization temperatures are expected to result in com- quent martensite is smaller leading to lower stress generation. Thus,
plete dissolution of carbides and consequently formation of a harder the thickness of the brittle martensite layer is reduced when preheat
martensite [31]. Fig. 5 shows the microhardness at different depths is not used and this leads to smaller cracks. These results are supported
below the interface. It is clearly observed that the prior-austenite by the hardness variation along the depth presented in later section. In
grain size and hardness reduces as we approach the cracked region. case of annealed substrate without preheat, higher cooling rate and high
This indicates that the crack passes through the martensite having temperature gradient must have contributed more to the initiation of
cracks. In the case of annealed substrate with preheat, the residual
stresses as well as the restraint offered by the relatively softer substrate
is lower and the martensitic transformation is accommodated without
cracking. From these four cases, it is concluded that coating must be car-
ried out on annealed H13 with a preheat of 400 °C to avoid HAZ
cracking.

3.3. Hardness vs. depth profile of the coatings

The effect of the changes in the microstructure and composition at

various depths of the coating and substrate was studied by measuring
hardness and taking micrographs in different depths on the polished
cross section. Fig. 7 shows the variation of the microhardness from top
of the coating to the substrate. Fig. 7a shows the hardness profile for
coating on Q&T substrate with and without preheat condition. In both
cases the average hardness of coating was found to be around 400 ±
20 HV0.5 and for the unaffected substrate it was around 620 ±
15 HV0.5. In case of Q&T substrate without preheat (Q&TWOPH), the
hardness reached a maximum value nearly 580 HV0.5 just below the in-
terface which was followed by a zone of lower hardness (420 HV0.5)
followed by the unaffected substrate. The width of the peak corresponds
Fig. 5. Optical micrograph showing the microhardness indents and values at different well to the thickness of the HAZ region as seen in Fig. 3 (brighter region
locations from the interface to the crack. below the interface) and is observed to be larger for the coatings
 R. G.P. et al. / Surface & Coatings Technology 326 (2017) 63–71 67

Fig. 6. Schematic illustrating the reasons for build-up of stresses which can lead to the formation of HAZ crack.

prepared with preheat. In case of coating with preheat (Q&TWPH), the down to 220 ± 10 HV0.5 over a depth of ~ 2 mm. This is again due to
hardness profile significantly changes. The microhardness close to inter- the fact that in the presence of preheat, austenitization occurs to a larger
face is nearly 700 HV0.5 and is found to decrease gradually over a large depth (~ 2 mm) and there is complete dissolution of carbides which
depth (~ 2 mm). This was explained based on the microstructures leads to the formation of a thicker and harder martensite. Thus, it is
shown in Fig. 4. In the case of quenched and tempered substrate with seen that the Q&T substrates are affected significantly during the
or without preheat, it is observed that there is formation of a zone hav- welding process as compared to annealed substrates. Hence, the crack-
ing lower hardness (~350 HV0.5) in between the hardened region and ing tendency is observed to be severe for Q&T substrates (Fig. 3a and b).
the unaffected substrate. The microstructure of this region shown in
Fig. 8 indicates that it is the over-tempered region having coarse car- 3.4. Volumetric dilution of coatings
bides. It is noted that the crack does not pass through this low hardness
region since it possesses higher toughness than the untempered mar- Since, crack free coatings could be produced using annealed H13
tensite. It is observed that preheating increases the width of the hard- with preheat (AWPH), further studies on microstructure, dilution and
ened as well as over-tempered regions from ~ 0.5 mm to ~ 2 mm and heat treatment were carried out on AWHP coating. Dilution of Stellite
~ 1.5 mm to ~ 3.5 mm, respectively. This is expected since a larger 21 coating due to mixing of molten substrate can have significant effect
depth of the substrate gets austenitized upon cladding with preheating. on mechanical and wear properties of the coating. The volumetric dilu-
Similarly, a larger depth of the substrate is expected to get over-tem- tion of coating was calculated according to Eq. (1) below [32].
pered. Fig. 7b shows the hardness profile for coatings deposited on  
annealed substrate with and without preheat. In both cases, the average ρ Fes −Fec −1
Dv ¼ 1 þ s ð1Þ
hardness of coating was found to be around 382 ± 15 HV0.5 and for the ρw Fec −Few
unaffected substrate it was around 220 ± 10 HV0.5. A peak in the hard-
ness profile can be observed, which corresponds to the HAZ region. In Here ρs and ρw represent the densities of H13 and Stellite 21 wire,
case of annealed substrate without preheat (AWOPH), the hardness respectively. Fes, Fec and Few represent the iron content (in wt%) in
reached a maximum of nearly 582 HV0.5 and decreased down to 220 the substrate, coating and the Stellite 21 wire, respectively, and were
± 10 HV0.5 over a depth of ~ 1.5 mm. The high hardness is attributed measured using EDS technique in SEM. The variation of weight percent-
to formation of martensite which resulted from rapid cooling of HAZ. age of Fe was measured using area scan method as shown in Fig. 9a and
Unlike the Q&T substrates, here there is no formation of softened re- c. The calculated percentage volumetric dilution (% Dv) is plotted as a
gions in the substrate. In case of annealed substrate with preheat function of depth (mm) as shown in Fig. 9b and d. It is important to
(AWPH), the hardness of HAZ reached a maximum value nearly note that the dilution near the top of the monolayer coating of average
684 HV0.5 below the interface and gradually decreased and reached thickness 2.5 mm was found to be about 3.4% (Fe = 5.9 wt%) which is

Fig. 7. Microhardness vs. depth profile for Stellite 21 coatings on a) Q&T H13 with and without preheat, and b) annealed H13 with and without preheat.
68 R. G.P. et al. / Surface & Coatings Technology 326 (2017) 63–71

3.5. Microstructural characterization of the coatings

The SEM images of the microstructure of the as deposited coating on

AWPH substrate is shown in Fig. 10. The microstructure consists of pri-
mary Co rich matrix (dark contrast) along with Mo and Cr rich phase
and carbides in inter-dendritic region. Variation in the scale of micro-
structure from the top region to the interface of the coating is as
shown in Fig. 11 (A — top, B — middle, and C — near interface). The den-
drites can be seen to grow perpendicular to the interface and they be-
come finer towards the surface. The secondary arm spacing decreases
from 5 to 6 μm in the interface to 2.5 μm in the surface region, which
is in accordance with the increase in solidification rate towards the coat-
ing surface. However, this change seems to have insignificant effect on
the microhardness of the coating as seen from Fig. 7b.

Fig. 8. SEM back scattered electron image of the over tempered region of the HAZ in the 3.6. Effect of heat treatment on microstructure of Stellite 21 coated H13
substrate.

The desirable microstructure of H13 material in die application is

much below the dilution levels obtained by conventional hardfacing Q&T condition, which possesses adequate strength and toughness. It is
methods such as MMAW, GTAW, conventional GMAW where it ranges general practice to heat treat the H13 steel after welding. The coated
from a minimum of 10–12% to a maximum of 40% in the first layer [33]. sample on AWPH substrate was subjected to quenching treatment by
As shown in Fig. 9d, the dilution level in regions approximately 2 mm heating to 1050 °C and soaking for 30 min followed by air cooling. Tem-
away from the interface was found to be 8% (Fe = 10 wt%) and about pering treatment was done for 2 h at 550 °C followed by furnace cooling.
16% (Fe = 16.4 wt%) near the interface. The obtained dilution values The microstructural features and microhardness variation of the coating
using CMT process are comparable with processes like PTA process, were studied after heat treatment.
where it ranges from a minimum of 6% to a maximum of 70% [34,35]. A comparison of the microstructure of the as deposited and heat
In case of laser cladding, the dilution values range from a minimum of treated coating are shown in Fig. 12. X-ray element maps showing the
3% to a maximum of 20% [36]. Thus, CMT can be an attractive econom- distribution of various elements in the microstructure are also shown
ical process compared to laser cladding. It is to be noted that the dilution in Fig. 12. Fig. 12a represents the as-deposited Co rich matrix (dark
can also be further reduced by optimizing the process parameters and is phase) with the interdendritic phase (grey area) and carbides (dark
the scope for future work. areas within the grey area). From the elemental maps, it is evident

Fig. 9. SEM images and corresponding percentage dilution variation for region close to the surface (a and b) and across the interface (c and d).
 R. G.P. et al. / Surface & Coatings Technology 326 (2017) 63–71 69

Fig. 10. SEM back scattered electron images of a) cross section of coating and substrate b) the central region of as deposited coating.

that Cr and Mo segregate in the interdendritic phase and a correspond- 3.7. Effect of heat treatment on hardness of Stellite 21 coated H13
ing depletion of Co is observed in these regions. Fe was also found to be
distributed in the Co rich areas whereas Ni and Mn showed a uniform The microhardness vs. depth profiles of the as-deposited and heat
distribution. The carbides in the interdendritic phase are found to be treated coating are shown in Fig. 14. It is observed that the hardness
rich in Cr and Mo. As shown in Fig. 12b, after the quenching treatment, of Stellite 21 coating increases slightly after quenching from 382 ± 15
the dendritic structure of the coating was changed to relatively coarse to 400 ± 20 HV0.5. This could be due to the homogenization of the
grain structure with randomly distributed islands of Mo and Cr rich cast structure and increased solid solution strengthening of the matrix.
phase and carbides in the grain boundaries. Some of the carbide remains The hardness of the H13 increased from 220 ± 10 to 677 ± 40 HV0.5
undissolved by the heat treatment. After tempering treatment, the car- after quenching. The tempering treatment is found to reduce the hard-
bide precipitation increases as observed in Fig. 12c. It is clearly observed ness of Stellite 21 from 400 ± 20 HV0.5 to 343 ± 15 HV0.5, which could
that the carbides are rich in Cr and Mo and do not contain Fe and Co. It is be due to the coarsening of the grains. It is noted that the final hardness
known that these carbides are of the type M23C6 in Stellite 21 [37]. compares well with the Stellite 21 coating deposited using GTAW pro-
Higher carbon Stellite alloys such as Stellite 6 are known to contain cess (320 HV for single layer and 350 HV for multi-layer) as reported
M7C3 type carbides. Fig. 13 depicts X-ray diffraction pattern of the coat- in literature [33]. In the quenched condition, the peak hardness of mar-
ing in the as deposited as well as in the heat treated condition. The tensite reduced from 684 HV0.5 to 563 HV0.5 close to the interface. This
major phase identified is the fcc γ-Co matrix and it also contains some could be due to the coarsening of the martensite due to repeated heat
amount of hcp ε-Co, which is also observed by other researchers [37]. treatment near the interface. In the unaffected base metal, the hardness
The M23C6 carbide peaks are found to overlap with that of the matrix increased from 220 ± 10 HV0.5 from annealed condition to 572 ±
phase and match with that for Cr23C6. 15 HV0.5 corresponding to Q&T condition. No cracks were observed in

Fig. 11. Optical micrograph showing the variation in microstructure of Stellite coating from the surface to the interface (A–C).
70 R. G.P. et al. / Surface & Coatings Technology 326 (2017) 63–71

Fig. 12. SEM backscattered electron images and EDS X-ray elemental maps of Stellite 21 coating in a) as deposited b) quenched c) tempered condition.

the substrate after the heat treatment. Thus, the properties of the sub- 3. Preheat was found to result in higher thickness of brittle martensite
strate and the coating could be restored successfully after heat region leading to easy crack propagation in Q&T substrates.
treatment. 4. Crack free deposition could be achieved only in the annealed H13
substrate with preheat of 400 °C.
4. Conclusions 5. A very low dilution of about 3–4% was measured in the first layer
which substantiates the low heat input characteristic of the CMT pro-
cess and its feasibility to be used in low dilution cladding/hardfacing
1. Thick Stellite 21 coatings (~2.5 mm) could be deposited using CMT
applications where conventional arc welding processes like MMAW,
process in a single pass
GTAW and conventional GMAW cannot be used.
2. Underbead cracks were observed in the Q&T substrates with and 6. The coating deposited on annealed substrate with preheat was suc-
without preheat and the annealed substrates without preheat. Crack- cessfully subjected to Q&T heat treatment without any cracking.
ing was found to be through the brittle martensitic region of the HAZ The microhardness of Stellite 21 coating after heat treatment was
of the substrate.

Fig. 13. XRD pattern of the Stellite 21 coating on annealed H13 with preheat in as- Fig. 14. Effect of heat treatment on the microhardness vs. depth profile of Stellite 21
deposited, hardened and tempered condition. coating deposited on annealed H13 with preheat.
 R. G.P. et al. / Surface & Coatings Technology 326 (2017) 63–71 71

found to be 343 ± 15 HV0.5 while the substrate hardness was re- [18] A.S.C.M. d'Oliveira, R. Vilar, C.G. Fede, High temperature behavior of plasma trans-
ferred arc and laser Co-based alloy coatings, Appl. Surf. Sci. 201 (2002) 154–160.
stored to 572 ± 15 HV0.5. [19] T.S. Sidhu, S. Prakash, R.D. Agrawal, Studies of the metallurgical and mechanical
Acknowledgments properties of high velocity oxy-fuel sprayed Stellite 6 coatings on Ni- and Fe-
based superalloys, Surf. Coat. Technol. 201 (2006) 273–281.
[20] K. Prasad Rao, R. Damodaram, H. Khalid Rafi, G.D. Janaki Ram, G. Madhusudhan
Authors are grateful to Naval Research Board of India (MET/11-12/ Reddy, R. Nagalakshmi, Friction surfaced Stellite 6 coatings, Mater. Charact. 70
124/NRBX/MKAM) for funding the CMT welding machine. We would (2012) 111–116.
like to acknowledge help of Mr. K. Rajesh and Mr. A. Elamaran for help [21] Yogesh Kumar Singla, D.K. Dwivedi, Navneet Arora, On the modeling of dry sliding
adhesive wear parameters of vanadium additive iron-based alloys at elevated tem-
in arranging the welding set-up. Authors would like to acknowledge
peratures, Surf. Coat. Technol. 283 (2015) 223–233.
help of Ambattur Metal Treaters (Chennai, India) for helping in heat [22] S. Atamert, H.K.D.H. Bhadeshia, Comparison of the microstructures and abrasive
treatment experiments. wear properties of Stellite hardfacing alloys deposited by arc welding and laser clad-
ding, Metall. Trans. A 20 (1989) 1037–1054.
[23] C.G. Pickin, K. Young, Evaluation of cold metal transfer (CMT) process for welding
References aluminium alloy, Sci. Technol. Weld. Join. 11 (5) (2006).
[24] H.T. Zhang, J.C. Feng, P. He, B.B. Zhang, J.M. Chen, L. Wang, The arc characteristics and
[1] J. Kohopoaa, H. Hakonen, S. Kivivuori, Wear resistance of hot forging tools surfaced metal transfer behaviour of cold metal transfer and its use in joining aluminium to
by welding, Wear 130 (1989) 103–112. zinc-coated steel, Mater. Sci. Eng. A 499 (2009) 111–113.
[2] E. Summerville, K. Venkatesan, C. Subramanian, Wear processes in hot forging press [25] G. Lorenzin, G. Rutili, The innovative use of low heat input in welding: experiences
tools, Mater. Des. 16 (1995) 289–294. on ‘cladding’ and brazing using the CMT process, Weld. Int. 23 (8) (2009) 622–632.
[3] H. Kashani, A. Amadeh, H.M. Ghasemi, Room and high temperature wear behaviors [26] C.G. Pickin, S.W. Williams, M. Lunt, Characterisation of the cold metal transfer (CMT)
of nickel and cobalt base weld overlay coatings on hot forging dies, Wear 262 process and its application for low dilution cladding, J. Mater. Process. Technol. 211
(2007) 800–806. (2011) 496–502.
[4] George Roberts, George Krauss, Richard Kennedy, Tool Steels: 5th Edition, Chapter [27] G.P. Rajeev, M. Kamaraj, S.R. Bakshi, Al-Si-Mn alloy coating on aluminum substrate
13, Hot-Work Tool Steels, ASM International, 1998 219–250. using cold metal transfer (CMT) welding technique, JOM 66 (6) (2014) 1061–1067.
[5] T. Altan, G. Ngaile, G. Shen, Cold and Hot Forging: Fundamentals and Applications, [28] O.T. Ola, F.E. Doern, A study of cold metal transfer clads in nickel-base Inconel 718
ASM International, 2005 278–280. superalloy, Mater. Des. 57 (2014) 51–59.
[6] ASM International, ASM Handbook Volume 6: Welding, Brazing and Soldering, ASM [29] M. Vedani, B. Previtali, G.M. Vimercati, A. Sanvito, G. Somaschini, Problems in laser
International, Materials Park, OH, 1993. repair-welding a surface-treated tool steel, Surf. Coat. Technol. 201 (2007)
[7] K.C. Antony, Wear resistant cobalt base alloys, JOM 35 (2) (1983) 52–60. 4518–4525.
[8] H.R. Shetty, T.H. Kosel, N.F. Fiore, A study of abrasive wear mechanisms in cobalt- [30] R.W. Messler, Principles of Welding: Processes, Physics, Chemistry, and Metallurgy,
base alloys, Wear 84 (1983) 327–343. John Wiley, New York, 1999.
[9] K.J. Bhansali, Adhesive wear of nickel- and cobalt-base alloys, Wear 60 (1980) [31] O.E. Okorafor, Fracture toughness of N2 and H13 alloy tool steels, Mater. Sci.
95–110. Technol. 3 (1987) 118–124.
[10] K. Ishida, T. Nishizawa, Bull. Alloy Phase Diagr. II (4) (1990) 357–370. [32] J.T.M. De Hasson, L. De Mol van Otterloo, Surface engineering with lasers: applica-
[11] L. Remy, A. Pineau, Twinning and strain-induced F.C.C. to H.C.P. transformation on tion to co based materials, Surf. Eng. 13 (6) (1997) 471–481.
the mechanical properties of Co-Ni-Cr-Mo alloys, Mater. Sci. Eng. 26 (1976) [33] H. Kashani, M. Sadeghi Laridjani, A. Amadeh, M. Khodagholi, S. Ahmadzadeh, The in-
123–132. fluence of volumetric dilution on the strain induced ϒ → ε martensitic transforma-
[12] K.J. Bhansali, A.E. Miller, The role of stacking fault energy on galling and wear behav- tion in GTAW processed Co–Cr–Mo alloy, Mater. Sci. Eng. A 478 (2008) 38–42.
ior, Wear 75 (1982) 241–252. [34] A.S.C.M. D'Oliveira, R.S.C. Paredes, R.L.C. Santos, Pulsed current plasma transferred
[13] Welding Handbook, 8th ed., American Welding Society, FL, 1991, vol. 2. arc hardfacing, J. Mater. Process. Technol. 171 (2006) 167–174.
[14] M. Farhani, A. Amadeh, H. Kashani, A. Saeed-Akbari, The study of wear resistance of [35] R. Ravi Bharath, R. Ramanathan, B. Sundararajan, P. Bala Srinivasan, Optimization of
a hot forging die, hardfaced by a cobalt-base superalloy, Mater. Forum 30 (2006) process parameters for deposition of Stellite on X45CrSi93 steel by plasma trans-
212–218. ferred arc technique, Mater. Des. 29 (2008) 1725–1731.
[15] L. Fouilland, M. El Mansori, A. Massaq, Friction-induced work hardening of cobalt- [36] J.L. De Mol Van Otterloo, J.Th.M. De Hosson, Microstructural features and mechanical
base hardfacing deposits for hot forging tools, J. Mater. Process. Technol. 209 properties of a cobalt-based laser coating, Acta Mater. 45 (3) (1997) 1225–1236.
(2009) 3366–3373. [37] J. Chen, X.Y. Li, H. Dong, Formation and characterisations of S phase in plasma
[16] A. Frenk, W. Kurz, Microstructural effects on the sliding wear resistance of a cobalt carburised high carbon Stellite 21 CoCr alloy, Surf. Eng. 26 (4) (2013) 233–241.
based alloy, Wear 174 (1994) 81–91.
[17] E. Yaedu, A.S.C.M. D'Oliveira, Cobalt based alloy PTA hardfacing on different sub-
strate steels, Mater. Sci. Technol. 21 (4) (2005) 459–466.