Journal of Materials Processing Technology 212 (2012) 402– 407
Contents lists available at SciVerse ScienceDirect
Journal of Materials Processing Technology
j o ur nal ho me p age : www.elsev ier .com/ locate / jmatprotec
ool steel and copper coatings by friction surfacing – A thermography study
. Prasad Raoa,∗, A. Veera Sreenua, H. Khalid Rafia, M.N. Libinb, Krishnan Balasubramaniamb
Materials Joining Laboratory, Department of Metallurgical and Materials Engineering, Indian Institute of Technology Madras, Chennai 600 036, IndiaMachine Design Section, Department of Mechanical Engineering, Indian Institute of Technology Madras, Chennai 600 036, India
r t i c l e i n f o
rticle history:eceived 6 August 2011eceived in revised form0 September 2011ccepted 30 September 2011
a b s t r a c t
Infrared thermography was used to record thermal profiles during friction surfacing. Thermal profilesfor different sets of consumable rod/substrates (tool steel/steel; copper/steel and copper/copper) wererecorded and analyzed. The thermal profiles showed distinct stages of plastic deformation with respectto temperature. The mechanism of bonding or no-bonding was discussed based on thermal profile data.It was found that a metallurgically bonded coating can be obtained if the flow stress of the plasticized
vailable online 7 October 2011
eywords:nfrared thermographyolid state coatinghermal profile
Friction surfacing is a useful technique of depositing solid stateoatings on substrates for obtaining enhanced surface properties.t involves an initial rubbing of a consumable rod against a flat sub-trate for a definite dwell time followed by a relative movement ofod and substrate (Fig. 1).
Earlier studies carried out by Madhusudhan Reddy et al. (2009)nd Khalid Rafi et al. (2010a,b) showed few metallic rod/substrateombinations that resulted in friction surfaced coatings while thetudies carried out by Arunsankar (2010) on few other metallicombinations could not produce any successful coatings. Further,he study by Liu et al. (2008) on understanding the mechanismuring friction surfacing was limited to material transfer and theonding mechanisms of friction surfaced coatings are not yetlearly probed. It is necessary to understand the thermal phenom-na that take place during the entire process of coating formations it can influence the coating formation. However, measuringemperatures at the interface of rotating components will haveractical difficulties if conventional methods are applied. InfraredIR) thermography is a convenient, non-contact method for mea-urement of temperature fields for friction surfacing.
The IR camera measures temperature variations based on irradi-nce of an object surface. IR camera has been used for temperatureeasurement in the past for different domains. Thomann and Frisk
(1968) used IR camera for heat-transfer studies, Matte et al. (2009),Huang et al. (2007), Bruggemann et al. (2000), and Mathieu et al.(2006) utilized the non-contact feature of IR camera for weldingand fatigue studies, Pastor et al. (2008) used it for non-destructiveflaw testing, and Marinetti and Vavilov (2010) studied differentaspects of corrosion using IR camera. Reported studies in open lit-erature on thermal profiles during friction surfacing process arelimited (Meola and Carlomagno, 2004; Khalid Rafi et al., 2011). Inthis work, it is aimed to study the feasibility of obtaining surfacecoatings of various ferrous and non-ferrous rods on different sub-strates. It was also aimed to understand the coating mechanismsby recording thermal profiles generated at the coating/substrateinterface by using infrared thermography.
2. Experimental work
Consumable rod/substrate combinations studied are as follows:(a) tool steel D2 rod on steel substrate (‘D2′ is a tool steel grade des-ignated by AISI-SAE containing high carbon and high chromium),(b) commercially pure (CP) copper rod on steel substrate and (c)CP Cu rod on CP Cu substrate. Dimension of the substrate was300 mm × 200 mm × 10 mm; and the consumable rod was: 15 mmdiameter and 100 mm length. The range of parameters (Axial load:2–20 kN; rod rotation speed (RPM); 100–2500 RPM; dwell time30–80 s) used in earlier studies (Bedford et al., 2001; Sreenu, 2010),
were applied for the above consumable rod/substrate combina-tions. Chisel test was carried out to find any de-bonding at theinterface between coating and substrate. Chisel test gives first handinformation about the coating integrity. A chisel was hammered
K. Prasad Rao et al. / Journal of Materials Processing Technology 212 (2012) 402– 407 403
Fig. 1. Different stages in friction surfacing.
– IR
aitTioCtt
iitcRtezc(oaD
Ndufiasrtiwc
oi
perature range of 100–1450 ◦C (and an integration time of 30 �s)was used. Therefore, in the current study, the starting temperaturerecorded was 100 ◦C.
Fig. 2. Experimental set up
gainst the small vedge which formed at the region where the coat-ng gets completed. If the coating was not adequately bonded thenhe coating would come off from the substrate during hammering.hough the chisel test does not give any quantitative information,t can give a basic understanding about the mechanical propertiesf the coating. For tool steel D2 rod on steel and CP Cu rod on CPu, parameters resulting in metallurgical bonding based on chiselest were selected. Button shaped coatings were made for infraredhermography study.
Surface temperature profiles generated at the rod/substratenterface were recorded using an infrared camera (Fig. 2). Thenfrared camera used in this study was calibrated to measureemperature between 100–1450 ◦C. The detail of infrared cameraalibration is available in an earlier published work done by Khalidafi et al. (2011). The same calibration method was adopted andhe same IR camera was used for this work. The calibrated IR cam-ra was placed at an appropriate distance (0.5 m) from surfacingone and focused to the surfacing region. The CEDIP JADE Mer-ury Cadmium Telluride (MCT) camera was operated in the LWIRLong Wave Infrared) band (7.9–9.7 �m) with a focal plane arrayf 320 × 240 detectors (each detector is a pixel of size 25 �m) and
pixel pitch of 30 �m. The mean Noise Equivalent Temperatureifference (NETD) of the camera was 20 mK.
Once the field of view was set, the IR camera was subjected toon-Uniformity Correction (NUC). This was done by exposing theetectors first to a black body at room temperature and then to aniformly heated plate. This procedure ensures same energy levelsor all the detectors before the start of the experiment. The thermalmages were captured as continuous frames. Thermal images can becquired rapidly and a rate of 200 frames/s (fps) was used to ensureufficient sampling of the temperature values. The higher frameate effectively nullifies the diffusivity effects at higher operatingemperatures. An integration time (the time required for processingndividual frames) of 30 �s was set and lens of focal length 25 mm
as used to capture the thermal images. The image data can be
alibrated and expressed as location dependent temperature.
There are different calibration files available for measurementf different temperature ranges and integration times for captur-ng thermal images using LWIR camera and processing by Altair
camera, rod and substrate.
software. For the current study, a calibration file having a tem-
Fig. 3. (a) Interface microstructure between D2 coating and steel substrate – SEM-BSE. (b) Rubbed surface of tool steel D2 rod compared to the original rod surface –rough surface indicates metal transfer and coating formation.
404 K. Prasad Rao et al. / Journal of Materials Processing Technology 212 (2012) 402– 407
D2 ro
3
1
due to plastic deformation (Meola and Carlomagno, 2004).Once the temperature reached to 900 ◦C, the friction compo-nent of heat generation get diminished and gradually taken
Fig. 4. Thermal profile obtained from the interface of tool steel
. Results and discussion
. Tool steel D2 consumable rod on low carbon steel substrateCoatings could be obtained with tool steel D2 rod on low
carbon steel substrate (Fig. 3) in a wide range of process param-eters (10–20 kN and 600–1600 RPM). Fig. 3a shows interfacemicrostructure between coating and substrate. Defect free andcontinuous bonding was observed. Chisel test showed no de-bonding at the interface of coating/substrate. The rubbed surfaceof tool steel D2 rod on steel is shown in Fig. 3b. It showed a veryrough surface indicating metal transfer from the tip of the rodforming a coating on the substrate. Fig. 4 shows typical thermalprofile (time–temperature plot) obtained during coating forma-tion. A thermal image obtained at a time period of 65 s is alsoshown on the left side of Fig. 4.
Four stages were noticed in the above thermal profile.1. Stage 1 consisted of a slow and gradual increase in temper-
ature (from room temperature to ∼200 ◦C) with respect todwell time (0 to ∼18 s).
2. Stage 2 consisted of sudden rise in temperature from ∼200 to900 ◦C (18 to ∼22 s).
3. Stage 3 consisted of gradual rise in temperature from 900 to∼1075 ◦C (22 to ∼45 s).
4. Stage 4 attains a steady state temperature (∼1075 ◦C) duringwhich the coating starts forming (45 s onwards).
First stage commences as soon as the two metal surfacescome into the contact. The initial friction between the surfaceswill be playing an important role in the temperature gener-ation during Stage 1. The heat generation during this stageis predominantly decided by the coefficient friction betweenconsumable rod and the substrate. Higher the coefficient offriction (COF), higher the rate of heat generation.
After a gradual increase in temperature (from room tem-perature to 200 ◦C) during Stage 1, the temperature increased
to ∼900 ◦C during Stage 2. At the end of Stage 1, the oxidelayers would have been removed from the faying surfacesleading to the formation of two nascent surfaces which comesinto an intimate contact. The interface experiences a sudden
d/steel substrate and the thermal image (left) obtained at 65 s.
increase in the torque and “sticking and slipping” phenomenamay take place where the surface asperities will get weldedand be broken. To overcome the effect of the torque, the sys-tem has to generate large amount of energy leading to severeplastic deformation at the interface. The heat generates ina much faster rate than heat dissipation, resulting to a con-dition of adiabatic heating. The sudden rise in temperatureduring Stage 2 can be attributed to viscous heat dissipation
Fig. 5. Rubbed end of CP Cu rod-smooth surface indicating no metal transfer.
K. Prasad Rao et al. / Journal of Materials Processing Technology 212 (2012) 402– 407 405
Fig. 6. Thermal profile obtained from the interface of CP Cu rod/steel substrate and the thermal image (left) obtained at 15 s.
over by the viscous heat dissipation (Stage 3). This gradualtransition from friction and viscous heat dissipation to com-plete viscous heat dissipation for heat generation is depictedby the slope of the temperature profile at Stage 3 (between900 and 1075 ◦C). When the peak temperature of 1075 ◦C wasattained, the process reached a steady state value which canbe termed as Stage 4 where coating was formed. The plas-tic deformation initiated during Stage 2 attains a steady stateduring Stage 4 with continuous generation of heat till the endof the process.
2. CP Cu rod on low carbon steel substrateIn the entire range of process parameters used, CP Cu rod
could not be coated on low carbon steel substrate. The rubbedsurface of CP Cu on steel is shown in Fig. 5. It can be noted thatthe rubbed surface of rod deformed at the tip and formed flashrather than coating. The smooth surface at the rubbing end ofthe rod is another indication that there was no metal transferfrom rod to the substrate. This was in contrast to the rubbedsurface of tool steel D2 rod (Fig. 3b), which was very roughindicating metal transfer and coating formation.
The thermal profile for this system (obtained with 6 kN loadand 1500 RPM) is shown in Fig. 6 with a thermal image takenat 15 s. The thermal profile showed three different stages, sim-ilar to that of tool steel D2 coating on steel substrate. DuringStage 1, the temperature of the rod tip rose to ∼600 ◦C in ∼14 s.During the end of Stage 2, peak temperature of ∼975 ◦C wasattained within a short time of ∼1.5 s. During Stage 3, the sys-
tem attained a steady state value of ∼975 ◦C. Despite such ahigh temperature attained, CP Cu coating could not be formedon steel substrate. This could be attributed to the sudden
reduction in the flow stress of copper at elevated tempera-ture, which is less than that of the localized stress due to axialload. At the maximum temperature obtained during Stage 3(975 ◦C), CP Cu will have very low flow stress (ASM Handbook,2006). When the material does not exhibit enough flow stressto involve in “stick and slip mechanism”, it easily deforms atthese applied axial load and coating will not form.
3. CP Cu rod on CP Cu substrateIn contrast to friction surfacing of CP Cu on steel substrate,
a coating could be obtained when CP Cu rod was frictionsurfaced over CP Cu substrate though in a narrow range ofprocess parameters (6 kN axial and 1500 RPM). Fig. 7a showsinterface microstructure between CP Cu coating and CP Cusubstrate. Continuous bonded area with no gross defects wasobserved. No de-bonding was noticed at the interface of coat-ing/substrate after the chisel test. The rubbed surface of CPCu rod is shown in Fig. 7b. The rough surface of the rubbedsurface indicates that there was a metal transfer from the tipof the rod forming a coating on the substrate.
The thermal profile obtained is shown in Fig. 8 with a ther-mal image taken at 19 s. During Stage 1, the heat generationwas lower compared to previous systems. The maximum tem-perature generated was ∼120 ◦C. During Stage 2, like previouscases, there was an increase in temperature reaching ∼700 ◦C.The steady state temperature of 700 ◦C was maintained dur-ing Stage 3. Unlike the friction surfacing of CP Cu rod oversteel, the temperature did not exceed 700 ◦C during Stage 3.
This could be attributed primarily to the high thermal con-ductivity of both consumable rod and the substrate materials.The rapid heat conduction through both the rod and substrate
406 K. Prasad Rao et al. / Journal of Materials Processing Technology 212 (2012) 402– 407
Table 1Steady state temperature during Stage 3 and forging temperatures of rod materials.
Rod-substrate Steady state temp(◦C)
Forging temp of rodmaterial (◦C) (ASMHandbook, 2006)
Hot extrusion coatingformation temp of rod material(◦C) (ASM Handbook, 2006)
Coating formation
D2 steel on steel 1075 1050–1100 – Formed
FSf
CP Cu on steel 975 730–845
CP Cu on CP Cu 700 730–845
materials would have resulted in avoiding temperature buildup at the interface. CP Cu will exhibit sufficient flow stressto cause shearing of plasticized layers and formation of sur-face coating at this temperature (ASM Handbook, 2006). Theshearing of rubbed surface after the surfacing process can be
seen from Fig. 6b indicating metal transfer.
ig. 7. (a) Interface microstructure between CP Cu coating and CP Cu substrate –EM-BSE. (b) Rubbed surface of CP Cu rod-rough surface indicating metal transferorming coating.
800–950 Not formed800–950 Formed
3.1. Mechanism of friction surfaced coating formation
Whether or not consumable rod forms a coating on the substratedepends on it’s flow stress. The flow stress of the consumable rodis a function of local strain rate and temperature (ASM Handbook,2006). Flow stress decreases significantly with both strain rateand temperature. Friction surfacing process involves relatively highstrain rates. Based on the study by Johnson et al. (1983) on responseof various metals to large torsional strains over a large range ofstrain rates, the strain rate in friction surfacing is expected to beabout 60 s−1. During friction surfacing process relatively high tem-peratures are generated, close to that of melting points (Khalid Rafiet al., 2011). The hot plasticized metal that lies in between the rodand substrate in friction surfacing is similar to a hot metal that issubjected to hot extrusion or hot forging. Friction stir welding isoften described as a combination of in situ extrusion and forgingprocess (Chang et al., 2004). If the flow stress of the hot plasticizedmaterial becomes too low due to higher temperature, the excessiveplasticization of the material will result in an uncontrolled defor-mation. Therefore, both in hot extrusion or hot forging processes,the temperature of material is maintained in a particular range forcontrolled deformation to take place.
In the friction surfacing technique also, the hot plasticized layeris compressed in between the rod and substrate. In this case too,a large reduction in flow stress due to exposure to very high tem-peratures is undesirable to get a surface coating. The plasticizedmaterial at the interface should posses some amount of flow stress,which should at least match with the localized stress experienceddue to axial loading. If the flow stress of plasticized material is lowfor a given temperature, then it extrudes out in the form a flash.If the flow stress is comparable with the localized stress experi-enced, the material will experience some resistance to plastic flowand will interact with the substrate to form a coating.
In the current study it was found that tool steel D2 coating couldbe coated on steel substrate with ease, because at the steady statetemperatures obtained (1075 ◦C), D2 steel had enough flow stressto offer resistance to plastic flow and interact with the substrate toform a coating. However, CP Cu could not form a coating on steelsubstrate. This can be attributed to high temperature that was gen-erated (∼975 ◦C), where flow stress of CP Cu will be relatively lowoffering no resistance to plastic flow. When CP Cu substrate wasused in place of steel substrate, temperature at the interface couldbe maintained at ∼650 ◦C and it had enough flow stress to offerresistance to plastic flow and formed a coating. Therefore, it can beinferred that the primary requirement for obtaining a friction sur-faced coating is to maintain sufficient flow stress for the plasticizedmaterial. Incidentally, the steady state temperature attained duringStage 3 of systems where coating was obtained (D2 steel coatingon steel as well as CP Cu on CP Cu), was close to the forging and hotextrusion temperatures of the materials (Table 1) (ASM Handbook,2006).
Therefore, the study indicates that by maintaining the steady
state temperature during Stage 3 close to that of the rod materialhot extrusion or hot forging temperatures, various coatings can beobtained. Methods can be adopted to maintain the flow stress witha prior knowledge of deformation temperatures.
K. Prasad Rao et al. / Journal of Materials Processing Technology 212 (2012) 402– 407 407
rod/Cp
4
1
2
A
bt
R
B
A
B
Fig. 8. Thermal profile obtained from the interface of CP Cu
. Conclusions
. Under the experimental conditions used, metallurgically bondedfriction surfaced coatings of tool steel (D2) on steel substrate andCP Cu on CP Cu substrates could be obtained.
. 4 stages were noticed in the IR thermal profiles obtained duringfriction surfacing.a) Stage 1 consisted of a slow and gradual increase in tempera-
ture with respect to dwell time.b) Stage 2 consisted of a sudden jump in temperature with in a
very short dwell time followed by gradual rise in temperature(Stage 3).
c) Stage 4 consisted of steady state temperature during whichcoating formed/not formed.
3. Under the applied experimental parameters, there seems tobe a relation between the feasibility of coating formation andthe steady state temperature during Stage 3.
4. It was found that a metallurgically bonded coating can beobtained if the flow stress of the plasticized material is compa-rable with the localized stress developed due to axial loading.
5. By maintaining the steady state temperature during Stage 3close to that of the rod material hot extrusion or hot forgingtemperature, various coatings can be obtained.
cknowledgements
The authors gratefully acknowledge the financial help renderedy Naval Research Board, DRDO, Government of India in supportinghis work.
eferences
auser, M., Sauer, G., Siegert, K. (Eds.), 2006. ASM Handbook, vol. 2, 2nd ed. ASMInternational, Materials Park, OH 44073-0002, p. 917 (vol. 14, p. 566).
runsankar, M., 2010. Feasibility studies of friction surfacing on non-ferrous sub-strates, M Tech Thesis, Indian Institute of Technology Madras, Chennai 600036.
edford, G.M., Vitanov, V.I., Voutchkov, I.I., 2001. On the thermo-mechanical
Cu substrate and the thermal image (left) obtained at 19 s.
events during friction surfacing of high speed steels. Surf. Coat. Technol. 14,134–139.
Bruggemann, G., Mahrle, A., Benziger, T., 2000. Comparison of experimental deter-mined and numerical simulated temperature fields for quality assurance at laserbeam welding of steels and aluminium alloying. NDT&E Int. 33, 453–463.
Chang, C.I., Lee, C.J., Huang, J.C., 2004. Relationship between grain size andZener–Holloman parameter during friction stir processing in AZ31 Mg alloys.Scripta Mater. 51, 509–514.
Huang, R.S., Liu, L.M., Song, G., 2007. Infrared temperature measurement and inter-ference analysis of magnesium alloys in hybrid laser-TIG welding process. Mater.Sci. Eng. A 447, 239–243.
Johnson, G.R., Hoegfeldt, J.M., Lindholm, U.S., Nagy, A., 1983. Response of variousmetals to large torsional strains over a large range of strain rates—I: ductilemetals. J. Eng. Mater. Technol. 105, 42–47.
Khalid Rafi, H., Janaki Ram, G.D., Phanikumar, G., Prasad Rao, K., 2010a. Frictionsurfaced tool steel (H13) coatings on low carbon steel: a study on the effects ofprocess parameters on coating characteristics and integrity. Surf. Coat. Technol.205, 232–242.
Khalid Rafi, H., Janaki Ram, G.D., Phanikumar, G., Prasad Rao, K., 2010b. Microstruc-ture and properties of friction surfaced stainless steel and tool steel coatings.Mater. Sci. Forum 638–642, 864–869.
Khalid Rafi, H., Balasubramaniam, K., Phanikumar, G., Prasad Rao, K., 2011. Thermalprofiling using infrared thermography in friction surfacing. Met. Mater. Trans.42A, 3425–3429.
Liu, X.M., Zou, Z.D., Zhang, Y.H., Qu, S.Y., Wang, X.H., 2008. Transferring mech-anism of the coating rod in friction surfacing. Surf. Coat. Technol. 202,1889–1894.
Madhusudhan Reddy, G., Srinivasa Rao, K., Mohandas, T., 2009. Friction surfacing:novel technique for metal matrix composite coating on aluminium–silicon alloy.Surf. Eng. 2, 25–30.
Marinetti, S., Vavilov, V., 2010. IR thermographic detection and characterization ofhidden corrosion in metals: general analysis. Corros. Sci. 52, 865–872.
Mathieu, A., Mattei, S., Deschamps, A., Martin, B., Grevey, D., 2006. Temperaturecontrol in laser brazing of a steel/aluminium assembly using thermographicmeasurements. NDT&E Int. 39, 272–276.
Matte, S., Grevey, D., Mathieu, A., Kirchner, L., 2009. Using Infrared thermographyinorder to compare laser and hybrid (laser +MIG) welding processes. Opt. LaserTechnol. 41, 665–670.
Meola, C., Carlomagno, G.M., 2004. Recent advances in the use of Infrared thermog-raphy. Meas. Sci. Technol. 15, 27–58.
Pastor, M.L., Balandraud, X., Grédiac, M., Robert, J.L., 2008. Applying infrared ther-mography to study the heating of 2024-T3 aluminium specimens under fatigue
loading. Infrared Phys. Technol. 51, 505–515.
Sreenu, May 2010. Feasibility studies of friction surfacing on non-ferrous substrates,M Tech Thesis, Indian Institute of Technology Madras, Chennai 600036.
Thomann, H., Frisk, B., 1968. Measurement of heat transfer with an infrared camera.Int. J. Heat Mass Transfer 2, 819–826.
