Surface and Coatings Technology, 64 (1994) 173—181
Influence of plasma nitriding on wear performance of TiN coating* M. Zlatanoviéa, D. Kakas”, Lj. Mazibrada”, A. Kunosiéa and W.-D. Münz’~ aFaculty of Electrical Engineering, Bulevar Revolucije 73, P0 Box 816, 11001 Belgrade (Yugoslavia) bFacUlty of Technical Science, Institute of Production Engineering, V. Perica Valtera 2, 21000 Novi Sad (Yugoslavia) aHauzer Techno Coating Europe, Groethofstraat 22h, NI~5916PB Venlo (Netherlands) (Received April 20, 1993; accepted in final form November 12, 1993)
Abstract The wear properties of uncoated, TiN-coated and plasma-nitrided—TiN-coated samples made ofsteel grades AISI M2 and AISI 4140 were studied. A wheel 50mm in diameter made of cementation steel grade AISI 5115 was used as a countermaterial for sliding wear tests against samples made in cubic form. The sliding speed was 56.5 m min’ and the normal load used was 10 or SON. Under the conditions used in these experiments it was found that plasma-nitrided—TiN-coated samples have superior wear resistance. The wear mechanism depends strongly on the surface treatment of the samples. Abrasive wear dominates in the case of uncoated samples, while in the case of TiN-coated samples cracking and plucking of parts of the coating dominate in the initial stage of the wear test and no gradual wear of the coating itself was found. The formation of a protective glassy layer on TiN coatings deposited on high speed steel substrates was observed.
I. Introduction Many laboratory and mass production tests have been performed in order to investigate the wear and friction properties of various wear-resistant coatings deposited on tool steel substrates , carbide inserts  or other substrate materials. TiN coatings have been widely tested and are sometimes used as a reference case for the comparative study of wear resistance . The coating properties, as well as the properties of the substrate material and the conditions at the interface, influence the tribological properties at the sample material— countermaterial contact surface. Combined plasma nitriding—physical vapour deposition (PVD) surface treatment has been shown to have beneficial effects on the coating-to-substrate adhesion , the load-bearing capacity of substrate materials  and some cutting tool performances . The results of various wear tests have shown that a composite TiN-coating—nitrided layer exhibits excellent wear resistance in the case of several tribocouples of different materials [5, 7]. The enhancement of tool life of some cutting tools was also observed after combined nitriding—coating surface treatment. In the case of forming tools for backward extrusion, the wear conditions are very complex owing to the high load at the tool—
*paper presented at the 20th International
Metallurgical Coatings and Thin Films, San Diego, CA, USA, April 19—23, 1993.
workpiece contact surface . During this formi: process, sliding without lubrication at the tool—wor piece contact surface under high specific load conditio causes severe wear of the tool material which can reduced by plasma nitriding . In the case of substrat made of structural steel with relatively low hardness the base material, the application of combined surfa treatment may reduce the wear intensity of some trib logical pairs. Various wear tests and various criteria for measuri: wear intensity have been used in the experiments do so far. In ref. 5 pure sliding ball-on-wheel tests we successfully carried out in order to investigate the perfc mance of the composite nitrided—TiN layer, while results of rolling—sliding wear tests were reported ref. 10. Some new methods for testing the tribologi properties of hard coatings have also been propos . 2. Experimental details In the experiments described below an Amsler testi; machine was used for testing the wear properties samples untreated and treated by plasma nitriding at plasma deposition of TiN. Wheel-on-flat surface we tests were performed in which a wheel made of cement tion steel AISI 5115 carburized and heat treated to Rockwell hardness of 60 ±1 HRC was run against t test specimen. The wheel was 50 mm in diameter wi a contact surface radius of 120 mm (Fig. 1(a)). T specimen dimensions were 38 x 38 x 10 mm3. Two th
M. Zlatanovk~et al.
Wear performance of TiN coating
3. Results and discussion -
3.1. TiN coating structure and adhesion
The chemical compositions of the sample materi and countermaterial are given in Table 1. The samç made of steel grade AISI M2 were quenched at 1200 and tempered at 550 °C.Table 2 lists the Rockwell c hardness and Vickers surface microhardness data
untreated and surface-treated samples. The TiN coat I
thickness was approximately the same in all ca
q ~ ~
(3 The ±0.5fractured i-tm)onasAISI measured by AIS1 the calotest. deposited cross-sections M2 and of 4140 thesteel TiNgrade coatii si strates are shown in Figs 2(a) and 2(b) respectively. both cases the coating structure is columnar and v dense. In the case of the HSS substrate the surf~ topography is smooth (Fig. 2(a)) and the interfi
Fig. 1. (a) Shape of wheel made of cementation steel AISI 5115 used as countermaterial. (b) Test configuration: 1. countermaterial; 2. sample.
mocouples were mounted on one side of the specimen in order to measure the temperature during wear testing, and a special mechanism for adjustment and control of the load at the specimen—wheel contact surface was used (Fig. 1(b)). The specimens were made of two steel grades: high speed steel (HSS) AISI M2 quenched and tempered to a hardness of 65 HRC and structural steel AISI 4140 quenched and tempered to 40 HRC. To investigate the influence of the initial surface roughness of specimens on the wear properties, two specimen surfaces were ground and the other two polished before coating deposition. in these experiments the wheel was used as a countermaterial for tribocouple testing. The samples were plasma nitrided in a laboratory-scale plasmanitriding unit and the TiN coating was deposited in an industrial-size arc ion-plating equipment. The wheel was slid against the sample surface at loads of 10 and 50 N, while the sliding speed was kept constant in all experiments at 56.5 m mm During testing, three experiments (with uncoated, TiN-coated and plasmanitrided—TiN-coated samples) were performed for both substrate materials (AISI M2 and AISI 4140 steel grades). The wear scars are found to be elliptical in form. Instead of measuring (calculating) the volume of the sample material worn out during testing, the change in one of the ellipse axes was used as a wear intensity criterion and plotted against the testing time. ~.
indicates very good coating-to-substrate adhesion. 1 column boundaries are not pronounced and in so places cohesive failure of the coating has occurred. 1 TiN coating deposited on the AISI 4140 substrate also columnar, with the columns growing continou from the substrate to the coating surface (Fig. 2(b)). cohesive failure of the coating was observed. The coating-to-substrate adhesion was investigated the scratch test method. The acoustic emission (A signal was compared with metallographic examinati of the scratch channel in order to measure the criti load for coating failure. In Fig. 3(a) the acoustic emissi signal during scratching of the TiN coating deposi on the structural steel sample is shown. Metallograp investigation shows the critical load for coating faili to be 26 N. Despite a relatively soft substrate, the st of coating failure was well defined. The coatingsubstrate adhesion represented by the scratch test criti load was found to be better in the case of the HSS st substrate (Fig. 3(b)). The mean value of the critical Ic calculated from the three scratches in the case of HSS substrate was L=42 N. From previous experiments  it was found that careful control of the plasma-nitriding process parar ters a composite TiN coating—plasma-nitrided la could be obtained on several steel grade samples w improved coating-to-substrate adhesion. The scratch results (acoustic emission signals) obtained in the c of the plasma-nitrided--TiN coating surface structure shown in Fig. 4. The critical load was enhanced in b cases, i.e. HSS and AISI 4140 steel substrates, amount] to 78 and 34 N respectively. 3.2. Wear characteristics and wear mechanisms
Preliminary wheel-on-flat surface tests done with s eral tribological couples of materials with known w characteristics have shown that both axes of the ellipti wear scar increase with increasing sliding distance. 1
M. Ziatanovid et al.
Wear performance of TiN coating
TABLE 1. Chemical compositions (weight per cent) of sample materials (balance Fe) Steel grade
5115 4140 M2
0.14—0.19 0.38—0.45 0.82
0.80—1.10 0.90—1.20 4.00
TABLE 2. Core hardness and surface microhardness values of samples Steel grade
M2 M2 4140 4140
Type of layera
TiN PN+TiN TiN PN+TiN
Core hardness (HRC)
3048 3075 1560 1820
1892 2734 1175 1498
65 65 40 40
ap~ plasma-nitrided samples.
(hi 1g. 2. Scanning electron micrographs of TiN-coated samples: (a) AISI M2; (b) AISI 4140.
wear scar area can be directly related to the wear volume. This is made possible by the use of the length of one axis as an indication of wear instead of measuring an actual wear rate. Generally speaking, the shapes of the elliptical wear scars were regular, except in the case of plasma-nitrided and coated substrates and TiN-coated HSS substrates, in which cases many irregularities were found, The same countermaterial wheel was used in all experiments, running against uncoated, TiN-coated and plasma-nitrided--TiN-coated HSS and AISI 4140 steel samples. The sliding speed was kept constant at 56.5 m min~. The wear scar width as a function of testing time during wheel-on-flat substrate tests is shown in Fig. 5
for uncoated, TiN-coated and plasma-nitrided--Ti coated AISI 4140 samples. The samples were tested a contact load of 10 N in the case of ground and polish surfaces (Fig. 5(a)). Plasma-nitrided samples sho~’t better wear resistance than non-nitrided samples. T initial roughness at the sample surface was found influence the wear resistance. The wear rate was reduc in the case of ground samples as compared with p ished surfaces. With an increased contact load of 50 N (at the sai sliding speed) and for surface-ground samples, plasn nitrided—TiN-coated specimens showed superior wt resistance (Fig. 5(b)). This effect could be partially d to the enhanced load-bearing capacity of plasrr nitrided samples.
M. Zlatanovié et al. / Wear performance of TiN coating
Fig.3. AE signals during scratch tests: (a) AISI M2; (b) AISI 4140.
PN+TIN Fig. 5. Wear scar width as a function of test time for uncoated, Ti coated (TiN) and plasma-nitrided—T1N-coated (PN+TiN) AISI 41 AIS
samples: (a) ground surface, load 10 N; (b) ground and polished surfa load SON.
—~---———-——~o~—~tal conditions, the two substrate with the same countermaterial
Fig. 4. AE signals during testing of plasma-nitrided--TiN-coated AISI M2 and AISI 4140 samples.
The wear mechanisms of untreated samples were found to be similar to those obtained in other laboratory tests . At the low sliding speeds used in these experiments, abrasive wear dominated for HSS samples, while superficial plastic flow of material was found in the case of AISI 4140 steel. The detailed wear mechanisms of TiN-coated HSS and structural steels are not yet completely known and in general there is a lack of data concerning the wear of plasma-nitrided—TiN-coated steel samples. Since the wear processes are strongly dependent on the experimen-
materials in conta will be considerm
separately. 3.2.1. Wear of TIN-coafed and plasma-nitrided-TiN-coated AISI 4140 steel samples In the case of carburized steel wheel vs. TiN-coab AISI 4140 substrates, scanning electron microscol (SEM) and optical microscopy were used for surfa wear analyses. At a contact load of 50 N and after 5 mm of testin a difference in surface topography at the front and re. parts of the scar was found on the TiN-coated samp] On the front part of the wear scar cracking of tI coating in semicircular form is visible (Fig. 6(a)). TI cracks are predominantly due to tensile stresses intr duced during sliding of the sample against tI
U~‘arpcrf ‘rinanec of Ii\ coating
.\i. Llaran( rh’ ei a!.
Fig. 6. %~earmechanisms of FiN—coated AISI 4140 samples: (a) front part of wear scar. L= 50 N. test time 5 mm; ( h) periphcr~of front scar 7C (C) central part of front scar lone: (d ) metaIIogr~tphtecross—section of central ione of wear scar: )e) plastic flow of inateri~tlin central ione; (f ) /one of the ssear scar (200
]\~I. Zlatanom:u/ ci a)., / Wear perjorismane of TiN coating
countermaterial wheel. SEM images of the sample crosssection (Figs. 6(b) and 6(c)) show the dominant wear mechanisms at the front of the wear zone. Near the periphery of the front zone (Fig. 6(b)) cracks in the coating are visible and fragments of the coating are embedded in the substrate material. In the central part of the front zone cracking of the coating and plastic deformation of the substrate material in which the coating fragments are embedded are more intensive ( Fig. 6(c)), while in the middle of the wear scar fragments of coating are detached from the surface together with the substrate material, as can be seen from Fig. 6(d) taken on the cross-section of the sample. The intensive (severe) abrasive wear and superficial plastic flow of material are dominant wear mechanisms (Fig. 6(e)). In the rear zone of the wear scar the TiN coating is concealed by adhered substrate and coating material
and the actual surface topography is not visii(Fig. 6(f)). The grooves in the sliding direction we formed by adhered material. In the SEM image of t cross-section nearly unaffected TiN coating und adhered material on the rear edge of the worn ar is visible. After plasma nitriding and subsequent TiN coatii deposition, the wear mechanism is partially altered. the very front edge zone of the wear scar some initi cracking of the TiN coating is visible ( Fig. 7 (a)) follow by the formation of the fractured fragments of coati; that are to be compressed against the plastica] deformed substrate material (Fig. 7(b)). A quantity tribocouple material has adhered to the coating. T fractured parts of the coating are further compress in the plastically deformed plasma-nitrided substre material, but no coating delamination is visit
Fig. 7. Wear mechanisms of plasma-nitrided—TiN-coaled \ISI 4140 samples: (a) very front edge ion~of wear scar: (h) front edge zone of w scar; (C) coating fracture wiihout delamination: (d) wol ml material adhered to coatmng in ream edge ,one.
M. Zlatanovk~et al.
Wear performance of TiN coating
(Fig. 7(c)). Again on the rear edge of the wear zone the adhered material covers the coating (Fig. 7(d)). The wear mechanism of the plasma-nitrided—TiN-coated sample is basically the same as that of the TiN-coated sample, except that the load-bearing capacity of the substrate material is enhanced owing to plasma nitriding and the plastic deformation of the substrate material is reduced. 3.2.2. Wear of TiN-coated and plasma-nitrided— TiN-coated HSS samples On the wear scar of the TiN-coated HSS substrates three zones are visible: two zones at the periphery with no cracks in the coating and a middle zone with a fractured coating (Fig. 8(a)). At the wear scar periphery grooves were formed in the sliding direction owing to
abrasive wear, while towards the central zone of I wear scar a glassy layer is visible (upper right side Fig. 8(b)). This glassy layer acts as a protective layer I the TiN coating and is composed of both substr~ material and countermaterial . In the central zo of the wear scar semicircular cracks are visible, w fragments of the coating compressed into the substr~ material (Fig. 8(c)). The fractured coating parts detached from the substrate in the places where glassy layer has formed and abrasive wear of the s~ strate material starts to appear. At the rear edge of I wear scar a step is formed on the surface between I TiN-coated surface and the surface with detached pa of the coating embedded (Fig. 8(d)). No gradual w of the TiN coating was found. After plasma nitriding of the HSS steel substrate a
Fig. 8. Wear mechanisms of TiN-coated HSS samples: (a) wear scar surface; (b) periphery ofwear scar; (c) central part of wear scar; (d) rear e of wear scar.
M. Zlatanovki et al. / Wear performance of TiN coating
TiN coating deposition, cracks in the coating were observed, as well as a quantity of tribocouple material adhered to the coating surface (Fig. 9(a)), but no intensive fracture of the coating and compression of the coating fragments to the substrate material were found. Instead, wear of the TiN coating itself was found in the central and rear edge areas of the wear scar (Figs. 9(b) and 9(c) respectively). This effect is due to the enhanced load-bearing capacity of the substrate material and improved coating-to-substrate interfacial adhesion which is sufficient to sustain high interfacial shear stresses during the wear test. 4. Conclusions In wheel-on-flat surface wear tests a wheel made of AISI 5115 steel grade was used as a countermaterial for
studying the wear mechanisms of uncoated, TiN-coat and plasma-nitrided—TiN-coated AISI M2 and Al 4140 steel samples. The wear mechanism of uncoat samples was found to be similar to that observ in other laboratory wear tests. In the case of TiN-coated AISI 4140 samples cracki; of the coating and compression of the coating fragmer in the plastically deformed substrate material we observed, followed by plucking of the coating fragmer and severe abrasive wear of the substrate material. Plasma-nitrided--TiN-coated AISI 4140 steel sampl showed less plastic deformation of the substrate mater] owing to the enhanced load-bearing capacity and cont quently reduced wear intensity. No gradual wear of coating was observed. During wear testing of TiN-coated HSS samples abi sive wear at the wear scar periphery and cracking at plucking of the coating in the central zone of the we
5 ins (a)
1-1g. 9. Wear mechanisms of plasma-nitrided—TiN-coated HSS samples: (a) front edge of wear scar: I b) central part of wear scar; (c) rear part wear scar.
M. Zlatanovié et a).
Wear performance of TiN coating
scar were found. In some places glassy layers were formed. After plasma nitriding and subsequent deposition of the TiN coating, the wear intensity was reduced owing to the enhanced coating-to-substrate adhesion and load-bearing capacity of the substrate material, and for the first time wear of the TiN coating itself was observed. Plasma-nitrided—TiN-coated samples have shown superior wear characteristics compared with uncoated and TiN-coated samples.
References 1 5. E. Franklin and J. Benger, Surf. Coat. Technol., 54—55 (1992)
2 W. Koning and R. Fritsch, Surf Coat. Technol., 54—55 (19 3 S. K. Glash and M. S. Kohler, Surf Coat. Technol., 54(1992) 466. 4 M Zlatanovic, Surf. Coat. Technol., 48 (1991) 19. 5 Y. Sun and T. Bell, Mater. Sci. Eng. A14O (1991) 419. 6 M. Zlatanovic and W.-D. Munz, Surf. Coat. Technol., 41(1990) 7 M. Zlatanovic, Ion Nitriding/Carburising Conf., ASM Internatior Metals Park, OH, 1989, p. 99. 8 D. Kakas, D. Lupuljev and M. Zlatanovic, Proc. First mt. Semi, on Plasma Heat Treatment, Senlis, 1987, L’Association Technic de Traitment Thermique, Paris, 1987, p. 353. 9 D. Kakas and M. Ziatanovic, Ion Nitriding/Carburizing, M International, Materials Park, OH, 1990, p. 141. 10 T. Bell and Y. Sun, Proc. 7th mt. Conf on Heat Treatment, Mosc 1990, in press. 11 P. Hedenqvist, M. Olsson and S. Soderberg, Surf Eng., 5 (1989) 141.