Influence of substrate pre-treatments by Xe+ ion bombardment and plasma nitriding on the behavior of TiN coatings deposited by plasma reactive sputtering on 100Cr6 steel

Influence of substrate pre-treatments by Xe+ ion bombardment and plasma nitriding on the behavior of TiN coatings deposited by plasma reactive sputtering on 100Cr6 steel

Materials Chemistry and Physics 177 (2016) 156e163 Contents lists available at ScienceDirect Materials Chemistry and Physics journal homepage: www.e...

2MB Sizes 10 Downloads 81 Views

Materials Chemistry and Physics 177 (2016) 156e163

Contents lists available at ScienceDirect

Materials Chemistry and Physics journal homepage: www.elsevier.com/locate/matchemphys

Influence of substrate pre-treatments by Xeþ ion bombardment and plasma nitriding on the behavior of TiN coatings deposited by plasma reactive sputtering on 100Cr6 steel S. Vales a, P. Brito b, F.A.G. Pineda a, E.A. Ochoa c, R. Droppa Jr. d, J. Garcia e, M. Morales c, F. Alvarez c, H. Pinto a, * ~o Paulo (USP), Escola de Engenharia de Sa ~o Carlos, Av. Trabalhador Sa ~o Carlense 400, Sa ~o Carlos, SP CEP 13566-590, Brazil Universidade de Sa lica de Minas Gerais (PUC-MG), Av. Dom Jos Pontifícia Universidade Cato e Gaspar 500, 30535-901 Belo Horizonte, MG, Brazil rio Zeferino Vaz, Bara ~o Geraldo, Campinas, SP CEP 13083-970, Brazil Universidade Estadual de Campinas (UNICAMP), Campus Universita d Universidade Federal do ABC (UFABC), Av. dos Estados, 5001, Santo Andr e, SP CEP 09210-580, Brazil e €gen 19, SE-12680, Stockholm, Sweden Sandvik Coromant R&D, Lerkrogsva a

b c

h i g h l i g h t s  Effect of Xe ion bombardment and plasma nitriding on TiN coatings was investigated.  Xe ion bombardment with 1.0 KeV increases nitrogen retention in plasma nitriding.  1.0 KeV ion impact energy causes sputtering, thus increasing surface roughness.  TiN coating wear is minimum after plasma nitriding due to lowest roughness.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 July 2015 Received in revised form 7 March 2016 Accepted 1 April 2016 Available online 15 April 2016

In this paper the influence of pre-treating a 100Cr6 steel surface by Xeþ ion bombardment and plasma nitriding at low temperature (380  C) on the roughness, wear resistance and residual stresses of thin TiN coatings deposited by reactive IBAD was investigated. The Xeþ ion bombardment was carried out using a 1.0 keV kinetic energy by a broad ion beam assistance deposition (IBAD, Kaufman cell). The results showed that in the studied experimental conditions the ion bombardment intensifies nitrogen diffusion by creating lattice imperfections, stress, and increasing roughness. In case of the combined pre-treatment with Xeþ ion bombardment and subsequent plasma nitriding, the samples evolved relatively high average roughness and the wear volume increased in comparison to the substrates exposed to only nitriding or ion bombardment. © 2016 Elsevier B.V. All rights reserved.

Keywords: Ion implantation Nitriding Ion beam-assisted deposition Tribology Wear

1. Introduction The performance of cutting tools and dies can be largely improved by surface modification techniques, such as plasma nitriding and hard coating deposition, which are widely applied to

* Corresponding author. E-mail addresses: [email protected] (S. Vales), [email protected] (P. Brito), [email protected] (F.A.G. Pineda), [email protected] (E.A. Ochoa), [email protected] (R. Droppa), [email protected] (J. Garcia), [email protected] (M. Morales), [email protected]fi.unicamp.br (F. Alvarez), [email protected] sc.usp.br (H. Pinto). http://dx.doi.org/10.1016/j.matchemphys.2016.04.010 0254-0584/© 2016 Elsevier B.V. All rights reserved.

steels leading to increase hardness, oxidation and corrosion resistance, wear resistance and overall superior tribological properties [1e5]. Plasma nitriding allows for excellent parameters control and high efficiency nitrogen diffusion at relatively low temperatures (300e450  C), which results in a more chemically uniform compound layer (“white layer”) and maintaining the material original hardness in comparison to conventional gas nitriding processes [6]. Typically, plasma nitrided steel work-pieces have a compound layer (“white layer”) formed at the surface which is composed by nitrides. This layer can be formed by ε-Fe2e3N and g0 -Fe4N or only a monolayer with one type of nitrides. Bellow of the compound layer there are a diffusion zone, which is composed by the a-phase

S. Vales et al. / Materials Chemistry and Physics 177 (2016) 156e163

saturated of nitrogen and the fine precipitates nitrides [7]. The nitrides ε-Fe2e3N and g0 -Fe4N can exhibit very high hardness, lower friction coefficient and good corrosion resistance, thus improving mechanical, chemical and tribological properties [8e11]. Deposition of hard coatings such as TiN, TiC, TiCN and TiAlN, which exhibit very high hardness, good wear resistance and low friction coefficient is also commonly utilized for enhancing surface characteristics of steel parts. The hard ceramic coating can be applied both directly on the metallic substrate as well as on a previously nitrided material [12e14]. In the latter case a duplex scale is formed with superior fatigue and wear resistance due to increased compatibility between the hard external coating and the strengthened underlying substrate [15,16], e.g. if the substrate is much softer than the hard coating, it might undergo plastic deformation in service causing the outer layer to collapse and fail prematurely [17]. Ion Beam Assisted Deposition (IBAD) is a Physical Vapor Deposition (PVD) process for hard coating deposition that may be combined with ion bombardment. The additional energy of the ion beam allows for coating deposition at lower temperatures, avoiding some of the difficulties associated with conventional PVD processes on nitrided substrates, such as bad adhesion, creation of porosities or decomposition of the nitride layer leaving a relatively soft iron layer behind [18,19]. The IBAD technique also allows for better densification of the hard coating in comparison to PVD processes, as well as independent control of several parameters [20,21]. The ion bombardment of metallic surfaces produces a variety of microstructural changes, such as the creation of lattice defects, phase transformations, material removal (sputtering) and amorphization on the annealed 100Cr6 steel [22]. Recent studies have demonstrated that the atomic bombardment of steel substrates with Xeþ ions improves the diffusion of nitrogen in the material by prompting stress, grain refinement, defects creation and diffusion channels in the 4140 (quenched and tempered) steel [23,25e27]. This process can also be used to modify the substrate surface on the atomic level by increasing the surface roughness and thus improving the adherence of hard coatings to the treated areas [7,28]. Thus, the development of processes which allow improving the performance of tools and dies can benefit from the understanding of pre-treating techniques for metallic substrates aiming at the subsequent deposition of ceramic coatings with enhanced hardness and thermal stability. In this work we report on individual and combined surface processing involving the bombardment with Xeþ ions and plasma nitriding at low temperature, i.e. 380  C, of annealed 100Cr6 steel substrates, as pre-treatment alternatives for the deposition of TiN hard coatings.

2. Materials and methods 2.1. Substrate material The substrate material was an annealed 100Cr6 bearing steel. Its chemical composition is given Table 1. The steel specimens were employed in the annealed instead of quenched and tempered condition in order to avoid phase transformations during processing by ion plasma nitriding. The samples used in all experiments were in the form of 20 mm diameter discs with 2 mm thickness. Prior to application of all surface modification techniques, the steel

Table 1 Chemical composition of the 100Cr6 steel. Element Composition [wt.%]

C 1.00

Si 0.25

Mn 0.35

Cr 1.5

157

samples were ground with SiC grinding paper, polished in 6, 3 and 1 mm diamond paste and finally polished in a 0.25 mm colloidal silica suspension. 2.2. Surface modification and TiN deposition Samples from the 100Cr6 steel were submitted to preliminary Xeþ ion bombardment and plasma nitriding prior to TiN film deposition. The Xeþ ion bombardment/implantation and TiN deposition processes were performed using an ion beam assisted deposition (IBAD) system located at the Ionic Implantation and Surface Treatment Laboratory of the Campinas State University (Unicamp, Brazil). The system consists mainly of a deposition chamber equipped with two Kaufman ion sources of 3 cm diameter, a support for four targets and a temperature controlled (<1000  C) sample holder. More details regarding this system can be found in the work of Hammer et al. [29]. For ion bombardment a xenon gas flux was introduced into the ~3 cm diameter ion beam Kaufman sources (9.9995%) and ions produced were accelerated to 1.0 keV. The samples were bombarded perpendicularly to the polished surface with a ~2.8 mA/cm2 current density for 30 min. During Xeþ bombardment, the substrate temperature was kept below 260  C and the work pressure in this case was 0.15 Pa. Plasma nitriding was performed using a 0.2 keV Nþ beam with ~2.7 mA/cm2 current density focused on the 100Cr6 sample surface for 30 min. The temperature of the process was set to 380  C and the working pressure was 0.015 Pa for all nitrided samples. The deposition of TiN coatings was accomplished by sputtering a titanium target with an argon ion beam of 1.45 keV (80 mA) in a N2 atmosphere. The deposition N2 pressure was 0.055 Pa and the substrate temperature was set to 400  C. The deposition time was 120 min for all the samples. All those processes were carried out independently in the following manner: initial sputter cleaning of the substrates using a 600 eV Arþ beam for 5 min, followed by Xeþ ion bombardment and/ or nitrogen implantation when applicable, and finally, TiN coating deposition. Thus, a series of four samples was produced by combining the processes described above yielding the samples described in Table 2. 2.3. Phase analyses and microstructure Angle-dispersive X-ray diffraction (XRD) using synchrotron radiation was carried out at the experimental station XPD of the Brazilian Synchrotron Radiation Laboratory (LNLS, Campinas) in order to assess the phase composition within the TiN coated samples subjected to the different surface pre-treatment procedures. The radiation energy was set to 10.5 keV (l ¼ 1.1823 Å) and the beam was 4 mm wide and 1 mm high. Diffractograms were obtained both under a grazing incidence angle (u) of 1, which corresponds to an average penetration depth t of 500 nm, and also using the symmetric qe2q mode. The microstructure of the 100Cr6 steel substrate and the TiN coatings was investigated by scanning electron microscopy (SEM) using a Philips XL-30 FEG microscope. The sample cross-sections were etched to reveal the material microstructure with 3% Nital

Table 2 Sample identification and structure. Sample

Sample structure

S SN S1000 S1000N

Steel Steel Steel Steel

þ þ þ þ

TiN nitrided layer þ TiN Xeþ bombarded layer þ TiN Xeþ bombarded and nitrided layer þ TiN

158

S. Vales et al. / Materials Chemistry and Physics 177 (2016) 156e163

solution. Transmission Electron Microscopy (TEM) of sample crosssections was used to characterize the size and distribution of the iron nitrides formed during plasma nitriding and to visualize the impact of atomic peening on the microstructure of the steel surface by using a JEOL 2100 microscope. Suitable thin samples were prepared applying a wedge shaped thin slice, i.e., polishing by gentle abrasive rubbing of a specimen with a slight tilt (0.3e0.7 ) to obtain an electron transparent wedge with smooth and highly polished surfaces (20 mm). Finally, electron transparent areas were then obtained by low energy ion polishing. 2.4. Topography, wear and coating adhesion The influence of each pre-treatment procedure on the surface topography of the coated samples was evaluated by atomic force microscopy (AFM) using the tapping mode in a FlexAFM 3 e Nanosurf equipment. To evaluate the adhesion of the coatings, an Anton Paar Nanoscratch-Tester (NST) was employed using a linear increase of the applied normal force up to 500 mN. The wear resistance of the TiN coatings deposited on the different substrate surfaces was evaluated by means of ball on disc tests. A ball of cemented carbide (WC-Co) with 3.0 mm radius and 5 N normal load were applied to the tests. The total sliding distance was set to 1000 m.

(II) a diffraction line (hkl) of the material under consideration is chosen, thus fixing the diffraction angle 2q ¼ 2qhkl; (III) a list of the azimuths 4 and the inclination j-angles has to be defined, taking into account the symmetry of the stress state, the crystallographic texture of the material and the required measurement accuracy; (IV) a list of diffractometer settings (u, c, f) has to be determined for each specimen direction (4, j) using Eqs. (1)e(3). Thus, in order to maintain a constant penetration depth during a sin2j-based stress analysis, the traditional c or u tilting, which are used individually at a certain 4-azimuth, has to be replaced by a coordinated movement of the diffractometer angles u, c and f. As a consequence of the controlled penetration depth, only a limited region within the sin2j-plot is accessible for a certain hkl reflection. For further details regarding this technique, the reader is referred to [30]. The lattice strains were measured for the {220} diffraction line of TiN. This reflection was chosen based on a compromise between the ease for separating this line from neighbouring ones, its intensity and the highest possible 2q-angle. The Diffraction Elastic €ner Constants (DEC) were calculated based on the Eshelby-Kro model using the single crystal elastic constants for TiN [31,32]. 3. Results and discussion

2.5. Residual stress analysis 3.1. Microstructure The effect of Xe bombardment and plasma nitriding on the average residual stresses of the TiN films was determined by applying the sin2j method at a fixed penetration depth of 500 nm using a four-circle diffractometer located at the experimental station XPD of the Synchrotron facilities (LNLS, Campinas, Brazil) [30,31]. The radiation energy was 10.5 keV (l ¼ 1.1823 Å) and beam dimensions were 1  1 mm. A fixed penetration depth was chosen in order to suppress the influence of stress gradients, which typically occur in thin surface films, on the stress evaluation procedure. Owing to the weak macroscopic texture of the IBAD deposited TiN films, unrestricted lattice spacing measurements could be performed for all sample directions defined by the inclination angle j and the azimuth 4. The penetration depth t of the synchrotron X-rays depends on the incident (a) and exit (b) angles of the incoming and diffracted Xrays, respectively. This can be also expressed as a function of the diffractometer angles according to Eq. (1):



sin a$sin b sin u$sinð2q  uÞ$cos c ¼ m½sin a þ sin b m½sin u þ sinð2q  uÞ

The microstructure of the 100Cr6 steel subjected to Xeþ bombardment followed by plasma nitriding and TiN deposition (sample S1000N) is presented in Fig. 1. The SEM image initially reveals that the annealed hypereutectoid steel substrate consists of a ferritic matrix with embedded spheroidized cementite particles. It can also be observed that the average thickness of the TiN coatings deposited by IBAD amounts to about 400 nm. Plasma nitriding conducted at 380  C did not lead to the formation of a white compound layer but only to a restricted diffusion zone, which cannot be visualized using SEM analyses. The results of the phase analyses conducted by XRD on the S, SN, S1000 and S1000N specimens are presented in Fig. 2. The XRD patterns of non-nitrided samples (S and S1000) contain reflections that correspond to ferrite and Fe3C from the substrate, as well as TiN from the deposited hard coating. For the SN sample, which was not subjected to Xeþ bombardment, the diffractograms reveal additionally the presence of the iron-rich g0 -Fe4N nitride. With

(1)

where m is the linear absorption coefficient of TiN and u, 2q and c represent the instrumental angles of a four-circle diffractometer. The azimuthal 4 and the inclination j angles, which define the sampled specimen directions, can be also expressed in terms of the diffractometer angles according to Eqs. (2) and (3):

j ¼ arccos½cos c$ cosðu  qÞ

(2)

  sin c f ¼ 4 þ arctan tanðu  qÞ

(3)

Considering Eqs. (1)e(3), a sin2j-based stress analysis can be carried out at a constant penetration depth by applying the following procedure: (I) the penetration depth t of interest is chosen taking into account e.g. the film thickness;

Fig. 1. Cross section micrograph of the S1000N sample showing the TiN coating (20,000).

S. Vales et al. / Materials Chemistry and Physics 177 (2016) 156e163

159

Fig. 3. TEM image from the S1000N sample showing increased defect density near the interface with the TiN coating.

Fig. 2. XRD spectra and phase identification for the S, SN, S1000 and S1000N samples.

regards to the S1000N sample, which was in turn bombarded with Xeþ prior to nitriding, both ε-Fe2e3N and g0 -Fe4N iron nitrides could be detected. The formation of ε-Fe2e3N nitride confirms that the previous ion bombardment was responsible for increasing nitrogen retention [26,27] on the steel surface during the plasma nitriding process. In addition, it is worth noticing that the volume fraction of g0 -Fe4N appears to be larger in the S1000N specimen than in the SN sample, as indicated by the difference in the relative intensity of the respective diffraction lines. This probably happens because bombardment with Xeþ ions texturized the treated surface, causing the generation of lattice defects and stress which accelerate the nitrogen diffusion into the first atomic surface layers [26,27]. Overall, the TiN reflections exhibit line broadening (e.g. in comparison with the reflections which stem from metallic iron in the steel substrate), an indication of small crystallite size and possibly nanostructured layer (i.e., disorder). In addition, there appears to be no strong preferential orientation in the TiN coatings deposited on the different substrates since in all cases several TiN reflections could be detected with no marked difference in their relative intensities, in agreement with a recent investigation [33]. The microstructure of the S1000N specimen was further investigated by TEM (Figs. 3 and 4). Observation of Fig. 3 indicates elevated dislocation density in the sub-surface region of the steel substrate (approximately 0.5 mm thickness), likely caused by plastic strain associated with the Xeþ ion bombardment. Possible sources of deformation caused by ion bombardment include: formation of gas bubbles from within the substrate lattice, phase transformations and the peening effect itself caused by ion collision [21,24]. The presence of lattice defects accounts for the higher diffusivity necessary for the formation of the nitrogen-rich ε-Fe2e3N nitride

Fig. 4. TEM image and electron diffraction pattern from the sample S1000N showing the presence of -Fe2-3N and g0 -Fe4N nitrides.

layer detected by XRD (Fig. 2). Fig. 4 displays a TEM image of the coating/substrate interface in S1000N sample. The interface between the steel substrate and TiN coating exhibit fringes due to very fine nitride formation (red arrows (in the web version)). Analyses of the electron diffraction data from the coating-steel interface indicate two distinct superposing patterns, corresponding to the ε-Fe2e3N and g0 -Fe4N nitrides, as indicated in the inset of Fig. 4, thus corroborating the results of the XRD analysis presented in Fig. 2. 3.2. Topography, wear and coating adhesion The surface topography of the samples investigated by AFM is displayed in Fig. 5. It is possible to verify that the different types of pre-treatment lead to marked differences in the surface roughness of the TiN films. The sample S exhibits the lowest levels of roughness, compatible with the surface finishing of the metallographic

160

S. Vales et al. / Materials Chemistry and Physics 177 (2016) 156e163

Fig. 5. AFM surface topography of the investigated samples obtained after wear testing.

preparation to which the substrates were initially subjected. Thus, the non-treated sample represents the reference state for the purpose of comparison. Plasma nitriding of the substrate (sample SN) leads to a small increase in roughness, which can be explained by different mechanisms such as: volumetric expansion of the substrate lattice due to incorporation of N in solid solution and formation of nitrides, sputtering of the surface due to the impact of N ions during deposition, as well as redeposition of the sputtered material [34,35]. The bombardment with Xeþ ions at 1.0 keV kinetic energy produces a more significant increase of roughness in the sample S1000 than the nitriding process. This occurs due to a more intense atomic attrition prompted by the heavy Xeþ ions. As previously shown reduction the Xeþ bombardment kinetic energy diminishes surface roughness [22,26]. Finally, the exposure of the 100Cr6 steel surface to prolonged atomic attrition caused by the combination of ion bombardment and plasma nitriding leads to the roughest surface profile in the sample S1000N. The evolution of the average surface roughness of the TiN coatings as a function of the pretreatment technique applied to the 100Cr6 substrates is presented in Fig. 6. The average worn volumes obtained after ball on disc tests are presented in Fig. 7. The lowest wear rates were observed for the samples subjected to individual treatments of nitriding (SN) or ion bombardment (S1000) instead of a combination of both plasma nitriding and Xeþ ion bombardment (S1000N). This indicates that the bombardment with Xeþ ions causes strain hardening of the steel surface that is comparable with the hardness increase promoted by nitride formation, or possibly a combination of surface hardening with slightly increased roughness that contributes to enhance the coating adhesion to the substrate. Both effects, i.e. increased substrate hardness and coating adhesion, ought to be considered here because in all cases the thin

Fig. 6. Evolution of average roughness Ra as a function of surface pre-treatment in TiN coating.

TiN coatings (~400 nm, Fig. 1) from all samples appear to have been completely removed during the wear tests. The wear behavior of the S1000N specimen was found to be comparable to the untreated sample S. In this case, excessive surface roughness causes insufficient TiN adhesion on the substrate. As the tribological system investigated (cemented carbide against TiN) involves materials of elevated hardness, the main wear mechanism appears to be abrasive, thus leading to appreciable breakup of the sharper coating asperities. Fig. 8 exhibits the scratch scars and the critical loads for coating

S. Vales et al. / Materials Chemistry and Physics 177 (2016) 156e163

161

delamination obtained during the nano-scratch tests. It is possible to observe chevron cracking, indicating cohesive failure. The dashed rectangles highlight the start of chipping failure (adhesive failure). These values indicate a transition point in the wear behavior, which can occur due to plastic deformation, increase in temperature [36] or initial failure of the coating (microcracking, surface flaking, loss of cohesion) [37]. In the present case, the critical load is an indicative of adhesion of the TiN coating to the substrate material. The highest critical load was observed on the SN sample, consistent with the minimum wear volume observed in Fig. 7. The same trend is observed for the remaining samples, within the error margins of the experiment.

3.3. Residual stress analyses Fig. 7. Average wear volume of the different samples after ball-on-disc tests.

Residual stress in thin coatings produced by deposition can usually be ascribed to two sources. On one hand, the energetic ions cause atoms from the substrate to be shifted from their original lattice positions producing residual interstitials. This leads to an outwards expansion of the thin film from the substrate. In the plane of the coating, however, expansion is restricted and the entrapped atoms cause elevated compressive stress [38]. In addition to these intrinsic stresses, additional residual stresses arise during cooling from the deposition temperature due to incompatibility of thermal expansion between substrate and coating [39,40]. These cooling stresses are therefore influenced by the surface pre-treatments which can alter the phase composition and the average thermal expansion coefficient near the interface with the hard metal coating. The sin2j versus lattice spacing plots are presented in Fig. 9. As can be observed for all samples, residual stresses are compressive and the {200} lattice spacing varies linearly with sin2j, in agreement with previous investigations [41,42]. It is also important to observe that XRD measurements for residual stress analysis on all materials could be performed over a large range of sample tilt positions, indicating that no strong preferential orientation is present in TiN coatings, as expected from the results presented in Fig. 2. The residual stress values obtained for each coating procedure are presented in Fig. 10. In all cases, stress values are highly compressive: (6.25 ± 0.2), (4.55 ± 0.05), (6.77 ± 0.12) and (5.95 ± 0.1) GPa for the S, SN, S1000 and S1000N specimens, respectively. By comparing the results obtained for samples S and

Fig. 8. a) Image of the scratch scars and b) critical load for TiN coatings deposited onto the different substrates.

Fig. 9. sin2j versus TiN (200) lattice spacing plots of the investigated substrates.

162

S. Vales et al. / Materials Chemistry and Physics 177 (2016) 156e163

impact energy in combination with plasma nitriding caused sputtering of substrate and significantly increased surface roughness of the TiN film. In such condition, the wear rate of the TiN became similar to that observed in the substrate without pre-treatment. Reduction of the Xeþ ion energy shall lead to less sputtering and surface roughness. This might enhance the wear performance in the case of a combined pre-treatment due to a more pronounced nitrogen retention. All TiN coatings evolved compressive residual stresses, ranging from 4.55 and 6.77 GPa. The lowest residual stress values were observed for the sample subjected to individual plasma nitriding due to the formation of only g0 -Fe4N with reduced TEC and the increased chemical compatibility by a Ti-rich interlayer between substrate and coating during IBAD. Acknowledgments

Fig. 10. Average residual stress values in the TiN as a function of surface pre-treatment.

S1000, it is possible to observe that Xeþ bombardment alone does not have a significant effect on the residual stress state on the deposited TiN coating. However, plasma nitriding caused a significant reduction in the residual stresses. In duplex coatings, where a hard ceramic layer is deposited onto a nitrided substrate, a transition zone appears between substrate and coating [37,43]. This happens because of the high ion energy during IBAD that promotes the transference of chemical elements across the substrate/coating interface, leading to the formation of a Ti-rich interlayer [33]. In the nitrided sample, this would lead to the reaction with nitrogen and an increase in adhesion and chemical compatibility between the TiN coating and substrate. This also contributes to reduce the residual stresses in the TiN coating. For similar reasons, the SN sample exhibited superior wear characteristics (Figs. 7 and 8) in comparison to the other tested materials. The application of Xeþ bombardment prior to plasma nitriding causes an increase in the residual stresses of the TiN film to values similar to those of the untreated substrate in the S sample. This is because of the occurrence of ε-Fe2e3N that has a Thermal Expansion Coefficient (TEC) greater than g0 -Fe4N: TEC(g0 -Fe4N) < TEC(εFe2e3N) < TEC(Fe). 4. Conclusions The impact of individual and combined pre-treatments based on the bombardment with 1.0 keV Xeþ ions and plasma nitriding at 380  C on the adhesion and wear of TiN coatings deposited by IBAD onto spheroidized 100Cr6 steel was investigated. The results showed that the bombardment with Xeþ ions causes grain refinement, texturization, and increases the density of lattice imperfections at the treated surface, thus enhancing nitrogen diffusion at 380  C and probably causing strain-hardening as well. The combined pre-treatment led therefore to a diffusion zone containing both ε-Fe2e3N and g0 -Fe4N nitrides, whereas simple plasma nitriding promoted the formation of only g0 -Fe4N. The presence of ε-Fe2e3N confirms an enhanced nitrogen retention in the diffusion zone and the formation of a more expressive nitride fraction. This corroborates the positive influence of Xeþ ion bombardment on the nitrogen retention during nitriding treatments. Superior wear behavior was however verified for the application of individual plasma nitriding. Xeþ ion bombardment with 1.0 keV

The authors HP and JG acknowledge the financial support of the NanoCom Network (Proposal FP7 n 247524) and the CREATeNetwork project (H2020-MSCA-RISE/644013), both funded by the European Union. SV is grateful to Conselho Nacional de Desenvolgico (CNPq) and Comissa ~o de Apervimento Científico e Tecnolo feiçoamento de Pessoal do Nível Superior (CAPES) for granting the scholarship for acting in this research. FA and HP are CNPq fellows. Part of this work was supported by FAPESP, projects 2010/11391-2 and 2012/10127-5. The Brazilian Nanotechnology National Laboratory (LNNano) is greatly acknowledged for support during the TEM analyses. References [1] E. Menthe, K.T. Rie, J.W. Schultze, S. Simson, Structure and properties of plasma nitrided stainless steel, Surf. Coat. Technol. 74e75 (1995) 412e416. [2] M.K. Lei, Z.L. Zhang, Microstructure and corrosion resistance of plasma source ion nitrided austenitic stainless steel, J. Vac. Sci. Technol. A 15 (1997) 421e427. [3] N. Renevier, P. Collignon, H. Michel, T. Czerwiec, Low temperature nitriding of AISI 316L stainless steel and titanium in a low pressure arc discharge, Surf. Coat. Technol. 111 (1999) 128e133. [4] C.H. Ma, J.H. Huang, H. Chen, Texture evolution of transition-metal nitride thin films by ion beam assisted deposition, Thin Solid Films 446 (2004) 184e193. [5] M.V. Leite, C.A. Figueroa, S.C. Gallo, A.C. Rovani, R.L.O. Basso, P.R. Mei, I.J.R. Baumvol, A. Sinatora, Wear mechanisms and microstructure of pulsed plasma nitrided AISI H13 tool steel, Wear 269 (2010) 466e472. [6] L.F. Zagonel, J. Bettini, R.L.O. Basso, P. Paredez, H. Pinto, C.M. Lepienski, F. Alvarez, Nanosized precipitates in H13 tool steel low temperature plasma nitriding, Surf. Coat. Technol. 207 (2012) 72e78. [7] E.A. Ochoa, C.A. Figueroa, F. Alvarez, Nitriding of AISI 4140 steel by a low energy broad ion source, J. Vac. Sci. Technol. A 24 (6) (2006) 2113e2116. [8] S. Vales, E.A. Ochoa, P.P. Brito, R. Droppa Jr., J. Garcia, F. Alvarez, H.C. Pinto, Effect of low temperature nitriding of 100Cr6 substrates on TiN coatings deposited by IBAD, Mat. Res. 18 (1) (2015) 54e58, http://dx.doi.org/10.1590/ 1516-1439.266514. [9] S.M.M. Ramos, L. Amaral, M. Behar, G. Marest, A. Vasquez, F.C. Zawislak, The effects of xenon bombardment on the dissolution and reprecipitation of carbonitrides produced in nitrogen-implanted low carbon steel, Surf. Coat. Technol. 45 (1991) 255e262. [10] K. Gammer, M. Stoiber, J. Wagner, H. Hutter, R. Kullmer, C. Mitterer, Investigations on the effects of plasma-assisted pre-treatment for plasmaassisted chemical vapour deposition TiN coatings on tool steel, Thin Solid Films 461 (2004) 277e281. [11] Y.L.K.X. Shengli Ma, The composite of nitrided steel H13 and TiN coatings by plasma duplex treatment and the effect of pre-nitriding, Surf. Coat. Technol. 137 (2001) 116e121.  [12] P. Panjan, M. Cekada, R. Kirn, M. Sokovi c, Improvement of die-casting with duplex treatment, Surf. Coat. Technol. 180e181 (2004) 561e565.  , D. Kakas, Tribological behavior of TiN and TiAlN deposited on [13] B. Skori e substrates plasma nitrided at low pressure, Mater. Manuf. Process. 10 (1995) 321e326.  , D. Kakas, T. Gredi [14] B. Skori e c, Influence of plasma nitriding on mechanical and tribological properties of steel with subsequent PVD surface treatments, Thin Solid Films 317 (1998) 486e489. [15] Y. Sun, T. Bell, Plasma surface engineering of low alloy steel, Mater. Sci. Eng. A 140 (7) (1991) 419e434. [16] J. Garcia, R. Pitonak, The role of cemented carbide functionally graded outer-

S. Vales et al. / Materials Chemistry and Physics 177 (2016) 156e163

[17] [18] [19]

[20]

[21]

[22]

[23] [24]

[25]

[26]

[27]

[28] [29] [30]

layers on the wear performance of coated cutting tools, Int. J. Refract. Met. Hard Mater. 36 (2013) 52e59. H. Dong, Y. Sun, T. Bell, Enhanced corrosion resistance of duplex coatings, Surf. Coat. Technol. 90 (1e2) (1997) 91e101. J.D. Kaminga, D. Doerwald, M. Schreurs, G.C.A.M. Janssen, Industrial feasibility of the nitrocoat process, Surf. Coat. Technol. 200 (5e6) (2005) 1837e1841. J.D. Kamminga, H. Hoy, G.C.A.M. Janssen, E. Lugscheider, M. Maes, First results on duplex coatings without intermediate mechanical treatment, Surf. Coat. Technol. 174e175 (2003) 671e676.  , D. Kakas, M. Rakita, N. Bibi B. Skori e c, D. Peruskob, Structure, hardness and adhesion of thin coatings deposited by PVD, IBAD on nitrided steels, Vaccum 76 (2004) 169e172. D. Kakas, P. Terek, L. Kovacevic, A. Miletic, B. Skoric, Influence of interfacial layer thickness and substrate roughness on adhesion of TiN coatings deposited at low temperatures by IBAD, Surf. Rev. Lett. 18 (2011) 83e90. R. Droppa Jr., H.C. Pinto, J. Garcia, E.A. Ochoa, M. Morales, S. Cucatti, F. Alvarez, Influence of ion-beam bombardment on the physical properties of 100Cr6 steel, Mater. Chem. Phys. 147 (2014) 105e112, http://dx.doi.org/10.1016/ j.matchemphys.2014.04.016. N. Mottu, M. Vayer, R. Erre, Xe and Mo implantation on austenitic stainless steel: structural modification, Surf. Coat. Technol. 183 (2004) 165e173. R.G. Lacerda, M.C. dos Santos, L.R. Tessler, P. Hammer, F. Alvarez, F.C. Marques, Pressure-induced physical changes of noble gases implanted in highly stressed amorphous carbon films, Phys. Rev. B 68 (2003) 054104. R.L. Basso, R.J. Candal, C.A. Figueroa, D. Wisnivesky, F. Alvarez, Influence of microstructure on the corrosion behavior of nitrocarburized AISI H13 tool steel obtained by pulsed DC plasma, Surf. Coat. Technol. 203 (2009) 1293e1297. S. Cucatti, E.A. Ochoa, M. Morales, R. Droppa Jr., J. Garcia, H.C. Pinto, L.F. Zagonel, D. Wisnivesky, C.A. Figueroa, F. Alvarez, Effect of bombarding steel with Xeþ ions on the surface nanostructure and on pulsed plasma nitriding process, Mater. Chem. Phys. 149e150 (2015), 261e169. E.A. Ochoa, R. Droppa, R.L.O. Basso, M. Morales, S. Cucatti, L.F. Zagonel, T. Czerwiec, M.C. dos Santos, C.A. Figueroa, F. Alvarez, The effect of noble gas bombarding on nitrogen diffusion in steel, Mater. Chem. Phys. 143 (2013) 116e123. E.A. Ochoa, C.A. Figueroa, T. Czerviec, F. Alvarez, Enhanced nitrogen diffusion induced by atomic attrition, Appl. Phys. Lett. 88 (2006) 254109. P. Hammer, N.M. Victoria, F. Alvarez, Electronic structure of hydrogenated carbon nitride films, J. Vac. Sci. Technol. 16 (5) (1998) 2941e2949. A. Kumar, U. Welzel, E.J. Mittemeijer, A method for the non-destructive

[31]

[32]

[33]

[34] [35]

[36] [37]

[38] [39] [40]

[41]

[42]

[43]

163

analysis of gradients of mechanical stresses by X-ray diffraction measurements at fixed penetration/information depths, J. Appl. Crystallogr. 39 (5) (2006) 633e646, http://dx.doi.org/10.1107/S0021889806023417. €ntgenographische untersuchung von spanB. Eigenmann, E. Macherauch, Ro €nden in werkstofften, Fortsetz. Matwiss Werktofftechn 26 (1995) nungszusta 199e216. A. Kumar, U. Welzel, E.J. Mittemeijer, Depth dependence of elastic grain interaction and mechanical stress: analysis by x-ray diffraction measurements at fixed penetration/informationdepths, J. Appl. Phys. 100 (11) (2006), http:// dx.doi.org/10.1063/1.2363899 art. no. 114904. T.H.C. Miranda, S. Vales, E.O. Becerra, R. Droppa Jr., P.P. Brito, J.L. Garcia, F. Alvarez, H. Pinto, Influence of Xe ion-bombardment on the substrate microstructure and the residual stresses of TiN coatings deposited by plasma reactive sputtering onto AISI 4140 steel, Adv. Mater. Res. 996 (2014) 841e847, http://dx.doi.org/10.4028/www.scientific.net/AMR.996.841. F. Mahboudi, K. Abdolvahabi, The effect of temperature on plasma nitriding behaviour of DIN 1.6959 low alloy steel, Vaccum 81 (2006) 239e243. A. Akbari, R. Mohammadzadeh, C. Templier, J.P. Riviere, Effect of the initial microstructure on the plasma nitriding behavior of AISI M2 high speed steel, Surf. Coat. Technol. 204 (2010) 4114e4120. H. So, Characteristics of wear results tested by pin-on-disc at moderate to high speeds, Tribol. Int. 29 (1996) 415e423. M. Polok-Rubiniec, K. Lukaszkowicz, L.A. Dobrzanski, Comparison of nanostructure and duplex PVD coatings deposited onto hot work tool steel substrate, J. Achiev. Mater. Manuf. Eng. 41 (2010) 187e194. C.A. Davis, A simple model for the formation of compressive stress in thin films by ion bombardment, Thin Solid Films 226 (1993) 30e34. M. Birkholz, Thin Film Analysis by X-Ray Scattering, WILEY-VCH, 2006. ISBN: 3-527-31052-5. G. Totten, M. Howes, T. Inoue, Stress determination in coatings, in: Handbook of Residual Stress and Deformation of Steels, first ed., ASTM International, USA, 2002, pp. 118e124, http://dx.doi.org/10.1361/hrsd2002p118. C.H. Ma, J.-H. Huang, H. Chen, Residual stress measurement in textured thin film by grazing-incidence x-ray diffraction, Thin Solid Films 418 (2002) 73e78, http://dx.doi.org/10.1016/S0040-6090(02)00680-6. A. Akbari, J.P. Riviere, C. Templier, E. Le Bourhis, G. Abadias, Hardness and residual stresses in TiN-Ni nanocomposite coatings deposited by reactive dual ion beam sputtering, Rev. Adv. Mater. Sci. 15 (2007) 111e117. G.E. Totten, H. Liang, Surface Modification Mechanisms: Friction, Stress and Reaction Engineering, Marcel Dekker, Inc., New York, 2004. ISBN: 0-82474872-7.