Corrosion properties of CVD grown Ti(C,N) coatings in 3.5 wt-% NaCl environment

Abstract

The corrosion behaviour of Titanium carbonitride (Ti(C,N)) films grown by chemical vapour deposition was analysed in artificial sea water environment. From potentiodynamic polarisation curves, two passivation zones were detected, which originated from an initial oxidation of TiC and TiN to TiO₂ followed by growth of the TiO₂ layer upon increased polarisation. X-ray photoelectron spectroscopy analyses verified the mechanism by detecting a gradual decrease in Ti(C,N) peaks accompanied by a gradual increase of oxidised Ti (e.g. TiO₂). It was likewise found that carbon in TiC mainly decomposes into carbonate species while the nitrogen in TiN remains elemental and likely escapes as nitrogen gas. Accordingly, Ti(C,N) behaves like a superposition of TiC and TiN with their individual oxidation behaviour, resulting in a highly corrosion resistant material.

GRAPHICAL ABSTRACT

Keywords

Corrosion Ti(C,N)CVD XPS

Highlights

Corrosion properties of CVD grown Ti(C,N) film

Two passivation zones detected in the polarisation curve

TiC and TiN oxidises to TiO₂, TiCO₃ and N₂

Introduction

Titanium carbides and nitrides have been used as wear protective hard coatings for decades. A major field of use is the metal cutting industry, where coatings of TiC, TiN and Ti(C,N) are routinely fabricated by either chemical vapour deposition (CVD) or physical vapour deposition (PVD) [1–4]. These hard coatings are also interesting for other applications, such as medical implants, where the excellent chemical stability and low toxicity of the coating are required [5–7].

CVD is routinely used to grow uniform and dense coatings of Ti(C,N) with excellent mechanical properties [8,9]. The technique offers a high step coverage and is therefore particularly adapt to coat complicated geometrical shapes and porous substrates. The technique has for instance been used to coat steel substrates, due to the moderate temperature needed to grow Ti(C,N) thin films (i.e. <830°C) [1]. Insights into the chemical stability can be obtained by investigating the electrochemical properties and more specifically the corrosion properties in different environments. While plenty of corrosion studies of TiC and TiN can be found in the literature [10–14], Ti(C,N) has been significantly less investigated [1,7,15].

Recent advances in the development of highly oriented Ti(C,N) thin films have revealed superior mechanical properties [16,17], compared to the state of the art in Ti(C,N) coatings grown by CVD [17]. Cutting fluid for coolant and chip removal is commonly used in many cutting operations. As such, it is likely that the protective coatings typically applied to the tools (e.g. Ti(C,N)) will come in contact with the cutting fluid under severe conditions (e.g. high temperature and high pressure). These conditions put the material under both mechanical and chemical stress, possibly causing rapid oxidation of the material. The present study aims to investigate the chemical stability of the Ti(C,N) thin film in a simulated sea water environment by studying the corrosion properties of the material. The corrosion mechanism and the corrosion products will be investigated by electrochemical analyses coupled with surface and chemical characterisation (i.e. atomic force microscopy (AFM), scanning electron microscopy (SEM) and X-ray photoelectron spectroscopy (XPS)).

Material and methods

Synthesis and sample preparation

The deposition and characterisation of the CVD Ti(C,N) coating used in this study are described in a previous work [16]. The CVD deposition was performed for 155 min at a growth rate of 0.67 µm/h using the following parameters: T = 830°C, P _tot = 8.0 kPa, P _TiCl4 = 0.26 kPa, P = 0.04 kPa, P = 0.7 kPa and P = 7.0 kPa. Ti(C,N) films were grown on single crystal c-sapphire substrates, as these substrates are insulating and thus do not contribute to the electrochemical response during corrosion measurements. The composition of the coating was Ti(C_0.6,N_0.4)_0.8[16]. TiC and TiN grown by CVD onto sapphire substrates were used as references; deposition parameters for the reference TiC and TiN coatings can be found in Table A1.

Characterisation

The surface morphology was analysed before and after corrosion measurements by high-resolution scanning electron microscopy (HR-SEM, Zeiss Merlin). The surface roughness and effective surface area were investigated before and after the corrosion measurements by AFM (PSIA XE150) operated in non-contact mode and in a 5 × 5 µm scan area.

The chemical environment of the corrosion products was investigated by XPS using a PHI Quantera II Scanning electron spectroscopy for chemical analysis (ESCA) microprobe with monochromatic Al Kα radiation and a spot size of 100 µm. Before the spectral acquisition, a gentle Ar⁺ etch of 200 eV was carried out for 1 min in order to remove contaminants from handling the sample in air. To avoid charging of the samples, neutralisation by both electrons and low energy Ar⁺ ions was used during the measurement.

Corrosion measurements

Potentiodynamic polarisation measurements were carried out to evaluate the corrosion properties of the coatings. All electrochemical analyses were performed in 3.5 wt-% NaCl aqueous electrolyte using a VersaSTAT4 (Princeton Applied Research) potentiostat/galvanostat. A three-electrode cell composed of the sample as the working electrode, a Pt wire counter electrode and saturated Ag/AgCl reference electrode was used. The exposed area for all samples was 0.196 cm². Initial open circuit potential measurements for 1 h were followed by polarisation at −1.5 V for 300 s, after which linear sweep voltammetry (LSV) was used to scan from −1.5 to +1.5 V at 1 mV/s. The corrosion potentials (E _corr) and the passivation potentials (E _pass) were extracted from the polarisation curves.

Results and discussion

Corrosion properties of Ti(C,N)

Potentiodynamic polarisation curves measured for the Ti(C,N) coatings are shown in Figure 1. From the corrosion potential (i.e. potential of zero net current) at about −0.3 V, a steady current increase was observed until maximum at 0.32 V was reached. At higher potentials the current decreased until a second maximum appeared at 1.05 V. This behaviour indicates two reactions passivating the surface in different potential regions. To understand the origin of these passivating reactions, reference samples of CVD grown TiC and TiN thin films were analysed and the resulting polarisation curves are shown in Figure 1. Titanium carbide showed a similar polarisation behaviour as Ti(C,N) with the notable difference in the first peak position (i.e. at 0.51 V for TiC) as well as two additional minor passivation peaks at higher potentials (i.e. 0.73 and 1.16 V). A higher passivation current was then observed in the region 0.5-0.9 V, compared to Ti(C,N). Titanium nitride, on the other hand, showed a steady current increase up to 1.10 V, after which the current density remained constant to finally rise after 1.3 V. The Ti(C,N) polarisation curve is therefore, most probably, a result of both TiC and TiN oxidation and passivation [13,14]:

Figure 1

Potentiodynamic polarisation curves for Ti(C,N), TiN and TiC thin films in 3.5 wt-% NaCl electrolyte. Points where samples were stopped for XPS analyses are marked.

The mechanism behind Ti(C,N) oxidation is not surprisingly coupled to the corrosion properties of its TiC and TiN constituents. With its chemical composition of Ti(C_0.6,N_0.4)_0.8, the corrosion properties of the Ti(C,N) film is expected to be slightly more similar to the TiC corrosion behaviour. TiC oxidises to TiO₂ upon anodic polarisation resulting in the first observed peak. Previous studies of TiC films have also speculated that carbonate species could be formed and thus detectable on the electrode surface [13]. The multiple passivation peaks observed in the TiC polarisation curve (i.e. above 0.5 V) can be interpreted as a repeating precipitation of oxidised carbon species on the surface and as such agree with the latter hypothesis. TiN has been shown to readily oxidise to form TiO₂ and elemental nitrogen; however, formation of gas bubbles was observed during the LSV experiment. It has been speculated that a nitrogen-enriched layer situated between TiO₂ and TiN protects the latter from further oxidation. An applied potential exceeding 1.05 V seems to be required to overcome the passivating layer and oxidise underlying TiN, in agreement with previous studies of TiN films [14]. A summary of the corrosion properties for Ti(C,N) and the TiC and TiN reference samples can be found in Table 1.

Table 1

Corrosion properties of CVD grown Ti(C,N), TiC and TiN thin films in 3.5 wt-% NaCl electrolyte.

Specimen	E _corr	E _pass1	E _pass2	E _pass3
V (vs. Ag/AgCl)	V (vs. Ag/AgCl)	V (vs. Ag/AgCl)	V (vs. Ag/AgCl)
Ti(C,N)	−0.27	0.32	1.06	–
TiN	−0.21	1.10	–	–
TiC	−0.25	0.51	0.73	1.16

XPS analysis of corroded Ti(C,N)

In order to get a better insight on the corrosion mechanism of Ti(C,N), XPS was used to analyse Ti(C,N) films at different degrees of polarisation. In Figure 2(a–d), high-resolution core level spectra of Ti 2p, C 1 s, N 1 s and O 1 s are presented for the corroded Ti(C,N) samples stopped at 0.68 V and 1.5 V during the LSV experiment, as well as for the pristine Ti(C,N) reference sample. The core-level spectra for the reference sample agrees well with what has been reported in the literature on Ti(C,N) thin films [18–20]. Significant differences were observed in the XPS spectra of the corroded samples, compared to the reference sample. In the Ti 2p spectra, the Ti(C,N) peaks at about 455.1 and 461.0 eV decreased in intensity upon increasing polarisation (Figure 2(a)). New peaks also appeared at higher binding energies at about 459.3 and 465.2 eV, indicating an oxide environment on the surface, most probably corresponding to TiO₂[21]. However, a distinct peak shoulder below 458 eV was also observed for the fully polarised sample (i.e. 1.5 V), indicating the presence of Ti with oxidation number below +4 [22].

Figure 2

XPS spectra of core levels of (a) Ti 2p, (b) O 1 s, (c) C 1 s and (d) N 1 s for the Ti(C,N) samples that, during the polarisation curve, were stopped at 0.68 and 1.5 V, respectively, and for the pristine reference Ti(C,N) sample.

In the C 1 s, a peak at about 282.0 and 285.1 eV can be observed for all samples, which most probably correspond to a carbide environment and adventitious carbon, respectively [20,23]. As the Ti(C,N) samples were polarised, the carbide contribution decreased in intensity and new peaks at higher binding energies occurred (∼285.7, ∼286.7 and ∼289.9 eV). The latter is most likely corresponding to carbonate species and the former to oxidised carbon species of various degree of oxidation [23,24]. In the N 1 s spectrum, a peak at about 397.0 eV can be observed for all samples, which is most likely attributed to a nitride environment and follow a similar trend as the carbide peak in the C 1 s spectra, i.e. decreasing with increasing polarisation [18,24]. This is also in agreement with the observed decrease of the Ti(C,N) peaks in the Ti 2p spectra. Furthermore, in the N 1 s spectrum, there is a relatively small and broad feature for the fully polarised sample at about ∼400 eV, corresponding to oxidised N containing species [25]. In the O 1 s spectra, the O-Ti contribution at about 531 eV is increasing in intensity with increased degree of polarisation, which confirms the assumption of a more oxide character for the Ti(C,N) surface upon increased polarisation [19,20]. Additionally, a clear shoulder was observed in the O 1 s spectra at the high binding energy side of the O-Ti peak, most probably corresponding to an O-(C,N) environment [25–27]. Also, a small contribution at about 534.4 eV in the O 1 s spectra can be observed, that most probably could be ascribed to adsorbed water [28].

The XPS analysis of Ti(C,N) films confirms the assumption that the film is gradually corroded according to the individual TiC and TiN corrosion mechanisms. In particular, detection of oxidised carbon species on the surface after full polarisation confirms the previous hypothesis that carbonate species can be formed during oxidation of TiC. However, the analysis cannot exclude CO₂(g) as a secondary reaction product. The TiN peak evolution during semi polarisation (i.e. 0.68 V) indicates that initially an oxide surface layer is formed, which attenuated the TiN signal. After full polarisation (i.e. 1.5 V), the TiN peak shape is altered and the intensity decreased, which could be interpreted as a two-step oxidation process.

Surface morphology

The surface morphology was analysed after the corrosion measurements by SEM and AFM, and compared to the pristine surface, as is shown in Figure 3. The smooth film surface was composed on sub-micrometre-sized crystals that generated an effective surface area gain of 8.8% (i.e. effective surface area 27.2 µm² compared to the 25 µm² footprint area analysed). After full polarisation to 1.5 V (i.e. LSV scan to 1.5 V), no significant change in surface morphology or roughness was observed and likewise no signs of pitting corrosion. As expected, the high corrosion resistance of TiC and TiN are maintained in CVD grown Ti(C,N) films in the artificial sea water environment.

Figure 3

Surface morphology (a and c) before and (b and d) after corrosion analyses imaged by (a–b) SEM and (c–d) AFM.

Conclusions

The corrosion properties of CVD grown Ti(C,N) films were investigated through electrochemical analyses coupled with XPS analyses of the corroded surface. Two distinct passivation zones were observed for Ti(C,N) with peaks at 0.32 and 1.05 V which aligned with reference measurements of TiC and TiN films. XPS measurements confirmed the gradual oxidation of TiC and TiN, in Ti(C,N), to TiO₂ by analysing the corrosion product after each passivation peak. Carbon oxidation, in TiC, was also detected as carbonate species were found on the surface. Unlike TiC, the main corrosion product of TiN is elemental nitrogen that most likely leaves the surface as N₂(g). Also, titanium species of lower oxidation state than +4 were detected, possibly originating from compounds having a composition between TiO₂ and TiO. The corrosion measurement did not produce any noticeable difference in surface morphology or roughness, as confirmed by SEM and AFM. As such, it can be concluded that CVD-grown Ti(C,N) films are highly resistant to corrosion in artificial sea water and undergo an individual passivation of TiC and TiN to TiO₂.

Footnotes

Acknowledgements

The authors acknowledge the valuable inputs of Dr Lewin and Dr Maibach on the XPS analyses.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Linus von Fieandt

Kristina Johansson

Erik Lindahl

David Rehnlund

Appendix A

References

Bonetti

Wiprächtiger

Mohn

Cvd of titanium carbonitride at moderate temperature: properties and applications. Met Powder Rep. 1990: 45: 837–840. doi: 10.1016/0026-0657(90)90575-2

Rebenne

Bhat

DG.

Review of CVD TiN coatings for wear-resistant applications: deposition processes, properties and performance. Surf Coat Technol. 1994; 63: 1–13. doi: 10.1016/S0257-8972(05)80002-7

Vuorinen

Hoel

RH.

Interfacial characterization of chemically vapour deposited titanium carbide on cemented carbide. Thin Solid Films. 1993; 232: 73–82. doi: 10.1016/0040-6090(93)90765-H

Jindal

Santhanam

Schleinkofer

Performance of PVD TiN, TiCN, and TiAlN coated cemented carbide tools in turning. Int J Refract Met Hard Mater. 1999; 17: 163–170. doi: 10.1016/S0263-4368(99)00008-6

Hollstein

Louda

Bio-compatible low reflective coatings for surgical tools using reactive d.c.-magnetron sputtering and arc evaporation – a comparison regarding steam sterilization resistance and nickel diffusion. Surf Coat Technol. 1999; 120-121: 672–681. doi: 10.1016/S0257-8972(99)00357-6

Antunes

Rodas

ACD

Lima

Study of the corrosion resistance and in vitro biocompatibility of PVD TiCN-coated AISI 316L austenitic stainless steel for orthopedic applications. Surf Coat Technol. 2010; 205: 2074–2081. doi: 10.1016/j.surfcoat.2010.08.101

Chatterjee-Fischer

Mayr

Investigations on TiCN – layers obtained at moderate temperatures. Proc. Fifth Eur. Conf. Chem. Vap. Depos. 1985. p. 395-405.

Paseuth

Fukui

Okuno

Microstructure, mechanical properties, and cutting performance of TiCxN1-x coatings with various x values fabricated by moderate temperature chemical vapor deposition. Surf Coat Technol. 2014; 260: 139–147. doi: 10.1016/j.surfcoat.2014.09.068

Larsson

Ruppi

Microstructure and properties of Ti(C,N) coatings produced by moderate temperature chemical vapour deposition. Thin Solid Films. 2002; 402: 203–210. doi: 10.1016/S0040-6090(01)01712-6

10.

Hintermann

HE.

Exploitation of wear- and corrosion-resistant cvd-coatings. Tribol Int. 1980; 13: 267–277. doi: 10.1016/0301-679X(80)90090-0

11.

Rie

Gebauer

Wöhle

Plasma assisted CVD for low temperature coatings to improve the wear and corrosion resistance. Surf Coat Technol. 1996; 86-87: 498–506. doi: 10.1016/S0257-8972(96)03177-5

12.

Mantyla

Helevirta

Lepisto

Corrosion behaviour and protective quality of TiN coatings. Thin Solid Films. 1985; 126: 275–281. doi: 10.1016/0040-6090(85)90321-9

13.

Beverskog

Carlsson

J-O

Deblanc Bauer

Corrosion properties of TiC films prepared by activated reactive evaporation. Surf Coat Technol. 1990; 41: 221–229. doi: 10.1016/0257-8972(90)90170-H

14.

Piippo

Elsener

Böhni

Electrochemical characterization of TiN coatings. Surf Coat Technol. 1993; 61: 43–46. doi: 10.1016/0257-8972(93)90200-8

15.

Zhang

Xue

Microstructure and corrosion behavior of TiC/Ti(CN)/TiN multilayer CVD coatings on high strength steels. Appl Surf Sci. 2013; 280: 626–631. doi: 10.1016/j.apsusc.2013.05.037

16.

von Fieandt

Johansson

Larsson

On the growth, orientation and hardness of chemical vapor deposited Ti(C,N). Thin Solid Films. 2018; 645: 19–26. doi: 10.1016/j.tsf.2017.10.037

17.

von Fieandt

Fallqvist

Larsson

Tribological properties of highly oriented Ti(C,N) deposited by chemical vapor deposition. Tribol Int. 2018; 119: 593–599. doi: 10.1016/j.triboint.2017.11.040

18.

Didziulis

Lince Jr Stewart

Photoelectron spectroscopic studies of the electronic structure and bonding in Tic and TiN. Inorg Chem. 1994: 33: 1979–1991. doi: 10.1021/ic00087a039

19.

Alves

CFA

Oliveira

Carvalho

Influence of albumin on the tribological behavior of Ag – Ti (C, N) thin films for orthopedic implants. Mater Sci Eng C. 2014; 34: 22–28. doi: 10.1016/j.msec.2013.09.031

20.

Soto

Aes, EELS and XPS characterization of Ti (C, N, O) films prepared by PLD using a Ti target in N2, CH4, O2 and CO as reactive gases. Appl Surf Sci. 2004; 233: 115–122. doi: 10.1016/j.apsusc.2004.03.212

21.

Saha

Tompkins

HG.

Titanium nitride oxidation chemistry: An X-ray photoelectron spectroscopy study. J Appl Phys. 1992; 72: 3072–3079. doi: 10.1063/1.351465

22.

Guillot

Fabreguette

Imhoff

Amorphous TiO2 in LP-OMCVD TiNxOy thin films revealed by XPS. Appl Surf Sci. 2001; 177: 268–272. doi: 10.1016/S0169-4332(01)00220-3

23.

Biesinger

Lau

LWM

Gerson

Resolving surface chemical states in XPS analysis of first row transition metals, oxides and hydroxides: Sc, Ti, V, Cu and Zn. Appl Surf Sci. 2010; 257: 887–898. doi: 10.1016/j.apsusc.2010.07.086

24.

Moulder

J. F.

Stickle

W. F.

Sobol

P. E.

. ‌Standard spectra of the elements. In: Chastain

King

R. C. J.

, editors. Handbook of X-ray photoelectron spectroscopy. Eden Prairie (MN): Physical Electronics, Inc.; 1992. p. 40–45.

25.

Jouan

P-Y

Peignon

M-C

Cardinaud

Characterisation of TiN coatings and of the TiN/Si interface by X-ray photoelectron spectroscopy and Auger electron spectroscopy. Appl Surf Sci. 1993; 68: 595–603. doi: 10.1016/0169-4332(93)90241-3

26.

Jaeger

Patscheider

A complete and self-consistent evaluation of XPS spectra of TiN. J Electron Spectros Relat Phenomena. 2012; 185: 523–534. doi: 10.1016/j.elspec.2012.10.011

27.

Chappé

Marco de Lucas

Cunha

Structure and chemical bonds in reactively sputtered black Ti–C–N–O thin films. Thin Solid Films. 2011; 520: 144–151. doi: 10.1016/j.tsf.2011.06.108

28.

Braic

Balaceanu

Vladescu

Preparation and characterization of titanium oxy-nitride thin films. Appl Surf Sci. 2007; 253: 8210–8214. doi: 10.1016/j.apsusc.2007.02.179