Influence of bias voltage on structure and tribocorrosion properties of TiSiCN coating in artificial seawater

Accepted Manuscript Influence of bias voltage on structure and tribocorrosion properties of TiSiCN coating in artificial seawater

Yue Wang, Jinlong Li, Chaoqun Dang, Yongxin Wang, Yuejin Zhu PII: DOI: Reference:

S1044-5803(17)30701-5 doi: 10.1016/j.matchar.2017.03.012 MTL 8592

To appear in:

Materials Characterization

Received date: Revised date: Accepted date:

13 August 2016 26 December 2016 7 March 2017

Please cite this article as: Yue Wang, Jinlong Li, Chaoqun Dang, Yongxin Wang, Yuejin Zhu , Influence of bias voltage on structure and tribocorrosion properties of TiSiCN coating in artificial seawater. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Mtl(2017), doi: 10.1016/ j.matchar.2017.03.012

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Influence of bias voltage on structure and tribocorrosion properties of TiSiCN coating in artificial seawater Yue Wang a, b, Jinlong Li a, Chaoqun Dang a, Yongxin Wang a, Yuejin Zhu b a

Key Laboratory of Marine Materials and Related Technologies, Zhejiang Key Laboratory of Marine

PT

Materials and Protective Technologies, Ningbo Institute of Materials Technology and Engineering,

Faculty of Science, Ningbo University, Ningbo 315211, PR China

SC

b

RI

Chinese Academy of Sciences, Ningbo 315201, PR China

Abstract

NU

The TiSiCN coatings deposited at different bias voltages were fabricated on Ti6Al4V alloy by

MA

arc ion plating. The structure and properties of the TiSiCN coating were characterized using scanning electron microscope, X-ray diffraction, X-ray photoelectron spectroscopy, transmission electron

D

microscopy, nanoindentation, potentiostat and ball-on-plate wear tests. As the bias voltage increases,

PT E

the TiSiCN coating shows a nanocystallite/amorphous structure, whereas the phase constitution and grain size changed slightly, and its hardness and tribocorrsoion properties also change

CE

correspondingly. When the bias voltage is -100 V, the coating has a composite structure of typical

AC

nanocystallite/amorphous, and a small amount of MAX phase of Ti3SiC2. Moreover, the protection potential applied on the coating effectively prevent the electrochemical corrosion of the coating. However, the applied protection potential will accelerate the degradation of the coating when the channel formed between the surface of the wear track and substrate.



Corresponding author. E-mail address: [email protected] (J.Li).

ACCEPTED MANUSCRIPT Abstract The TiSiCN coatings deposited at different bias voltages were fabricated on Ti6Al4V alloy by arc ion plating. The structure and properties of the TiSiCN coating were characterized using scanning electron microscope, X-ray diffraction, X-ray photoelectron spectroscopy, transmission electron

PT

microscopy, nanoindentation, potentiostat and ball-on-plate wear tests. As the bias voltage increases,

RI

the TiSiCN coating shows a nanocystallite/amorphous structure, whereas the phase constitution and

SC

grain size changed slightly, and its hardness and tribocorrsoion properties also change correspondingly. When the bias voltage is -100 V, the coating has a composite structure of typical

NU

nanocystallite/amorphous, and a small amount of MAX phase of Ti3SiC2. Moreover, the protection

MA

potential applied on the coating effectively prevent the electrochemical corrosion of the coating. However, the applied protection potential will accelerate the degradation of the coating when the

D

channel formed between the surface of the wear track and substrate.

PT E

Keywords: TiSiCN coating; Bias voltage; Arc ion plating; Structure; Tribocorrosion properties; 1. Introduction

CE

Ti6Al4V alloy has been widely used in vessel, shipbuilding and key maritime industrial parts

AC

due to its excellent combination of high strength-to-weight ratio and good corrosion resistance [1]. However, the applications of Ti6Al4V are limited due to their low wear resistance and high friction coefficient [2]. Therefore, the specific coating materials consisting of high hardness and self-lubricating are expected to improve the properties of Ti6Al4V. It is well known that the wear-corrosion performance of key components can be improved by nanostructure hard coating. In recent years, some researchers have fabricated the quaternary TiSiCN coating by various techniques, because the coating has ultrahigh hardness and outstanding wear

ACCEPTED MANUSCRIPT resistance. S.L. Ma et al. [3, 4] reported that the TiSiCN coating with Si concentration of ~12 at.% and C concentration of ~30 at.% fabricated by plasma enhanced chemical vapor deposition exhibits high hardness, low friction coefficient and wear rate. The TiSiCN coating deposited at higher power densities exhibiting excellent corrosion resistance has been reported [5]. Moreover, the coating was

PT

fabricated by arc ion plating technique possessing inherent benefits of high ionization rates and good

RI

adhesion compared with magnetron sputtering [6]. Above all, we believe that the TiSiCN coating

SC

consisting of specific nanostructure (nc-TiN/TiC and a-Si3N4/SiC) has great potential application value in marine environment [7].

NU

The Mn+1AXn phase has fascinating mechanical properties due to the combination of ceramic and

MA

metallic characteristics properties [8]. The Ti3SiC2 with particular lamellar structure is widely considered to have a similar structure with the graphite, and it exhibits excellent self-lubricating

D

tribological properties (ultra-low friction coefficient of 0.005) under low-micrometer-scale frictional

PT E

condition. Unfortunately, friction coefficient cannot maintain at a low value in normal loads because the Ti3SiC2 can easily delaminate and kink resulting in serious wear [9-11]. In addition, Ti3SiC2

CE

exhibits low hardness that cannot resist deformation resulting in serious wear and delamination

AC

failure [12]. As the lubrication phase, the Ti3SiC2 is incorporated into the superhard TiSiN coating to form the TiSiCN coating with an excellent wear resistance. In this study, the TiSiCN composite coatings with amorphous-nanocrystalline and MAX phase coupling structure are synthesized on the Ti6Al4V substrate using arc ion plating technique by conducting different bias voltages. The effects of bias voltage on microstructure and tribocorrosion property of the TiSiCN coating and the failure mechanism of the coating in corrosive medium are studied.

ACCEPTED MANUSCRIPT 2. Experimental 2.1 Deposition of TiSiCN coating The TiSiCN coatings were fabricated using arc ion plating technique (Hauzer Flexicoat 850) equipped with TiSi targets (90 at.% Ti, 10 at.% Si; purity 99.99 at.%) in gas mixture of C2H2, N2 and

PT

Ar. Ti6Al4V alloy (30 mm × 20 mm× 3 mm) was as the substrates which were grounded and

RI

polished to mirror, then ultrasonically cleaned for 15 minutes in acetone and ethanol, respectively.

SC

The substrates were mounted on a substrate holder at a distance of 10 cm from the targets. The chamber was pumped down to a background pressure below 4 × 10-3 Pa. And then the substrates

NU

were cleaned by Ti+ etching at bias voltages of -900 V, -1100 V and -1200 V, respectively, to remove

MA

thin oxide layer and other pollutants. A gas mixture of Ar and C2H2 was used for the deposition of a TiC interlayer from two Ti targets using a 60 A target current and a −70 V substrate bias to induce

D

MAX phase and enhance the adhesion of TiSiCN coating. Then, the TiSiCN coatings were deposited

PT E

from two TiSi targets with target current of 65 A in N2, C2H2 and Ar (420 sccm, 60 sccm, and ~470 sccm, respectively) mixture, and the work pressure was maintained at 0.3 Pa. Syntheses of MAX

CE

phases need to be provided high energy, and thus we set the same base temperature of 510 ℃ and

AC

different bias voltages of -20 V, -40 V, -60 V, -80 V and -100 V for the deposition of the TiSiCN coatings. The detailed process parameters are listed in Table 1. 2.2 Coating characterization The cross-sectional images of the coating were observed using a field emission scanning electron microscope (FE-SEM, Hitachi S4800). The crystal structure of the coating were characterized by X-ray diffraction (Bruker D8 x-ray facility) using Kα radiation (λ = 0.154 nm) operated at 40 kV and 40 mA. The scanning angle ranged from 10° to 100° at a scanning speed of 4°/min with 0.02° step

ACCEPTED MANUSCRIPT size. X-ray photoelectron spectroscopy (XPS) (Kratos Axis UltraDLD) using an Al (mono) Kα X-ray source was used to investigate the chemical composition and state after eliminating surface contaminations by argon ion etching. The microstructure of TiSiCN coating was observed in details by high resolution transmission electron microscopy (HR-TEM) using a FEI Tecnai F20. The

PT

hardness and elastic modulus were performed by a nanoindenter (MTS G200) with a Berkovich

RI

diamond indenter and using the continuous stiffness measurement (CSM) mode. The hardness and

SC

elastic modulus were measured based on the model of Oliver and Pharr from the load-displacement curves [13]. The residual stress of the coatings was measured by a stress tester (J&L Tech) by means

NU

of calculating the change in radius of curvature of the coating/substrate using Stoney's equation.

MA

2.3 Tribocorrosion testing

The tribocorrosion tests of the TiSiCN coatings against SiC balls with a diameter of 6 mm were

D

performed in artificial seawater using a reciprocating ball-on-plate tribometer (Rtec) and a

PT E

Potentiostat (Modulab) at room temperature of about 18 ± 3 ℃ and relative humidity of 52 ± 5% keeping the normal load of 5 N, the sliding speed of 20 mm/s and the wear track length of 5 mm. The

CE

coatings serve as working electrode and its potential can be monitored using potentiostat, and

AC

platinum and saturated Ag/AgCl were as the counter electrode and the reference electrode respectively. The tribocorrosion experiments were performed at open circuit potential (OCP) and cathodic protection (CP) conditions. The potentials, currents and friction coefficient were recorded before, during and after the wear process. Prior to the tribocorrosion tests, all samples were immersed in artificial seawater for 30 minutes. Potentiodynamic polarization measurements were carried out from −700 to +500 mV at a sweep rate of 2 mV s-1 under sliding and corrosion-only conditions when the tests were relatively stable. In order to simulate the fluid dynamics in

ACCEPTED MANUSCRIPT corrosion-only experiments, the ball was placed above (~5 mm) the coating surface and slid without any contact with the specimen surface for potentiodynamic polarization measurements. Three tests were conducted for each coating. The morphology of wear track of the TiSiCN coatings was observed by field emission scanning electron microscope (FEI Quanta FEG 250) equipped with EDS

PT

(OXFORD X-Max). The depth profiles of wear tracks and the roughness of the coating were

RI

examined using a surface profilometer (Alpha-Step IQ) by taking average measurements along the

SC

wear tracks. Then, the formula K=V/FS was used to calculate the volume loss rates, where the V is the volume of the wear loss of the coatings, S is the total sliding distance and F is the normal load.

NU

3. Results and discussion

MA

3.1 Microstructure and composition

Fig.1 shows the typical cross-sectional images of the TiSiCN coatings with different bias voltages.

D

It is clear that many droplets embed in the coatings, which is the feature of the ion plating technique

PT E

[14]. The TiSiCN coating deposited at -20 V shows slightly loose columnar structure. As the bias voltage increases to -100 V, a dense and compacted structure is found. This dense structure is due to

CE

the high kinetic energy of the positive ions from the targets, which fills the voids among grains and

AC

peens on the depositing coating under a high bias voltage. The thickness and roughness of the TiSiCN coatings with different bias voltages are listed in Table 2. As seen in Table 2, with the increase of the bias voltage, the thickness first increases and then decreases, and the maximum thickness of 4.11 um was obtained at -60 V. The fluctuation of the thickness of the coating may be related to the ion energy. The higher bias voltage will lead to greater attraction of substrates to ions. The coating harvesting more incoming ions in unit time results in the increase of deposition rate. However, with further increase of the bias voltage, the deposition rate gradually decreases due to

ACCEPTED MANUSCRIPT densification of the coating and re-sputtering [15]. The decrease of the deposition rate may also be due to an increase in the energy of the impinging ions with bias voltage increasing resulting in the increase of the defect density so as to restrict the local epitaxial growth of individual grains. [16] As shown in Table 2, the roughness of the TiSiCN coating slightly decreases from 0.148 μm to 0.132 μm

PT

with the bias voltage increase from -20 V to -100 V. This may be due to that a large amount of ion

RI

bombardment at higher bias voltage results in enhanced etching to the asperities and thus smooths

SC

the coating surface [17]. Moreover, with the bias voltage increase, the ion-irradiation effect strengthens, which enhances surface atomic mobility and decreases the atomic shadowing resulting

NU

in smoother surfaces [18].

MA

XRD patterns of the TiSiCN coatings at various bias voltages are shown in Fig. 2. Ti(C, N) and Ti are only crystalline phases detected in TiSiCN coatings except the peaks from substrate. The peaks at

D

38.4° and 40.2° can be attributed to Ti phase ( PDF # 44-1294), in which the Ti peaks in the patterns

PT E

are owing to the droplets of evaporation from TiSi targets [19]. It’s hard to differentiate TiC and TiN since they have NaCl-type (B1) face-centered cubic (FCC) structure and the atom radius of C and N

CE

is similar which results in that N atoms are substituted by C atoms in the TiN structure or C atoms

AC

are substituted by N atoms in the TiC structure. The XRD patterns exhibit the similar (111), (200), (220) and (222) peaks at about 36.1°, 42.0°, 61.2° and 76.8° diffraction angles, which are between TiN peaks ( PDF # 87-0629) and TiC peaks ( PDF # 71-0298), respectively. And the XRD patterns do not exhibit an obvious preferred orientation. Whereas, Yao et al. [20] reported that the TiSiN coating has obvious preferred orientation of TiN (200) with the increase of the bias voltage, because the energy of the coating tend to be at the lowest state. The total energy consists of strain energy and surface energy, of which the strain energy of (111) is the lowest, while the surface energy of (200) is

ACCEPTED MANUSCRIPT the lowest in B1 FCC structure. The coating preferring a (111) orientation at a low bias voltage, since the largest quantity of atoms per unit area can be incorporated at a low energy site. The increasing of ion energy and mobility induce depositing by lowest surface energy (200) plane with the increase of the bias voltage. In addition, the variation of densest planes and interplanar crystal spacing by bias

PT

voltage may play an important role in the change of diffraction planes pointed by Kong et al [21].

RI

However, there is only TiN crystal in their TiSiN coating compared with TiSiCN coating comprising

SC

of two TiC and TiN crystals. The C atoms and N atoms can replace each other in the crystal structure and the ratios of TiN/TiC can self-adjust when the coatings suffer high-energy ion bombardment in

NU

the deposition process. Hence, the lattice deformation and binding energy of two crystals can absorb

MA

and transfer the energy which comes from high bias voltage so that the coating maintains at a lower energy state. The peaks of Ti(C, N) shift to lower angles with the increase of bias voltage. This shift

D

can be attributed to the formation of more TiC phase. This result agrees well with the above analysis.

PT E

However, the atom is replaced by another atom forming substitutional solid solution or the atom occupying interstitial positions forming interstitial solid solution, which form the defective crystals

CE

and high residual stress in the TiSiCN coating [4]. The residual stress of the TiSiCN coatings was

AC

characterized and listed in Table 2. The residual compressive stress witnesses an increase on the whole, although the residual stress shows a little decrease as the bias voltage increases from -80 V to -100 V. The residual compressive stress increases from 1.092 GPa at -20 V to 2.011 GPa at -80 V, which may be due to the presence of solid solution and more point defect (Frenkel pairs and anti-Schottky defects) inducing crystal lattice deformation under high energy ion bombarding at high bias voltage [4, 21]. However, at the bias voltage of -100 V, the residual stress slightly decreases compared with the coating deposited at -80 V may be owing to the formation of some Ti3SiC2 whose

ACCEPTED MANUSCRIPT hardness is ~3.6 GPa which benefits to stress release for the TiSiCN coating [22]. Furthermore, Zhang et al. [23] pointed out that low deposition rate results in the coating oxidation, which leads to its amorphous structure and stress relaxation. In addition, D.F. Franceschini et al. [24] found that the substitution of nitrogen atoms by carbon atoms in the coating causes an increase of residual stress,

PT

which agrees well with the chemical composition of the TiSiCN coating listed in Table 3. The

RI

average grain size of Ti(C, N) at (111), (200), (220) and (222) planes was measured by Scherrer

SC

equation and the results were listed in Table 2 [25]. The grain size decreases from ~46 nm to ~23 nm with the bias voltage increase from -20 V to -80 V, which is ascribed to that the high-energy ion

NU

bombardment can inhibit columnar crystal growth. Meanwhile, the increase of micro-defects results

MA

in more nucleation sites. At higher bias voltage, there are more ion-irradiation-generated defects which can decrease the energy needed for nucleation and make nucleation more competitive relative

D

to grain growth [26]. At the bias voltage of -100 V, the grain size is a little bigger than that deposited

PT E

at -80 V, which is owing to ions with higher energy exhibiting higher mobility, and thus the ions migrate to the grain boundaries [21]. From the X-ray diffraction patterns, the Si3N4 and SiC phases

AC

the coating.

CE

are not be found in the TiSiCN coatings and this implies the silicon may be as amorphous phase in

The chemical states of the TiSiCN coatings were investigated by XPS. The core levels spectra for the Ti 2p, Si 2p, C 1s and N 1s are presented in Fig. 3. In order to identify the steady-state compositions the coating surface were sputtered by argon ion at 2 keV for 5 min before measurements. The Ti 2p spectra can be fitted using four pairs of peaks, and one of them located at 454.1 and 459.9 eV are pure Ti [27]. This is due to that the titanium oxides on the surface were removed by sputtering and thus the Ti 2p spectra reveal the pure Ti from the droplets, which agrees well with the XRD results.

ACCEPTED MANUSCRIPT The peaks at 454.8 and 460.5 eV are corresponding to Ti(C, N), which can be attributed to that Ti atoms may bond to both C and N atoms, and TiN and TiC have close value of binding energies, and the C and N atoms may substitute each other in crystal lattice [2, 28]. The peaks located at 456.3/451.8 eV and 458.1/463.5 eV are in agreement with the Ti2O3 and TiO2, respectively [29].The

PT

Si 2p spectra (Fig. 3b) were composed of Si-N (BE = 101.4 eV) and Si-C (BE = 100.2 eV) [30, 31].

RI

From XRD results there are no crystal phases related to Si, so it can be determined that Si exists as

SC

amorphous phases of Si3N4 and SiC in the TiSiCN coating deposited at various bias voltages. The C 1s spectra of the as-deposited TiSiCN coatings are deconvoluted using five peaks centered at 281.1,

NU

281.8, 283.3, 284.8 and 286.4 eV, which are attributed to C-Si [32], C-Ti [27], C-H [33], C-C [29.]

MA

and C-N bonds [14], respectively. The C-H is due to that the C2H2 gas can release C ions and H ions when it absorbs enough energy (Fig. 3c). The N 1s spectra of the coatings could be fitted by three

D

peaks located at 396.1, 397.7 and 399.0 eV, which attribute to Ti-N [7], Si-N [7] and N-C bonds [14],

PT E

respectively (Fig. 3d). Ti–N bonds shift slightly toward lower binding energy, which may be owing to the presence of C [14]. Meanwhile, the N-C bond can confirm the replacement between N and C

CE

atoms on the other side. The chemical composition of the TiSiCN coatings are measured by XPS and

AC

summarized in Table 2. It is obvious that the atomic concentrations of Ti and Si decrease slightly, whereas the C content increases gradually, as the bias voltage increases. This is ascribed to that the high bias voltage provides higher energy in the chamber, which is beneficial to the decomposition of carbon source gas due to that the bond breaking energy of C2H2 is higher than N2. Meanwhile, intensely re-sputtering occurs at high bias voltage. In addition, the binding between non-metal and non-metal is stronger than that between metal and non-metal. The energy of ion bombardment is strong, the metallic atoms may re-sputter from the surface of coating, which is in accordance with the

ACCEPTED MANUSCRIPT report by Y. Zhang et al. [34] who have pointed out that re-sputtering of Si and Ti elements which can result in Si and Ti contents decrease. Moreover, C.L. Chang [35] have reported that chemical affinity of the C is higher than N, resulting in that the C easily combines with Ti and Si atoms. The high resolution TEM images, together with the selected area electron diffraction (SEAD)

PT

patterns of the TiSiCN coatings are shown in Fig. 4. The coatings deposited at various bias voltages

RI

all exhibit nanocrystal TiN/TiC embedded in the amorphous Si3N4 and SiC matrix. As shown in Fig.

SC

4a, the coating fabricated at -20V shows a clear crystal orientation of TiN (111)and TiC (200), and Ti(C, N) (200), of which the deformed lattice of Ti(C, N) shows (200) lattice spacing of ~ 2.14 Å

NU

caused by high energy ion bombardment resulting in the originally atoms replaced by other atoms.

MA

The crystal planes (200) spacing is larger than the (111) planes’, and thus the atoms (N or C) of the (200) planes can be easily replaced by other atoms (C or N) at relative low bias voltage. It also shows

D

special nanostructure consisted of the TiN (200) and TiN (111) crystals, which can form

PT E

semi-coherent low-angle grain boundaries that will increase the toughness of the coating. As the bias voltage increases to -60 V, more deformed lattice can be found and the coating exhibits the Ti(C, N)

CE

(111) phase compared with the coating deposited at -20V in Fig. 4b. Meanwhile, the selected-area

AC

electron diffusion (SAED) pattern reveals a polycrystalline cubic TiN (TiC) structure, in which the discrete diffraction spots of the (111) and (200) are identified. The coating exhibits a typical nanocystallite/amorphous microstructure, in which the Ti(C, N) (111), Ti(C, N) (200) and TiC (111) crystals embeds in amorphous matrix, when the bias voltage increases to -100 V (Fig. 4c). It was observed that the crystallographic texture of the coating exhibits a gradual transformation, of which the main TiN and a little deformed lattice of Ti(C, N) in -20 V, TiN, TiC and deformed lattice of Ti(C, N) in -60 V, and the main deformed lattice of Ti(C, N) and a little TiC in -100 V, which is consistent

ACCEPTED MANUSCRIPT with our aforementioned XRD inference. 3.2 Mechanical properties The hardness and elastic modulus of the TiSiCN coatings are shown in Fig. 5. The hardness and elastic modulus of the TiSiCN coating gradually increase from 28.3 to 38.4 GPa and 307.1 to 350.0

PT

GPa, respectively, as the bias voltage increases from -20 to -100 V. Energetic ion bombardment

RI

caused by high bias voltage leads to lattice deformation and more defects forming in the coating,

SC

which can fix the dislocations and result in dislocation pile-up, thus improving the hardness. And it can also cause grain refining, which is in accordance with the XRD grain size listed in Table 2. The

NU

fine crystal embeds in the amorphous matrix that can hinder the movement and slip of the

MA

dislocations, which can also improve the hardness [7]. The grain size of the coating deposited at -100 V is slightly bigger than that at -80 V, but the coating still exhibits a higher hardness compared with

D

the coating fabricated at -80 V. It may be due to that the fine crystals alternatively distribute with

PT E

amorphous phases (Fig. 6), forming alternating stress field and restricting the motion of dislocations across the interfaces more effectively [23]. H/E and H3/E*2 mechanical indexes (where E* = E/ (1 -

CE

ν2) is the effective elastic modulus and ν is the Poisson’s ratio) are ranking parameters to evaluate the

AC

mechanical performance of the coating. The H/E index is used to characterize the ability of resistance elastic strain, and the H3/E*2 is considered as the key parameter of resistance plastic deformation. In general, the coating with high hardness and low elastic modulus implies a reduced contact pressure through redistribution of the applied load over a large area, delaying failure of the coating and resisting crack of the coating [5]. As shown in Fig. 5, the value of H/E and H3/E*2 of the TiSiCN coatings increase from 0.226 to 0.433 and 0.092 to 0.11, respectively, as the bias voltage increases from -20 to -100 V. In addition, residual compressive stress also plays an important role in

ACCEPTED MANUSCRIPT restraining crack, which may influence the hardness of the coating. 3.3 Tribocorrosion properties The potentiodynamic polarization curves for the Ti6Al4V substrate and TiSiCN coatings at ball motion with non-contact and sliding condition are depicted in Fig. 7a and b, respectively, and the

PT

results using Tafel extrapolation method are listed in Table 3. As shown in Fig. 7a, the coating

RI

deposited at -20 V shows a low corrosion potential (Ecorr = − 181.2 mV) and a higher corrosion

SC

current density (Icorr = 29.75 × 10−7 A/cm2) at non-contact condition due to its loose columnar structure. With the increase of the bias voltage, the corrosion potential increases and corrosion

NU

current density decreases progressively. The coating deposited at -100 V exhibits the best corrosion

MA

resistance ascribed to more dense and grain refinement of the coating by ion bombardment. The Ti6Al4V exhibits a low corrosion current density of 8.99 × 10−7 A/cm2 which is between the TiSiCN

D

coatings deposited at -20 V and -100 V, indicating good corrosion resistance at ball motion with

PT E

non-contact condition. This may be due to the Ti6Al4V substrate surface forming a high stability passive film of titanium oxide [36]. For all samples, the Ecorr and Icorr measured at non-contact

CE

condition are much nobler than it taking at sliding condition. In particular, the corrosion current

AC

density of the Ti6Al4V substrate increases to 62.27 × 10−7 A/cm2 and the corrosion potential decreases to -633.4 mV, which indicates that the Ti6Al4V substrate exhibits poor corrosion resistance at sliding condition. S.A. Naghibi et al. [37] pointed out that wear can accelerate corrosion, and the synergy effect of wear and corrosion can further increase the degradation of the materials. J. Chen et al. [38] found that wear-accelerated corrosion plays an important role in material loss in tribocorrosion. This may be due to mechanical wear destroying the passive film which results in the increase of corrosion current density and the decrease of corrosion potential.

ACCEPTED MANUSCRIPT Fig. 8a shows the open circuit potential (OCP) values recorded versus time before, during and after sliding for the Ti6Al4V substrate and the TiSiCN coatings in artificial seawater. When the sliding starts the potential drops sharply to a more negative value, which attributes to the passive film covering on the coating destroyed by mechanical wear and exposing active regions on wear track.

PT

Afterward, the whole potential witnesses a decrease with fluctuation, because the friction is in a

RI

running-in period due to high surface roughness that mechanical depassivation is irregular. Then, a

SC

gradually shift in potential and reach a relatively steady-state are observed as long as sliding. This implies that mechanical depassivation exceeds the electrochemical repassivation in this period. The

NU

coatings deposited at different bias voltages have a similar evolutionary trend, and the potential drop

MA

is listed in Table 3. Sh. Hassani et al. [39] pointed out that high hardness can resist the mechanical destruction of passive film during the tribocorrosion tests. K. Kato [40] found the relationship

D

between hardness and wear loss, indicating that the requirement for excellent wear resistance was

PT E

high hardness. Certainly, a high hardness results in a relative narrow width of wear track so that the ratio of active to passive areas is small, and thus the potential drop is low, and vice versa. In addition,

CE

the intrinsic corrosion resistance of coating may also influence the potential drop. Thus, the coating

AC

deposited at -100 V exhibits the smallest potential drop, which is due to its best corrosion resistance and high hardness, while the Ti6Al4V substrate has large potential drop owing to its lower hardness and poor wear resistance [41]. When the load was removed from the coating, the potential rises sharply at initial period and reaches the initial values eventually, which indicates repassivation occurred after sliding. Fig. 8b shows the current density recorded versus time before, during and after sliding for the TiSiCN coatings under cathodic protection (CP) condition in artificial seawater. Appropriate cathodic potential can protect the coating from corrosion, and the excessively negative

ACCEPTED MANUSCRIPT protection potential will make the coating suffer hydrogen embrittlement. On the contrary, the higher protection potential applied on the coating cannot compensate the loss of electron of the coating in tribocorrosion tests [42]. The protection potentials applied on the coatings were chosen by reference the potential measured at OCP condition listed in the Table 3. Under CP condition, it can exclude the

PT

factor of corrosion and only mechanical wear occurred in tribocorrosion tests [43]. A significant rise

RI

of the current density of the coatings is noticed at the start of the sliding, which indicates that wear

SC

accelerates corrosion and degradation of materials. For Ti6Al4V substrate, the current density with a downtrend is also in the cathodic region, which means that it can protect the Ti6Al4V substrate. At

NU

the end of sliding, the current density returns to its initial value before the start of the sliding.

MA

Fig. 9 presents the friction coefficient and volume loss rates of the Ti6Al4V substrate and the TiSiCN coatings deposited at different bias voltages against SiC ball at OCP and CP conditions. It is

D

obvious that the Ti6Al4V exhibits higher friction coefficient both at OCP and CP conditions

PT E

compared with the substrates coated TiSiCN coating. It means that the TiSiCN coatings possess friction reduction ability to some extent. The friction coefficient of the TiSiCN coatings first

CE

increases and then decreases at the OCP condition as the bias voltage increases. The coating

AC

deposited at -60 V has the highest friction coefficient of about 0.37. As seen in Fig. 11c, there are more wear debris accumulated and fine furrows compared to Fig 11a and e, this can be explained that abrasive wear occurred in sliding process which results in the increase of the friction coefficient. With the bias voltage further increase, the friction coefficient gradually decreases and the friction coefficient of the coating deposited at -100 V is about 0.32. This may be ascribed to some Ti3SiC2 MAX phase forming which can act as solid lubricant in the TiSiCN coating (Fig. 10). Yi Zhang et al. [22] reported that Ti3SiC2 ceramic exhibited very low friction coefficient sliding against diamond

ACCEPTED MANUSCRIPT pairs. However, S. Gupta et al. [44] revealed that the friction coefficient of Ti3SiC2 is low at the initial period, but it rapidly increases in tribological tests. Furthermore, J. Emmerlich et al. [11] also pointed out that the basal plane rupture and kink formation through remove surface parts, which can increase the friction coefficient. However, nanocystallite/amorphous microstructure can act as

PT

skeleton guaranteeing the hardness of the coating, and filling in some Ti3SiC2 MAX phase can act as

RI

a lubricating in the TiSiCN coating, which is different from the above mentioned pure Ti3SiC2. From

SC

Fig. 10, it can be observed that the Ti3SiC2 phase exists in special place in the TiSiCN coating, of which one is sandwiched between TiC (111) and TiC (200) (Fig. 10a), and another was surrounded

NU

by amorphous phase (Si3N4/ SiC) (Fig. 10b). These special locations provide the material basis for

MA

generating Ti3SiC2 MAX phase. In addition, power assurance was provided by the intrinsic high ionization rate, high temperature (510 ℃) and high bias voltage of the cathodic arc deposition. Thus,

D

some lubricating Ti3SiC2 phase can form in deposition process. The friction coefficient at CP

PT E

condition follows a similar trend except the values are slightly higher than that at OCP condition as shown in Fig. 9. J. Takadoum et al. [45] have found that Si3N4 and SiC can react with H2O forming

CE

Si(OH)4 tribo-film which can act as a lubricant in sliding process. J. Wang et al. [46] have pointed

AC

out Mg(OH)2 and CaCO3 will cover on the counterpart, which are key factors to reduce friction coefficient in seawater. Thus, the protection potential applied on the coating may influence the friction chemical reaction and inhibit the formation of the lubricating film. The volume loss rates of Ti6Al4V substrate and TiSiCN coatings with different bias voltages against SiC ball in OCP and CP conditions are shown in Fig. 9. The Ti6Al4V substrate exhibits poor wear resistance and the volume loss rate is about 2-order of magnitude higher than the TiSiCN coating’s. This fluctuant trend of the coatings is consistent with the variations of H/E and H3/E*2

ACCEPTED MANUSCRIPT ratios. The coating deposited at -20 V exhibits the highest volume loss rate as a result of the loose column structure. With the increase of the bias voltage, the hardness of the coating gradually increases and the structure is denser, which is beneficial to tribocorrosion resistance [39]. For the same coating, the volume loss rate at OCP condition is higher than that at CP condition, indicating

PT

that mechanical wear and electrochemical corrosion accelerate the coating degradation in seawater.

RI

However, the corrosion loss accounts for a small fraction compared with wear loss, which indicates

SC

that mechanical wear is mainly responsible for the material failure.

Fig. 11 shows the SEM images of wear tracks of the typical TiSiCN coatings at OCP or CP

NU

condition. The coating deposited at -20 V can be observed that the surface of the wear track are

MA

polished smooth at OCP and CP condition, indicating that the wear mechanism is mainly polishing wear (Fig. 12a and b). Simultaneously, partial delamination can be found on the wear track and this

D

can be ascribed to that the micro-defects of the coating may form stress concentration on sliding

PT E

process which can induce plastic deformation and promote the initiation of delamination [47]. And some oxide layer flakes can be found in Fig. 12b and the EDS analyses (Table 4) reveal that the Ti,

CE

Si, C, N and O are detected on the area marked by rectangle A and the Ti, Si, C, N, O, Na, Mg and K

AC

are detected on the area B. A high concentration of O element indicating that oxidation occurred during sliding forming oxide layer flake on the area marked by rectangle B compared with A, which indicates an adhesive wear. Na, Mg and K elements are identified on the wear track, which indicates that the deposition of elements in the artificial seawater during tests. According to Fig. 12c and d, some wear debris and fine furrows paralleling to the sliding direction of the coatings can be clearly observed, indicating the wear mechanism is abrasive wear. Fig. 12e shows smooth surface of the wear track, while Fig. 12f shows deep and rough grooves paralleling to the sliding direction. This

ACCEPTED MANUSCRIPT results indicating different wear mechanisms, the former is mainly polishing wear and the latter is mainly abrasive wear. In addition, the micro-cracks in the vicinity of the droplets removed on the wear track of the coatings can be found, which can be ascribed to that micro-cracks initiate and propagate from weak area when suffering alternating stress.

PT

The morphology of the wear tracks of -20 V shows delamination at CP condition, while it is not

RI

found at OCP condition through a close inspection of the whole wear track. Furthermore, for the

SC

coating deposited at -60 V, more corrosion pits can be found on the wear track at CP condition than it at the OCP condition. Generally, it is effective to compensate for electrons loss which balances the

NU

anodic dissolution by electrochemical corrosion and improves the repassivation rate and further

MA

protects the coating at CP condition shown in Fig. 12a. However, plastic deformation occurs during repetitive sliding tests, resulting in cracks originating from surface defects and propagating at

D

sub-surface. When seawater intrudes into the inner along the cracks or crystal boundaries and then

PT E

the channel between the surface of the wear track and substrate forms (Fig. 12b). Then, potential difference exists between the upper and lower surfaces of the coating, owing to Ti6Al4V with a low

CE

potential (˗770 mV) compared with the coating. This has two effects, one is galvanic corrosion

AC

occurred between substrate and coating, resulting in the consumption of Ti6Al4V and the worse adhesion between the coating and substrate. The appropriate protection potential is applied on the coating but it is too high for substrate, which has an adverse impact on the substrates occurring anodic dissolution even oxygen-consuming corrosion. This is another effect. The above two points result in the formation of corrosion pits. Even worse, when the cracks propagate and connect with the adjacent one, the delamination occurs. According to SEM images of the coating, the corrosion character becomes inconspicuous with the increase of the bias voltage, resulting from the high

ACCEPTED MANUSCRIPT hardness and H3/E*2 value deposited at -100 V which can benefit to plastic deformation resistance. Another reason attributes to the residual compressive stress, which can restrain the crack propagation [48]. 4. Conclusion

PT

The TiSiCN coatings deposited at different bias voltages were fabricated on Ti6Al4V alloy using

RI

arc ion plating technique. At the bias voltage of -20 V, the coating consists of coarse TiN and a little

SC

deformed lattice of Ti(C, N) which embed in amorphous matrix (Si3N4 and SiC), exhibiting lower hardness and poor wear resistance. For the coating deposited at -60 V, it is characterized as TiN, TiC

NU

and deformed lattice of Ti(C, N) that embed in amorphous matrix, and its deposition rate and

MA

hardness increase, while the friction coefficient rises. At the bias voltage of -100 V, the coating exhibits composite structure of typical nanocystallite/amorphous and some Ti3SiC2 MAX phase

D

which not only maintains a high hardness (~ 38.4 GPa), but also acts as a lubricating, and that the

PT E

coating exhibits an excellent tribocorrosion resistance. In addition, the applied protection potential is effective to protect the coating from electrochemical corrosion. However, it accelerates failure of the

AC

corrosion channel.

CE

coating, when the seawater invaded into the coating along with the cracks resulting in forming

Acknowledgements

This work has been supported by the National Natural Science Foundation of China (Grant No. 51575510), Zhejiang Provincial Natural Science Foundation of China (LY14E010005).

ACCEPTED MANUSCRIPT

References [1] M. Jahedi, S. Zahiri, S. Gulizia, B. Tiganis, C. Tang, D. Fraser, Direct manufacturing of titanium parts by cold spray, Mater. Sci. Forum. 618 (2009) 505–508.

PT

[2] P.D. Miller, J.W. Holliday, Friction and wear properties of titanium, Wear 2 (1958) 133-140.

RI

[3] S.L. Ma, D.Y. Ma, Y. Guo, B.Xu, G.Z. Wu, K.W. Xu, Paul K. Chu, Synthesis and characterization

SC

of super hard, self-lubricating Ti–Si–C–N nanocomposite coating, Acta Mater. 55 (2007) 6350–6355. [4] D.Y. Ma, S.L. Ma, H.S. Dong, K.W. Xu, T. Bell, Microstructure and tribological behaviour of

NU

super-hard Ti–Si–C–N nanocomposite coating deposited by plasma enhanced chemical vapour

MA

deposition, Thin Solid Films 496 (2006) 438–444.

[5] A. M. Abd El-Rahman, R. Wei, Effect of ion bombardment on structural, mechanical, erosion and

D

corrosion properties of Ti–Si–C–N nanocomposite coating, Sur. Coat. Tech. 258 (2014) 320–328.

PT E

[6] C.H. Hsu, K.L. Chen, Z.H. Lin, C.Y. Su, C.K. Lin, Bias effects on the tribological behavior of cathodic arc evaporated CrTiAlN coating on AISI 304 stainless steel, Thin Solid Films 518 (2010)

CE

3825–3829.

AC

[7] E. Thangavel，S. Lee，K.S. Nam，J.K. Kim，D.G. Kim, Synthesis and characterization of Ti-Si-C-N nanocomposite coating prepared by a filtered vacuum arc method, Appl. Surf. Sci. 265 (2013) 60-65.

[8] M.W. Barsoum, The Mn+1AXn Phases: a new class of solids; Thermodynamically Stable Nanolaminates, Prog. Solid State Chem. Ch. 28 (2000) 201-281. [9] A. Crossley, E.H. Kisi, J.W.B. Summers, S. Myhra, Ultra-low friction for a layered carbide-derived ceramic, Ti3SiC2, investigated by lateral force microscopy (LFM), J. Phys. D Appl.

ACCEPTED MANUSCRIPT Phys. 32 (1999) 632-638. [10] S. Myhra, J.W.B. Summers, E.H. Kisi, Ti3SiC2—a layered ceramic exhibiting ultra-low friction, Mater. Lett. 39 (1999) 6-11. [11] J. Emmerlich, G. Gassner, P. Eklund, H. Högberg, L. Hultman, Micro and macroscale

PT

tribological behavior of epitaxial Ti3SiC2 thin films, Wear 264 (2008) 914-919.

RI

[12] B.J. Kooi, R.J. Poppen, N.J.M. Carvalho, J.T.M.D. Hosson, M.W. Barsoum, Ti3SiC2: A damage

SC

tolerant ceramic studied with nano-indentations and transmission electron microscopy, Acta Mater. 51 (2003) 2859-2872.

NU

[13] W.C. Oliver, G.M.J. Pharr, An Improved Technique for Determining Hardness and Elastic

MA

Modulus Using Load and Displacement Sensing Indentation, J. Mater. Res. 7 (1992) 1564-1583. [14] A.O. Eriksson, J. Zhu, N. Ghafoor, J. Jensen, G. Greczynski, Ti–Si–C–N thin films grown by

D

reactive arc evaporation from Ti3SiC2 cathodes, J. Mater. Res. 26 (2011) 874-881.

PT E

[15] W.J. Shen, M.H. Tsai, Y.S. Chang, J.W. Yeh, Effects of substrate bias on the structure and mechanical properties of (Al1.5CrNb0.5Si0.5Ti)Nx coating, Thin Solid Films 520 (2012) 6183–6188.

CE

[16] C.L. Chang, C.S. Huang, Effect of bias voltage on microstructure, mechanical and wear

4923–4927.

AC

properties of Al-Si-N coatings deposited by cathodic arc evaporation, Thin Solid Films 2011 (519)

[17] X.L. Peng, Z.H. Barber, T.W. Clyne, Surface rough of diamond-like carbon films prepared using various techniques, Surf. Coat. Technol. 138 (2001) 23–32. [18] I. Petrov, P.B. Barna, L. Hultman, J.E. Greene, Microstructural evolution during film growth, J. Vac. Sci. Technol. A 21 (2003) S117-S128. [19] M. Tkadletz, C. Mitterer, B. Sartory, I.Letofsky-Papst, C. Czettl, C. Michotte, The effect of

ACCEPTED MANUSCRIPT droplets in arc evaporated TiAlTaN hard coating on the wear behavior, Sur. Coat. Tech. 257 (2014) 95–101. [20] Y.R. Yao, J.L. Li, Y.X. Wang, Y.W. Ye, L.H. Zhu, Influence of the negative bias in ion plating on the microstructural and tribological performances of Ti–Si–N coatings in seawater, Sur. Coat. Tech.

PT

2015 (280) 154-162.

RI

[21] Q.H. Kong, L. Ji, H.X. Li, X.H. Liu, Y.J. Wang, J.M. Chen, H.D. Zhou, Influence of substrate

SC

bias voltage on the microstructure and residual stress of CrN films deposited by medium frequency magnetron sputtering, Mater. Sci. Eng. B 176 (2011) 850-854.

NU

[22] Y. Zhang, G.P. Ding, Y.C. Zhou, B.C. Cai, Ti3SiC2—a self-lubricating ceramic. Mater. Lett. 55

MA

(2002) 285–289.

[23] Y.J. Zhang, Y.Z. Yang, Y.H. Zhai, P.Y. Zhang, Effect of negative substrate bias on the

D

microstructure and mechanical properties of Ti–Si–N films deposited by a hybrid filtered cathodic

PT E

arc and ion beam sputtering technique, Appl. Surf. Sci. 258 (2012) 6897-6901. [24] D.F. Franceschini, C.A. Achete, F.L. Freire, Internal stress reduction by nitrogen incorporation

CE

in hard amorphous carbon thin films, Appl. Phys. Lett. 1992 (60) 3229-3231.

AC

[25] V. Uvarov, I. Popov, Metrological Characterization of X-ray Diffraction Methods for Determination of Crystallite Size in Nano-Scale Materials, Mater. Charact. 58 (2007) 883-891. [26] J.J. Olaya, S.E. Rodil, S. Muhl, E. Sanchez, Comparative study of chromium nitride coating deposited by unbalanced and balanced magnetron sputtering, Thin Solid Films 474 (2005) 119-126. [27] J. Lauridsen, P. Eklund, J. Jensen, H. Ljungcrantz, Å. Öberg, E. Lewin, U. Jansson, A. Flink, H. HÖgberg, L. Hultman, Microstructure evolution of Ti–Si–C–Ag nanocomposite coating deposited by DC magnetron sputtering. Acta Mater. 58 (2010) 6592-6599.

ACCEPTED MANUSCRIPT [28] J.L. Lin, J.J. Moore, B. Mishra, M. Pinkas, W.D. Sproul, The structure and mechanical and tribological properties of TiBCN nanocomposite coating, Acta Mater. 58 (2010) 1554–1564. [29] C.Q. Dang, J.L. Li, Y. Wang, J.M. Chen, Structure, mechanical and tribological properties of self-toughening TiSiN/Ag multilayer coating on Ti6Al4V prepared by arc ion plating, Appl. Surf. Sci.

PT

386 (2016) 224–233.

RI

[30] A.O. Eriksson, N. Ghafoor, J. Jensen, L.Å. Näslund, M.P. Johansson, J. Sjölen, M. Odén, L.

SC

Hultman, J. Rosen, Arc deposition of Ti–Si–C–N thin films from binary and ternary cathodes — Comparing sources of C. Sur. Coat. Tech. 213 (2012) 145–154.

NU

[31] J.H. Jeon, S.R. Choi, W.S. Chung, K.H. Kim, Synthesis and characterization of quaternary

MA

Ti–Si–C–N coating prepared by a hybrid deposition technique, Sur. Coat. Tech. 188 (2004) 415– 419.

D

[32] J.R. Waldrop, R.W. Grant, Y.C. Wang, R.F. Davis, Metal Schottky barrier contacts to alpha

PT E

6H-SiC, J. Appl. Phys. 72 (1992) 4757-4760.

[33] W. Dai, G.S. Wu, A.Y. Wang, Structure and elastic recovery of Cr–C:H films deposited by a

CE

reactive magnetron sputtering technique, Appl. Surf. Sci. 257 (2010) 244-248.

AC

[34] X.J. Pang, L. Shi, P. Wang, G.A. Zhang, W.M. Liu, Influences of bias voltage on mechanical and tribological properties of Ti–Al–C films synthesized by magnetron sputtering, Sur. Coat. Tech. 203 (2009) 1537–1543.

[35] C.L. Chang, T.J. Hsieh, Effect of C2H2 gas flow rate on synthesis and characteristics of Ti–Si–C–N coating by cathodic arc plasma evaporation, J. Mater. Process. Tech. 209 (2009) 5521–5526. [36] V. Totolin, V. Pejaković, T. Csanyi, O. Hekele, M. Huber, M.R. Ripoll, Surface engineering of

ACCEPTED MANUSCRIPT Ti6Al4V surfaces for enhanced tribocorrosion performance in artificial seawater, Mater. Design 104 (2016) 10-18. [37] S.A. Naghibi, K. Raeissi, M.H. Fathi, Corrosion and tribocorrosion behavior of Ti/TiN PVD coating on 316L stainless steel substrate in Ringer's solution, Mater. Chem. Phys. 148 (2014)

PT

614-623.

RI

[38] J. Chen, Q. Zhang, Q.A. Li, S.L. Fu, J.Z. Wang, Corrosion and tribocorrosion behaviors of AISI

SC

316 stainless steel and Ti6Al4V alloys in artificial seawater, Trans. Nonferrous Met. Soc. China 24 (2014) 1022-1031.

NU

[39] S.H. Hassani, K. Raeissi, M. Azzi, D. Li, M.A. Golozar, J.A. Szpunarb, Improving the corrosion

MA

and tribocorrosion resistance of Ni–Co nanocrystalline coating in NaOH solution, Corros. Sci. 51 (2009) 2371-2379.

D

[40] K. Kato, Wear in relation to friction — a review, Wear 241 (2000) 151-157.

PT E

[41] F. Galliano, E. Galvanetto, S. Mischler, D. Landolt, Tribocorrosion behavior of plasma nitrided Ti–6Al–4V alloy in neutral NaCl solution, Sur. Coat. Tech. 145 (2001) 121-131.

CE

[42] K. Shin-Ichi, R. Maruyama, T. Misawa, Effect of Applied Cathodic Potential on Susceptibility

AC

to Hydrogen Embrittlement in High Strength Low Alloy Steel, ISIJ Int. Vol. 2003 (43) 475–481. [43] J. Geringer, B. Forest,P, Combrade, Wear analysis of materials used as orthopaedic implants, Wear 261 (2006) 971-979. [44] S. Gupta, D. Filimonov, T. Palanisamy, M.W. Barsoum, Tribological behavior of select MAX phases against Al2O3 at elevated temperatures, Wear 265 (2008) 560-565. [45] J. Takadoum, H. Houmid-Bennani, D. Mairey, The wear characteristics of silicon nitride, J. Eur. Ceram. Soc. 18 (1998) 553-556.

ACCEPTED MANUSCRIPT [46] J.Z. Wang, F.Y. Yan, Q.J. Xue, Tribological behavior of PTFE sliding against steel in sea water, Wear 267 (2009) 1634-1641. [47] J. Jiang, M.M. Stack, A. Neville, Modelling the tribo-corrosion interaction in aqueous sliding conditions, Tribol. Int. 35 (2002) 669-679.

PT

[48] E.L. Bourhis, P. Goudeau, M.H. Staia, E. Carrasquero, E.S. Puchi-Cabrera, Mechanical

AC

CE

PT E

D

MA

NU

SC

RI

properties of hard AlCrN-based coated substrates, Sur. Coat. Tech. 203 (2009) 2961-2968.

ACCEPTED MANUSCRIPT

Table 1 Deposition parameters for TiC and TiSiCN layers. Layer

TiC

TiSiCN

Substrate temperature (℃)

510

Target assignment

Two Ti targets (purity 99.99%);

Background pressure (Pa)

4×10

4×10

Work pressure (Pa)

1

3

Gas flow (sccm)

C2H2 (40), Ar (~350)

Cathode current (A)

60

Deposition time (minute)

10

Substrate bias (V)

-70

510

-3

SC

C2H2 (60), N2 (420), Ar (~470)

NU

65

PT E

D

MA

90

CE AC

RI

-3

PT

Two TiSi targets (90 at.% Ti, 10% at.% Si; purity 99.99%)

-20, -40, -60, -80, -100

ACCEPTED MANUSCRIPT

Table 2 Atomic concentration and properties of TiSiCN deposited at different bias voltages. Roughness, μm

Averaged Grain size, nm

Residual stress, GPa

46

-1.092 ± 0.0147

Si

C

N

O

-20 V

30.14

5.15

22.72

25.71

16.28

3.68

0.148

-40 V

30.12

4.49

24.44

25.71

16.53

4.05

0.142

33

-1.472 ± 0.0339

-60 V

29.33

4.14

26.90

23.97

15.66

4.11

0.129

28

-1.592 ± 0.0239

-80 V

27.13

3.64

27.24

23.72

18.27

3.56

0.130

23

-2.011 ± 0.0363

-100 V

26.74

3.95

28.28

23.84

17.19

3.25

0.132

26

-1.937 ± 0.0361

AC

CE

PT E

D

SC

NU

MA

Samples

PT

Ti

Thickness, μm

RI

Chemical composition, at. %

ACCEPTED MANUSCRIPT

Table 3 Open circuit potential, applied potential and Corrosion parameters extracted from the polarization curves in Fig. 7 using Tafel extrapolation method Ball motion with non-contact the coating

Sliding Open circuit potential (mV)

Applied potential (mV)

icorr -7

Ecorr

PT

Samples

Drop of potential (mV)

2

(×10 A/cm )

D PT E CE

-7

Ecorr 2

(mV) (×10 A/cm )

-633.4 -186.6 -189.2 -191.1 -191.5 -178.1

RI

62.27 39.63 14.66 7.98 6.60 4.74

SC

-870 -405 -378 -374 -373 -238

NU

-770 -305 -278 -274 -273 -238

MA

638 187 169 157 151 134

AC

Ti6Al4V -20 V -40 V -60 V -80 V -100 V

icorr

8.99 29.75 9.55 7.48 4.87 4.56

(mV) -289.1 -181.2 -163.6 -176.0 -158.8 -164.9

ACCEPTED MANUSCRIPT

B

Ti (at. %) Si (at. %)

42.0 5.4

32.6 4.1

C (at. %)

24.4

9.4

N (at. %)

24.3

4.3

O (at. %) K (at. %) Na (at. %) Mg (at. %)

3.9 -

48.3 0.2 1.0 0.2

RI

A

AC

CE

PT E

D

MA

NU

SC

Elements

PT

Table 4 EDS results of the area marked by rectangle on the worn surface from Fig. 12b.

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 1 Cross-section FESEM images of TiSiCN coatings deposited at various bias voltages: (a) -20 V, (b) -60 V, (c) -100 V.

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

Fig. 2 X-ray diffraction patterns of TiSiCN coatings as a function of the bias voltage.

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 3 X-ray photoelectron spectroscopy of TiSiCN coatings deposited at different bias voltages: (a)

AC

CE

PT E

D

MA

Ti 2p; (b) Si 2p; (c) C 1s; (d) N 1s.

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

D

Fig. 4 High resolution TEM images of TiSiCN coatings deposited at different bias voltages: (a) -20 V;

AC

CE

PT E

(b) -60 V; (c) -100 V.

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

MA

Fig. 5 Variation of hardness, elastic modulus, H3/E*2 and H/E for TiSiCN coatings deposited at

AC

CE

PT E

D

different bias voltages.

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

Fig. 6 High resolution TEM images of TiSiCN coating deposited at -100 V.

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

Fig. 7 Potentiodynamic polarization curves of TiSiCN coatings deposited at different bias voltages under different sliding condition in artificial seawater: (a) Ball motion with non-contact the coating; (b) Sliding.

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

Fig. 8 (a) Evolution of corrosion potential of TiSiCN coatings sliding against SiC ball in artificial seawater at open circuit potential; (b) Evolution of current density of TiSiCN coatings sliding against SiC ball in artificial seawater at cathodic protection condition.

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

MA

Fig. 9 Friction coefficients and volume loss rate of TiSiCN coatings deposited at different bias

AC

CE

PT E

D

voltages in the artificial seawater at OCP and CP condition.

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 10 High resolution TEM images of TiSiCN coating deposited at -100 V, red vertical lines indicate c-axis spacing of Ti.

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 11 SEM images of wear tracks of TiSiCN coatings deposited at different bias voltages at OCP or CP condition: (a) -20 V (OCP); (b) -20 V (CP); (c) -60 V (OCP); (d) -60 V (CP); (e) -100 V (OCP); (f) -100 V (CP);

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 12 Tribocorrosion model of TiSiCN coatings sliding against SiC ball in seawater.

ACCEPTED MANUSCRIPT

PT RI SC NU MA D PT E CE

  

Highlights A nanocystallites/amorphous and Ti3SiC2 MAX phase nanostructure was found in TiSiCN coating. The TiSiCN coating deposited at -100 V exhibits the best tribocorrosion properties. The synergistic effect of mechanical wear and electrochemical corrosion degraded coating. The applied protection potential has both beneficial and harmful effects on coating.

AC

