Effects of nitriding temperature on microstructures and vacuum tribological properties of plasma-nitrided titanium

Effects of nitriding temperature on microstructures and vacuum tribological properties of plasma-nitrided titanium

    Effects of nitriding temperature on microstructures and vacuum tribological properties of plasma-nitrided titanium Dingshun She, Wen ...

2MB Sizes 0 Downloads 34 Views

    Effects of nitriding temperature on microstructures and vacuum tribological properties of plasma-nitrided titanium Dingshun She, Wen Yue, Zhiqiang Fu, Chengbiao Wang, Xingkuan Yang, Jiajun Liu PII: DOI: Reference:

S0257-8972(15)00063-8 doi: 10.1016/j.surfcoat.2015.01.029 SCT 20044

To appear in:

Surface & Coatings Technology

Received date: Accepted date:

12 May 2014 12 January 2015

Please cite this article as: Dingshun She, Wen Yue, Zhiqiang Fu, Chengbiao Wang, Xingkuan Yang, Jiajun Liu, Effects of nitriding temperature on microstructures and vacuum tribological properties of plasma-nitrided titanium, Surface & Coatings Technology (2015), doi: 10.1016/j.surfcoat.2015.01.029

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 Type of contribution: Research paper, Full length article

RI P

T

Date of preparation: May 10, 2014

SC

Number of text pages: 23

MA NU

Number of tables & figures: 2 tables and 15 figures

Title: Effects of nitriding temperature on microstructures and vacuum tribological properties

ED

of plasma-nitrided titanium

PT

Authors: Dingshun She 1, 2, Wen Yue 1, 2*, Zhiqiang Fu 1, Chengbiao Wang 1, Xingkuan Yang 3,

CE

Jiajun Liu 4

1. School of Engineering and Technology, China University of Geosciences (Beijing), Beijing

AC

100083, PR China;

2. Key Laboratory on Deep Geo-drilling Technology of the Ministry of Land and Resources, China University of Geosciences (Beijing), Beijing 100083, PR China; 3. Metals & Chemistry Research Institute, China Academy of Railway Sciences, Beijing 100081, PR China; 4. Mechanical Engineering Department, Tsinghua University, Beijing 100084, PR China

*Corresponding author. Tel: +086 10 82320255, Fax: +086 10 82322624, E-mail: [email protected], [email protected]

ACCEPTED MANUSCRIPT Effects of nitriding temperature on microstructures and vacuum tribological properties of plasma-nitrided titanium

RI P

Jiajun Liu 4

T

Dingshun She 1, 2, Wen Yue 1, 2*, Zhiqiang Fu 1, Chengbiao Wang 1, Xingkuan Yang 3,

(1. School of Engineering and Technology, China University of Geosciences (Beijing),

SC

Beijing 100083, PR China;

MA NU

2. Key Laboratory on Deep Geo-drilling Technology of the Ministry of Land and Resources, China University of Geosciences (Beijing), Beijing 100083, PR China; 3. Metals & Chemistry Research Institute, China Academy of Railway Sciences, Beijing

ED

100081, PR China;

4. Mechanical Engineering Department, Tsinghua University, Beijing 100084, PR China)

PT

*Corresponding author. Tel.: +086 10 82320255; Fax: +086 10 82322624;

AC

CE

E-mail address: [email protected], [email protected]

ACCEPTED MANUSCRIPT ABSTRACT: Facing the outer space application of titanium and its alloys in drilling pipes and drilling robots, it is necessary to find an optimized surface engineering method to

T

improve their vacuum tribological properties. In order to find an optimized nitriding

RI P

temperature, a series of plasma nitriding experiments were carried out for commercial titanium (UNS R50400/Gr. 2/TA2/CP3) at various temperatures ranging from 650 to 950 ℃for

SC

8 h to investigate the effects of nitriding temperature on microstructures and mechanical

MA NU

properties. Vacuum tribological properties of the nitrided and untreated samples were examined using a ball-on-disc vacuum tribo-meter. The results show that the wear resistance of titanium tested under vacuum condition is effectively enhanced, due to the formation of the

ED

hard nitrided layer. The surface roughness, thickness and hardness of the nitrided layer increase with the increase of nitriding temperature. Nevertheless, the load bearing capacities

PT

of the samples nitrided at 900 and 950 ℃are much lower than those of the samples nitrided at

CE

800 and 850 ℃. Therefore, the wear volume keeps dropping at the nitriding temperature from 650 to 850 ℃, whereas, it converts to rise up when the nitriding temperature is above 850 ℃. In

AC

general, it is an optimized process to be nitrided at the temperature of 850 ℃for 8 h to improve the vacuum tribological properties of titanium.

Keywords: titanium; plasma nitriding; vacuum tribology; wear

ACCEPTED MANUSCRIPT 1. Introduction Due to the excellent mechanical properties such as excellent strength-to-weight,

T

exceptional corrosion resistance and inherent flexibility, Titanium (Ti) and its alloys have

RI P

been found to provide operational benefits in geological drilling systems, such as drilling pipes and drilling robots in space [1-3]. Nevertheless, Ti and Ti-based alloys exhibit low

SC

surface hardness, poor wear resistance and higher friction coefficient under vacuum condition

MA NU

[4-8]. Y. Liu et al. [8, 9] have demonstrated that plastic deformation and high flash temperature can melt and soften the metal, and then cause severe adhesion under vacuum condition. Accordingly, it is essential to find an optimized surface engineering method to

ED

improve the vacuum tribological properties of titanium and its alloys. Plasma nitriding (PN) has been proven to be one of the most effective methods improved

PT

the tribological properties of titanium and its alloy [10-13]. Plasma nitriding can produce a

CE

compound layer of TiN on the top of the matrix and Ti2N beneath, with a hardness of 3000 and 1500 HV, respectively [14, 15]. Generally, the composition, microstructure, thickness,

AC

and mechanical properties of the nitrided layer strongly depend on the nitriding temperature which usually can be chosen at a wide temperature ranging from 400 to 950 ℃[14-22]. The hardness and thickness of the nitrided layer increase with the increase of nitriding temperature in references [16-20]. A. Molinari et al. [15] have reported that plasma nitriding temperature is a basic and critical process parameter, and the tribological behaviours of the plasma-nitrided Ti6Al4V alloy are deeply influenced by the nitriding temperature. For space application, it is required to investigate the tribological behaviours of plasma-nitrided titanium under vacuum condition, and to optimize nitriding temperature to enhance the vacuum tribological properties of titanium. T. Spalvin [23, 24] investigated the

ACCEPTED MANUSCRIPT effects of the constitution of plasma-nitrided steel on friction coefficient under vacuum condition and revealed that the γ-Fe4N phase performed a lower friction coefficient than that

T

of ε-Fe3N, and the friction coefficient of the plasma nitrided steel displayed a significant

RI P

reduction. Previous studies [25, 26] have showed the predominant wear mechanism for plasma-nitrided surface under vacuum condition is abrasive, in complete contrast with

SC

oxidation wear and delamination under ambient condition. Unfortunately, few researches on

MA NU

the tribological properties of plasma-nitrided titanium under vacuum condition have been reported.

In this study, UNS R50400 titanium was plasma-nitrided under the temperatures of 650,

ED

700, 750, 800, 850, 900 and 950 ℃for 8 h. The effects of nitriding temperature on the surface morphology, compound layer thickness, constitution, hardness, load bearing capacity and

PT

vacuum tribological property have been investigated. It aims to optimize the nitriding

CE

temperature to improve the wear resistant performance of titanium under vacuum conditions.

AC

2. Experimental details 2.1 Materials

Commercial titanium (UNS R50400/Gr. 2/TA2/CP3) sheets with a thickness of 5 mm and a hardness of 140 HV were used. Its chemical composition was shown in TableⅠ. Samples were cut into discs with a diameter of 65 mm by a CNC wire-cut electric discharge machine. The samples were then ground and polished to a mirror surface. 2.2 Plasma nitriding treatment The discs were placed into a LDM 1-100 plasma nitriding furnace after cleaning with acetone in an ultrasonic cleaner. The discs were connected to the cathode, and the furnace

ACCEPTED MANUSCRIPT wall acted as the anode. The vacuum chamber was evacuated to ultimate vacuum (below 10 Pa) and the leakage rate of the equipment was less than or equal to 2 Pa per 15 min prior to

T

the nitriding treatment. Before the nitriding process, the plasma chamber was back-filled by

RI P

argon and hydrogen (3:1 ratio) for sputtering at 900 Pa gas pressure. During the process of sputtering, the maximum temperature was 200 ℃. After 45 min of sputtering, the ammonia

SC

gas was introduced for plasma nitriding. The nitriding was carried out at 700 Pa in an NH3

MA NU

containing atmosphere. Nitriding was performed at 650, 700, 750, 800, 850, 900 and 950 ℃ , and those nitrided discs were marked as 650PN, 700PN, 750PN, 800PN, 850PN, 900PN and 950PN, respectively. During the process of plasma nitriding, the glow discharge was operated

ED

with a potential voltage 700~850 V to obtain the prescribed nitriding temperature. Finally, the discs samples were cooled to room temperature in NH3 atmosphere with a pressure of 600 Pa.

PT

2.3 Characterizations

CE

Surface roughness and 3D topographies were characterized by 3D profiler (Nano-Map-D). For the roughness measurements, the tests were repeated ten times and the

AC

average value was calculated. JSM-7001 F model Scanning electron microscope (SEM) was adopted to investigate surface morphologies of the discs. Axio Imager M2m model optical Microscope (OM) was employed to observe the microstructure of cross-section of the untreated and nitrided discs. The cross-sectional samples mounted in the dental base acrylic resin was polished, and etched in Kroll’s reagent (2% HF and 4% HNO3). The constitution of the nitrided layers were identified by a D/max X-ray diffract-meter using a Cu-Kα radiation source (wavelength of 1.5406 Å), 2θ range of 30-80° and an increment of 0.04°/step with a time of 1.5 s per step.

ACCEPTED MANUSCRIPT Micro-hardness was used to characterize the hardness of the nitrided layer as a function of depth. A MH-6 model Vickers micro-hardness tester was employed to measure the surface

T

hardness and hardness profile along the depth at a load of 50 gf. According to the ASTM

RI P

standard E384-11e1, the indentation was corrected before the hardness test. Each measurement was repeated for five times and the average value was chosen. The thickness of

SC

nitrided layer was defined as the depth where the hardness was 10% HV above the core

MA NU

hardness [27, 28]. 2.4 Scratch tests

Scratch test is generally accepted as one of the simple means in assessing the wear and

ED

crack behavior of the nitride layer [29]. During the process of the scratch test, a diamond tip is driven over a nitride layer surface to produce a scratch, and the load on the diamond tip is

PT

increased linearly to induce a shear force that is proportional to the applied load and

CE

transmitted through the bulk of the nitrided samples [29, 30]. As the plasma nitrided samples possess a gradient nitrided layer, the mechanical properties of the nitrided samples are

AC

different along the scratching depth [10-22, 29]. Therefore, there is a discontinuity in the shear stress at the interface which, when sufficiently high, induces adhesive failure at a critical load. Generally, for hard nitrided layer, microcracks appear during scratching before the final adhesion failure [30, 31]. The minimum load at which the first crack occurs is termed as the critical load (Lc). Some researchers directly used this critical load to indicate cracking resistance of the hard coating (nitrided layer is also one of hard coatings) because the higher Lc, the more difficult it is to initiate a crack in the nitrided layer [29-35]. Therefore, what the lower critical load represents is a load bearing capacity: the minimum critical load of crack initiation [29, 30].

ACCEPTED MANUSCRIPT In order to characterize the load bearing capacity, a Rockwell-C diamond tip was adapted to produce a scratch on the surface of nitrided samples. A MFT-4000 scratch tester was used

T

to carry out the scratch tests. The machine was equipped with acoustic emission and frictional

RI P

force measurement. The loading rate was 100 N/min. The nominal maximum load was 50 N. The tests were carried out under ambient laboratory temperature and humidity conditions

MA NU

to obtain mean and standard deviation values.

SC

which were 22 ℃and 40-50% relative humidity. Five scratch tests per coating were performed

2.5 Tribo-tests

Tribo-tests were performed on a MSTS-1 model ball-on-disc vacuum tribo-meter. The

ED

schematic diagram of MSTS-1 vacuum tribo-meter was shown in Ref. [36]. AISI 52100 steel ball with a diameter of 9.525 mm, hardness of 770 HV and surface roughness of Ra 32 nm

PT

was chosen as the counter face materials of the untreated or nitrided discs. During the process

CE

of tribo-tests, the upper ball was fixed, and the lower disc was rotated with a rotating speed of 100 rpm and a wear scar diameter of 26 mm (the sliding velocity is about 0.136 m/s).

AC

Tribo-tests were carried out under a high vacuum (6.67×10-4 Pa) condition with a temperature of 25 ℃. The test duration was 1200 s, and the applied load was 3 N (corresponding to the mean Hertzian contact stress of 0.53 GPa). Wear volume of the discs was measured by a 3D profiler (Nano-Map-D). To reveal the wear mechanism of the untreated and nitrided Ti, the morphologies of worn surface were investigated by JSM-7001 F model Scanning electron microscope (SEM) equipped with Oxford EDX-450 model energy dispersion spectrum (EDS).

3. Results

ACCEPTED MANUSCRIPT 3.1 Morphologies and microstructures The SEM surface micrographs of the discs are shown in Fig. 1. It can be seen that the

T

surface of untreated discs is quite smooth. Nevertheless, some micro-pits and granular

RI P

micro-particles can be found on the surface of 750PN, 850PN and 950PN samples. Additionally, it can be found that nitride particles are easy to aggregate larger-size particles at

SC

a high nitriding temperature. The growing size of the nitrided micro-particles is accompanied

MA NU

with the increase of surface roughness (Ra) of the nitrided discs. As shown in 3D profiler images of the untreated and nitrided samples (Fig. 2), the 900 PN and 950PN samples exhibit much rougher surfaces with many sharp peaks and valleys. Average surface roughness of the

ED

untreated and nitrided samples is shown in Fig. 3. Obviously, the surface roughness of 900 PN and 950PN samples is much higher than those of the other samples. Average surface

PT

roughness of the nitrided discs keeps rising with the increase of nitriding temperature. S.R.

CE

Hosseini et al. [21] have reported that variation of surface roughness versus process temperature is similar to an exponential curve.

AC

Fig.4 shows optical microscopy images of the cross-section of untreated and nitrided discs. It can be seen that there is a uniform and continuous white compound layer on the surface of the 750PN, 850PN and 950PN samples, and lamellar structure can be found in the substrate of the 900PN and 950PN samples. The formation of lamellar structure is attributed to the nitriding temperature is higher than the transition temperature (882 ℃) of α-Ti to β-Ti phase [22]. In fact, it cannot be neglected that the formation of the lamellar structure in bulk is deleterious to the mechanical properties of Ti. In addition, the thickness of the white compound layer on the nitrided samples is listed in Fig. 5. Obviously, there is no white compound layer formed on the surface of the 650PN

ACCEPTED MANUSCRIPT samples. As the nitriding temperature is above 700 ℃, the compound layer starts to grow and its thickness gradually increases with the increase of nitriding temperature.

T

3.2 XRD analysis

RI P

XRD patterns of the untreated and nitrided discs are shown in Fig. 6. It can be seen that the peaks of α-Ti keep dropping, nevertheless, the peaks of ε-Ti2N and δ-TiN keep rising with

SC

the increase of nitriding temperature. For the 650PN and 700PN samples, a new phase ε-Ti2N

MA NU

with a tetragonal crystal can be indentified compared with the untreated samples. Furthermore, the conspicuous peaks of δ-TiN phase with NaCl-type crystal structure can be found on the 750PN samples. Accordingly, the results of XRD patterns demonstrate that a nitride layer was formed on the titanium discs and the increasing nitriding temperature results in the increasing

ED

growth of both ε-Ti2N and δ-TiN phase.

PT

3.3 Micro-hardness

CE

Micro-hardness of the untreated and nitrided samples is listed in Fig. 7. Obviously, the hardness of the nitrided samples is significantly enhanced by plasma nitriding. The higher

AC

nitriding temperature, the harder surface layer formed. The hardness of 950PN samples is approaching to 1560 HV, and that of the untreated samples is only about 140 HV. Fig. 8 shows the micro-hardness variation versus the cross-sectional depth of the samples. For the nitrided samples, the micro-hardness keeps dropping form the top surface to the substrate, and there is no sharp and abrupt change of hardness between surface layer and the substrate. The higher nitriding temperature, the harder and thicker nitrided layer is formed, which is in according with the thickness of the compound layer shown in Fig. 4 and 5. In addition, the core hardness of the 650PN, 700PN, 750PN, 800PN and 850PN is the same as the untreated samples, whereas, the core hardness of the 900PN and 950PN samples increases up to about 200 HV.

ACCEPTED MANUSCRIPT The formation of the hard β phase lamellar structure (as shown in Fig. 4 (d)) is one of the reasons for higher hardness of the bulk materials. Many literatures [16, 22, 23] have showed

T

that the increase of surface hardness and the thickness of the nitride layer is attributed to the

RI P

formation of hard phases (ε-Ti2N and δ-TiN), and an increased diffusion of nitrogen and a

processed temperature.

MA NU

3.4 Load bearing (cracking resistance) capacity

SC

rapid reaction rate promoted the formation of hard phases (ε-Ti2N and δ-TiN) at a high

Fig. 9 shows morphologies of the typical scratches on the untreated and nitrided samples. Visibly, the nitrided samples treated at the lower nitriding temperature (650, 700 and 750 ℃) shows a minimum critical load of crack initiation (Lc) at 10~20 N. Such a low minimum

ED

critical load of crack initiation results from a thin hard nitrided layer composed of α-Ti and

PT

some brittle ceramic phases such as ε-Ti2N and δ-TiN phase on a soft substrate with a poor

CE

mechanical support [29-31]. A suitable gradient of hardness (as shown in Fig. 8) can provide a better mechanical support to the compound layer, as a comparative thick nitrided layer forms

AC

on the 800PN and 850PN samples. As a consequence, the 800PN and 850PN samples exhibit a better load bearing capacity with a higher Lc of 30~40 N, comparing with the 650PN, 700PN and 750PN samples. Nevertheless, for the 900PN and 950PN samples, the spalling on the edge of scratches is severe, thus the Lc of is only about 6~8 N. The severe spalling and low Lc of the 900PN and 950PN samples can be attributed to the following two reasons. The one is that almost no α-Ti could be indentified on the surface of 900PN and 950PN samples, as shown in Fig. 6. The compound layer is consisted of ε-Ti2N and δ-TiN phases (brittle ceramic phase), and the other one is that the formation of the β phase lamellar structure, as shown in Fig.4d, is easy to cause a crack. Accordingly, despite a higher hardness and a thicker

ACCEPTED MANUSCRIPT nitrided layer are formed on the 900PN and 950PN samples, the load bearing capacity and cracking resistance of the 900PN and 950PN samples are much lower than those of the

T

800PN and 850PN samples.

RI P

3.5 Friction coefficient

Friction coefficient of the untreated and nitrided samples is shown in Fig. 10. Friction

SC

coefficient of the nitrided samples is deeply influenced by nitriding temperature. The average

MA NU

friction coefficients of the 650PN, 700PN, 750PN, 900PN and 950PN samples are around 0.65. Comparatively, the 800PN and 850PN samples exhibit the lowest friction coefficient about 0.54, and the untreated samples perform the highest average friction coefficient about

ED

0.73. The typical friction coefficient curves versus tribo-test time are shown in inner figures of Fig. 10. During the process of tribo-tests, the friction coefficient is fluctuant under vacuum

PT

condition. The 850PN samples perform the mildest friction coefficient fluctuation ranging

CE

from 0.25 to 0.95, and the untreated samples exhibits the most severe fluctuation ranging from 0.15 to 1.35.

AC

3.6 Wear behaviours

3D profiler images of the worn surface the untreated and nitrided samples are shown in Fig.11. Evidently, the untreated and 650PN samples suffer serious wear, and there are some plastic flows on the worn surfaces. Comparing with the 950PN samples, the 850PN samples exhibit a milder wear. Wear volumes of the untreated and nitrided discs are shown in Fig. 12. It can be seen that the wear resistance of the titanium discs tested under vacuum condition is effectively improved by plasma nitriding. According to Archard theory, the wear volume is inversely proportion to the hardness. Therefore, the wear volume of the nitrided sample decreases with

ACCEPTED MANUSCRIPT the increase of nitriding temperature when the nitriding temperature is lower than 850 ℃. Nonetheless, the wear volume of the nitride samples tends to increase when the temperature

T

exceeds 850 ℃. Accordingly, the 850PN samples present the lowest wear volume.

RI P

As listed in Fig. 13, the variation of wear scar diameter of counter balls is in accordance with the variation of wear volume. Wear scar diameter is initially decreased, and then

SC

increased with the increase of nitriding temperature. Wear scar diameter of the ball against the

MA NU

untreated discs is about 1.38 mm, which is higher than those of other counter balls. Counter ball against the 850PN samples shows the smallest wear scar diameter (about 0.42 mm).

ED

4. Discussion

The characteristics of the wear surface reveal quite different wear mechanisms between

PT

the untreated samples and the samples nitrided at various temperatures. As shown in Fig. 14a,

CE

typical characteristics of tongue-shaped wedges are clearly observed on the untreated samples, which indicates that severe adhesion and plastic deformation occurred on the untreated and

AC

650PN samples. Moreover, some scale-like debris and plowing along the sliding direction can be found on the worn surface of 700PN and 750PN samples. Wear damage of the samples nitrided at low temperature is a comprehensive wear model of abrasive wear and plastic deformation, adhesive wear. For the 800PN and 850PN samples, no plastic deformation is observed on the worn surface, whereas, some plowing along sliding direction can be found on the worn surface. Accordingly, the dominant wear mechanism is abrasive wear. The 900PN and 950PN samples show a feature of abrasive wear (plowing along direction) and spalling as shown in Fig 14d. A further EDS analysis has been carried out to discuss the wear mechanisms. As shown

ACCEPTED MANUSCRIPT in Table Ⅱ, Fe and Cr can be identified on the worn surface of the 900PN and 950PN samples, which indicates that material transfers from the counter balls to the discs. The materials

T

transfer is the main reason for the big wear scar on counter balls against 900PN and 950PN

RI P

samples. Almost no N was detected on the worn surface of the samples nitrided at 650 ℃, which also demonstrates that the nitrided layer has already been worn out.

SC

There is no air convection for cooling under vacuum condition, meanwhile it is different

MA NU

to regenerate the surface films (absorbed film and oxide film) with friction reduction property, and thus the fresh surface of ball and disc contact each other directly [5-9, 37,38]. In addition, the surface can be plowed and cut by hard particles easily, leaving behind deep grooves as

ED

shown in Fig. 11. Plastic deformation and high flash temperature can melt and soften the metal, which is able to cause severe adhesion [29, 30]. The above effects result in the

PT

breakage of friction stationary and the aggravation of adhesion. Under the shearing action of

CE

friction, the adhesion junctions with high strength are easy to be cut up and slide with each other. Finally, sliding wear transforms to the alternating process of “stick-slip behaviour”.

AC

Accordingly, the untreated samples show the most severe friction coefficient fluctuation and severe adhesion wear. For the 650PN, 700PN and 750 PN samples, the hardness has been improved. Nevertheless, the thin hard nitrided layer without an enough thick diffusion layer cannot provide a better mechanical support to the applied normal load. The load bearing capacity is not higher enough, and then the compound layers on the surface of the discs tend to be fractured and produced wear debris. These wear debris can cause slight scuffing and abrasive wear. Meanwhile, these wear debris also can transform into scale-like debris adhered on the worn surface after plastic deformation as shown in Fig. 11b. Furthermore, the wear debris is

ACCEPTED MANUSCRIPT able to prevent fresh surface of the counter balls and discs contact each other directly. As a consequence, adhesive wear tends to be milder, and the “stick-slip behaviour” is restrained

T

during the process of the tribo-tests. Therefore, the milder fluctuation of friction coefficient

RI P

can be observed on the 650PN, 700PN and 750 PN samples, and relatively lower friction coefficients comparing with the untreated sample. In addition, the hard nitrided layer can

SC

provide excellent abrasive wear resistance [23-26]. Therefore, the wear volumes of the 650PN,

MA NU

700PN and 750PN samples are lower than that of the untreated samples. The surface hardness of the 800PN and 850PN samples has been effectively enhanced by plasma nitriding. Additionally, a thicker nitrided layer with a hardness gradient can provide

ED

better mechanical support to the applied normal load. Therefore, the load bearing capacities of the 800PN and 850PN samples is perfect. The wear debris spalled from the discs acting as

PT

abrasive particles is hard to deform. As a result, no “adhesion-cut up-adhesion” and plastic

CE

deformation characteristics can be observed on the worn surface of the 800PN and 850PN samples, such as tongue-shaped wedges and scale-like debris. In another words, the nitride

AC

layer on the 800PN and 850PN samples acts as a barrier to protect titanium surfaces from adhesion wear and plastic deformation under vacuum condition. The wear mechanism of the 800PN and 850PN samples is abrasive wear. M.M. Yazdanian et al. [39] have reported the similar results that thermally oxidized layer on the titanium alloy can effectively protect titanium surfaces from adhesion wear and plastic deformation under vacuum condition. Accordingly, the fluctuation of friction coefficient is mildest, and the average friction coefficient of the 800PN and 850PN samples also decreased due to the decrease of adhesion. The high surface hardness requires the diffusion layer to provide better mechanical support for the compound layer. Otherwise, any cracking on the compound layer resulting

ACCEPTED MANUSCRIPT from substrate deformation is likely to lead to higher localized spalling which results in severe wear damage [13]. For the 900PN and 950PN samples, the hardness is enhanced to about

T

1500~1600 HV. As shown in Fig.9, the load capacities of the 900PN and 950PN samples were

RI P

poor. Meanwhile, obvious micro-cracks and spalling can be found on the scratches of the 900PN and 950PN samples. Therefore, a mixed failure mode of obvious micro-cracks and

SC

spalling is harmful to keep the wear resistance of the compound layer. Actually, as shown in

MA NU

Fig.15, the results of wear tests also demonstrate that micro-cracks and spalling can be observed on the worn surface of the 900PN and 950PN samples. Those wear debris spalled from the discs acting as hard abrasive particles plows and shears the worn surface of the

ED

900PN and 950PN samples and their counter ball. Consequently, the 900PN and 950PN samples and its counter balls suffer severe abrasive wear (as shown in Fig. 13). Comparing

PT

with the 850 PN samples, there are more and harder abrasive particles on the worn surface of

CE

the 900PN and 950PN samples during the wear tests. Therefore, the abrasive wear on the 900PN and 950PN samples is more serious than that on the 850 samples. D. Nolan et al. [13]

AC

have demonstrated that the more gradual transition of the hardness and elastic modulus across the nitrided layer/substrated interface, the more excellent load bearing capacity and sliding wear resistance. Additionally, M. Rahman et al. [22] has demonstrated that the rougher surface formed at a high nitriding temperature leads to more severe wear damage and high average friction coefficient. As a consequence, the wear volume and average friction coefficient of the 800PN and 850PN samples are lower than that of the 900PN and 950PN samples. Furthermore, previous studies [40-42] have shown that the size and shape of wear debris play basic role on the evolution of friction coefficient in dry sliding conditions. As a result, the 900PN and 950PN samples perform a wilder fluctuation of friction coefficient.

ACCEPTED MANUSCRIPT 5. Conclusions The following conclusions can be drawn from this work:

T

(1) Plasma nitriding can effectively enhance the wear resistance, and reduce the friction

RI P

coefficient of UNS R50400 titanium (Ti) under vacuum condition. In case of the nitriding temperature ranges from 650 to 950 ℃for 8 h, UNS R50400 Ti nitrided at the temperature of

SC

850 ℃exhibits the most excellent wear resistance and the lowest friction coefficient under

MA NU

vacuum condition.

(2) As nitriding temperature rises from 650 to 850 ℃, the wear resistance of nitrided UNS R50400 titanium shows a upward trend, which results from the increase of the thickness, hardness and load bearing (cracking resistance) capacity of the nitrided layer.

ED

(3) For the case of the UNS R50400 titanium nitrided at 900 and 950 ℃, a brittle ceramic

PT

top layer and a lamella structure in the substrate are formed, which results in the poor load

CE

bearing (cracking resistance) capacity. The poor load bearing capacity accompanied with the

AC

sharp rise of roughness leads to a higher wear rate.

Acknowledgments The authors are grateful for the financial support by the National Natural Science Foundation of China (51375466), the Beijing Higher Education Young Elite Teacher Project (YETP0646), the Beijing Natural Science Foundation (3132023), the Fundamental Research Funds for the Central Universities (2652013080) and the Tribology Science Fund of State Key Laboratory of Tribology (SKLTKF13B10). The authors would like to thank Prof. Haidou Wang and Dr. Guozheng Ma from National Key Lab for Remanufacturing, Academy of Armored Forces Engineering, for their help with the use of the vacuum tribotester.

ACCEPTED MANUSCRIPT References [1] K. Zacny, Y. Bar-Cohen, K. Davis, P. Coste, G. Paulsen, S. Sherrit, J. George, B.

T

Derkowski, S. Gorevan, D. Boucher, J. Guerrero, T. Kubota, B.J. Thomson, S. Stanley, P.

RI P

Thomas, N. Lan, C. McKay, T.C. Onstot, C. Stoker, B. Glass, S. Wakabayashi, L. Whyte, G. Visentin, E. Re, L. Richter, M. Badescu, X. Bao, R. Fincher, T. Hoshino, P. Magnani, C.

SC

Menon, Extraterrestrial drilling and excavation, Y. Bar-Cohen, K. Zancy (Eds.), Drilling

MA NU

in Extreme Environments, Wiley-VCH Verlag GmbH , New York, 2009, pp. 347–557 [2] P. Younse, A. Stroupe, T. Huntsberger, M. Garrett, J.L. Eigenbrode, L.G. Benning, M. Fogel, A. Steele, Sample acquisition and caching using detachable scoops for mars

ED

sample return, Aerospace conference, 10.1109/AERO.2009.4839312, 2009. [3] M. Anttila, Concept evaluation of mars drilling and sampling instrument, Ph.D. Thesis,

PT

Helsinki University of Technology.

CE

[4] K. Zacny, G. Cooper, Considerations, constraints and strategies for drilling on Mars, Planet. Space Sci. 54 (2006) 345-356.

AC

[5] K. Miyoshi, Aerospace mechanisms and tribology technology-case study, Tribol. Int. 32(11) (1999) 605-616. [6] K. Miyoshi, Considerations in vacuum tribology (adhesion, friction, wear, and solid lubrication in vacuum), Tribol. Int. 32(11) (1999) 673-685. [7] I.L. Lebedeva, G.N. Presnyakova, Adhesion wear mechanisms under dry friction of titanium alloys in vacuum, Wear. 148(2) (1991) 203-210. [8] Y. Liu, D.Z. Yang, S.Y. He, W.L. Wu, Microstructure developed in the surface layer of Ti6Al4V alloy after sliding wear in vacuum, Mater. Charact. 50 (2003) 275-279. [9] Y. Liu, D.Z. Yang, S.Y. He, W.L. Wu, Study on dry sliding wear of TC4 alloy in vacuum,

ACCEPTED MANUSCRIPT Rare Metal. Mat. Eng. 34 (1) (2005) 128-131. [10] D.G. Bansal, O.L. Eryilmaz, P.J. Blau, Surface engineering to improve the durability and

T

lubricity of Ti–6Al–4V alloy, Wear 271(9-10) (2011) 2006-2015.

RI P

[11] G. Cassar, J.A.B.S. Wilson, S. Banfield, J. Housden, A. Matthews, A. Leyland, A study of the reciprocating-sliding wear performance of plasma surface treated titanium alloy, Wear

SC

269(1-2) (2010) 60.

MA NU

[12] A.F. Yetim, F. Yildiz, Y. Vangolu, A. Alsaran, A. Çelik, Several plasma diffusion processes for improving wear properties of Ti6Al4V alloy, Wear 267(12) (2009) 2179-2185.

ED

[13] D. Nolan, S.W. Huang, V. Leskovsek, S. Braun, Sliding wear of titanium nitride thin films deposited on Ti–6Al–4V alloy by PVD and plasma nitriding processes, Surf. Coat.

PT

Technol. 200(20-21) (2006) 5698-5705.

CE

[14] S.L. Ma, K.W. Xu, W.Q. Jie, Wear behavior of the surface of Ti–6Al–4V alloy modified by treating with a pulsed d.c. plasma-duplex process, Surf. Coat. Technol. 185(2-3) (2004)

AC

205-209.

[15] A. Molinari, G. Straffelini, B. Tesi, T. Bacci, G. Pradelli, Effects of load and sliding speed on the tribological behaviour of Ti6Al4V plasma nitrided different temperatures, Wear 203-204 (1997) 447-454. [16] F.Yilidiz, A.F. Yetim, A. Alsaran, A. Çelik, Plasma nitriding behavior of Ti6Al4V orthopedic alloy, Surf. Coat. Technol. 202(11) (2008) 2471-2476. [17] M. Rahman, M.S.J. Hashmi, Saddle field fast atom beam source: A new low pressure plasma nitriding method for a alloy Ti–6Al–4V, Thin Solid Film 515(1) (2006) 129-134. [18] S.Taktak, H. Akbulut, Diffusion kinetics of explosively treated and plasma nitrided

ACCEPTED MANUSCRIPT Ti–6Al–4V alloy, Vacuum 75(3) (2004) 247-259. [19] S.Taktak, H. Akbulut, Dry wear and friction behaviour of plasma nitrided Ti–6AL–4 V

T

alloy after explosive shock treatment, Tribol. Int. 40(3) (2007) 423-432.

RI P

[20] D.S. She, W. Yue, Z.Q. Fu, Y.H. Gu, C.B. Wang, J.J. Liu, The effect of nitriding temperature on hardness and microstructure of die steel pre-treated by ultrasonic cold

SC

forging technology, Mater. Des. 49 (2013) 392-399.

MA NU

[21] S.R. Hosseini, A. Ahmadi, Evaluation of the effects of plasma nitriding temperature and time on the characterisation of Ti 6Al 4V alloy, Vacuum 87 (2013) 30-39. [22] M. Rahman, I. Reid, P. Duggan, D.P. Dowling, G. Hughes, M.S.J. Hashmi, Structural and

ED

tribological properties of the plasma nitrided Ti-alloy biomaterials: Influence of the treatment temperature, Surf. Coat. Technol. 201(9-11) (2007) 4865-4872.

PT

[23] T. Spalvin, Tribological and microstructural characteristics of ion-nitrided steels, Thin

CE

solid films 108(2) (1983) 157-163. [24] T. Spalvin, Frictional and structural characterization of ion‐ nitrided low and high

AC

chromium steels, J. Vac. Sci. Technol. A 3(6) (1985) 2329-2333. [25] J.Q. Yang, Y. Liu, Z.Y. Ye, D.Z. Yang, S.Y. He, Microstructure and tribological characteristics of nitrided layer on 2Cr13 steel in air and vacuum, Surf. Coat. Technol. 204(5) (2009) 705-712. [26] J.Q. Yang, Y. Liu, Z.Y. Ye, D.Z. Yang, S.Y. He, A study of friction and wear behavior of nitrided layer on 2cr13 steel in vacuum, Tribol. Lett. 40(3) (2010) 285-294. [27] S.D. Oliveira, A.P. Tschiptschin, C.E. Pinedo, Simultaneous plasma nitriding and ageing treatments of precipitation hardenable plastic mould steel, Mater. Des. 28(5) (2007) 1714-1718.

ACCEPTED MANUSCRIPT [28] S.Y. Sirin, E. Kaluc, Structural surface characterization of ion nitrided AISI 4340 steel, Mater. Des. 36 (2012) 741-747.

T

[29] H. Paschke, M. Weber, P. Kaestner, G. Braeuer, Influence of different plasma nitriding

RI P

treatments on the wear and crack behavior of forging tools evaluated by Rockwell indentation and scratch tests, Surf. Coat. Technol. 205 (20010) 1465-1469.

SC

[30] S. Zhang, D. Sun, Y.Q. Fu, H.J. Du, Toughness measurement of thin films: a critical

MA NU

review, Surf. Coat. Technol. 198 (2005) 74-84.

[31] S. Zhang, X.M. Zhang, Toughness evaluation of hard coatings and thinfilms, Thin Solid Films, 520 (2012) 2375-2389.

ED

[32] A.A. Voevodin, C. Rebholz, J.M. Schneider, P. Stevenson, A. Matthews, Wear resistant composite coatings deposited by electron enhanced closed field unbalanced magnetron

PT

sputtering, Surf. Coat. Technol. 73 (1995) 185-197.

CE

[33] E. Harrry, A. Rouzaud, P. Juliet, Y. Pauleau, M. Ignat, Failure and adhesion characterization of tungsten–carbon single layers, multilayered and graded coatings, Surf.

AC

Coat. Technol. 116–119 (1999) 172-175. [34] A.A. Voevodin, J.S. Zabinski, Supertough wear-resistant coatings with ‘chameleon’ surface adaptation, Thin Solid Films 370 (2000) 223-231. [35] A.A. Voevodin, J.S. Zabinski, Load-adaptive crystalline–amorphous nanocomposites, J. Mater. Sci. 33 (1998) 319-327. [36] G.Z. Ma, B.S. Xu, H.D. Wang, S.Y. Chen, Z.G. Xing, Excellent vacuum tribological properties of Pb/PbS film deposited by Rf magnetron sputtering and ion sulfurizing, Appl. Mater. Interfaces 6(1) (2014) 532-538. [37] S.Z. Wen, Tribology Principle, Tsinghua university press, Beijing, 2003. (in Chinese)

ACCEPTED MANUSCRIPT [38] H.D. Wang, G.Z. Ma, B.S. Xu, H.J. Shi, D.X. Yang, Microstructure and vacuum tribological properties of 1Cr18Ni9Ti steel with combined surface treatments, Surf. Coat.

T

Technol. 205(11) (2011) 3546-3552.

RI P

[39] M.M. Yazdanian, A. Edrisy, A.T. Alpas, Vacuum sliding behaviour of thermally oxidized Ti–6Al–4V alloy, Surf. Coat. Technol. 202(4-7) (2007) 1182-1188.

SC

[40] A.I. Dmitriev, W. Österle, H. Kloß, G. Orts-Gil, A study of third body behaviour under

MA NU

dry sliding conditions. Comparison of nanoscale modelling with experiment, Est. J. Eng. 18(3) (2012) 270-278.

[41] A. Zmitrowicz, Wear debris: a review of properties and constitutive models, J. Thero.

ED

Appl. Mech-pol.43(1) (2005) 3-35.

[42] J. Denape, Paper XI (iii) Wear Debris Action in Sliding Friction of Ceramics, Tribol.

AC

CE

PT

Series 21 (1992) 453-462.

ACCEPTED MANUSCRIPT Figure captions: Fig. 1 Typical SEM micrographs of the (a) Untreated, (b) 700PN, (c) 750PN and (d) 950PN

T

samples.

RI P

Fig. 2 3D profiler images of the (a) untreated (b) 700PN (c) 800PN and (d) 900PN samples. Fig. 3 Surface roughness of the untreated and nitrided samples.

SC

Fig. 4 Optical microscopy images of the cross-section of the (a) Untreated, (b) 750PN, (c)

MA NU

850PN and (d) 950PN samples.

Fig. 5 Thickness of the white compound layer forming on the nitrided samples. Fig. 6 XRD patterns of the nitrided samples compared with the untreated samples.

ED

Fig. 7 Surface hardness of the untreated and nitrided samples. Fig. 8 Micro-hardness variation versus cross-sectional depth.

PT

Fig. 9 Scrathes evaluated in scratch test showing the influence of the nitriding temperature on

CE

the load bearing capacity of the nitrided layer. Fig. 10 Friction coefficients of the untreated and nitrided samples.

AC

Fig. 11 3D profiler images of the untreated and nitrided samples. Fig.12 Wear volumes of the untreated and nitride samples. Fig. 13 Wear scar diameter of the counter balls. Fig. 14 SEM morphologies of worn surfaces of the (a) Untreated, (b) 750PN, (c) 850PN and (d) 950PN samples. Fig. 15 SEM morphology of worn surface of the 950PN sample.

Table captions:

RI P

T

ACCEPTED MANUSCRIPT

Table ⅠNominal composition of commercial titanium (wt.%).

AC

CE

PT

ED

MA NU

SC

Table ⅡEDS analysis results on the worn surfaces.

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 1 Typical SEM micrographs of the (a) Untreated, (b) 700PN, (c) 750PN and (d) 950PN samples.

Fig. 2 3D profiler images of the (a) untreated (b) 700PN (c) 800PN and (d) 900PN samples.

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

Fig. 3 Surface roughness of the untreated and nitrided samples.

Fig. 4 Optical microscopy images of the cross-section of the (a) Un-treated, (b) 750PN, (c) 850PN and (d) 950PN samples.

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 5 Thickness of the white compound layer forming on the nitrided samples.

Fig. 6 XRD patterns of the nitrided samples comparing with the untreated samples.

Fig. 7 Surface hardness of the untreated and nitrided samples.

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

Fig. 8 Microhardness variation versus cross-sectional depth.

Fig. 9 Scrathes evaluated in scratch test showing the influence of the nitriding temperature on the load bearing capacity of the nitrided layer.

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 10 Friction coefficients of the un-treated and nitrided samples.

Fig. 11 3D profiler images of the (a) 650PN, (b) 750PN, (c) 850PN and (d) 950PN samples.

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

Fig. 12 Wear volumes of the untreated and nitride samples.

Fig. 13 Wear scar diameter of the counter balls.

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

Fig.14 SEM morphologies of worn surfaces of the (a) 650PN, (b) 750PN, (c) 850PN and (d) 950PN samples.

Fig.15 SEM morphology of worn surface of the 950PN sample.

ACCEPTED MANUSCRIPT Table 1 Nominal composition of commercial titanium (wt.%). Ti

Fe

O

C

H

N

Si

Content

Bal.

≤0.30

≤0.35

≤0.08

≤0.015

≤0.03

≤0.15

AC

CE

PT

ED

MA NU

SC

RI P

T

Element

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

Table Ⅱ EDS analysis results on the worn surfaces. Spectrum Samples Element content (at.%) number Ti N Fe Cr 1 un-treated 99.96 ---0.04 ---2 650PN 98.33 1.65 0.02 ---3 700PN 86.73 13.61 0.18 ---4 750PN 82.33 17.49 0.16 ---5 800PN 78.78 19.99 1.23 ---6 850PN 71.52 26.21 2.26 ---7 900PN 63.37 28.76 7.28 0.59 8 950PN 57.52 30.27 11.95 1.26