Approach to controllable tribological properties of sintered polycrystalline diamond compact through annealing treatment

Carbon 116 (2017) 103e112

Contents lists available at ScienceDirect

Carbon journal homepage: www.elsevier.com/locate/carbon

Approach to controllable tribological properties of sintered polycrystalline diamond compact through annealing treatment Jiansheng Li a, b, Wen Yue b, c, *, Wenbo Qin b, Chengbiao Wang b, c a

Nano Structural Materials Center, School of Materials Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, PR China School of Engineering and Technology, China University of Geosciences (Beijing), Beijing 100083, PR China c Key Laboratory on Deep Geo-drilling Technology of the Ministry of Land and Resources, China University of Geosciences (Beijing), Beijing 100083, PR China b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 19 August 2016 Received in revised form 26 January 2017 Accepted 27 January 2017 Available online 31 January 2017

Tribological properties are the significant engineering factors to evaluate the quality of sintered polycrystalline diamond compact (PDC). An effective approach to control the tribological properties of PDC is extremely desired for engineers. The current paper has systematically investigated the tribological properties of PDC in relation to annealing temperatures. The friction coefficients and the wear rates of PDC and Si3N4 balls were evaluated by the morphologies and chemical conversions on tribological surfaces. Carbonaceous transfer films, induced by tiny diamond grains (TD grains, sizes of 0e5 mm), can effectively reduce the friction coefficients of the annealed PDC. The wear of the annealed PDC is caused by the extraction of medium size diamond grains (MD grains, sizes of 5e15 mm). The wear of Si3N4 balls is attributed to the abrasion effect of TD grains and the cutting action of the MD/BD (big size diamond with sizes of ~25 mm) asperities. It is specially noted that a markedly enhanced wear resistance of PDC can be achieved by the annealing treatment at 750
C. These results would give significant instructions for regulation and control of the tribological properties of PDC by simple and low cost annealing treatment. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Polycrystalline diamond compact Tribological property Annealing temperature Ambient air Si3N4 ball

1. Introduction Polycrystalline diamond compact (PDC), an ultrahard composite diamond related material, has been widely applied in machining tools, thrust bearings and drilling bits due to its perfect properties such as high hardness, excellent wear resistance and good impact toughness [1e3]. Tribological properties are significant engineering factors for evaluating the quality of PDC tools. In the past decades, various techniques for regulating and controlling the tribological properties of PDC have been extensively investigated by many scientific researchers and engineers. Most of the previous efforts have been devoted to improve the tribological properties of PDC and other diamond related materials by optimizing the chemical states and physical structures. Qian et al. [4] sintered polycrystalline diamond (PCD) with the replacement of metallicbinders such as cobalt by inorganic binders such as magnesium carbonate. The PCD with binder of magnesium carbonate behaved a higher wear resistance than that of the PCD with cobalt binder,

* Corresponding author. School of Engineering and Technology, China University of Geosciences (Beijing), Beijing 100083, PR China. E-mail addresses: [email protected], [email protected] (W. Yue). http://dx.doi.org/10.1016/j.carbon.2017.01.092 0008-6223/© 2017 Elsevier Ltd. All rights reserved.

which was ascribed to the higher thermal stability and chemically inert of magnesium carbonate. The refinement method of diamond grains had also been utilized to enhance the wear resistance of diamond tools in the previous researches [5,6]. Konicek et al. [7] explored the origin of ultralow frictional properties of the ultrananocrystalline diamond and they confirmed that the dissociative passivation either by H2O or H2 was responsible for the lower friction coefficients and wear rates. Detonation nanodiamonds (ND) and epoxy-ND composites had hold great potentials to improve tribological characteristics of composites, and it was recently confirmed by Neitzel et al. [8,9]. Special efforts were devoted to diamond like carbon (DLC) and metal-doped DLC coatings. DLC coatings can acquire the greatly low friction coefficients because of its low shearing properties [10,11], which could be acted as a solid lubricant coating covered on other diamond related materials. The doped metal elements had a markedly effect on modifying the microstructures and reducing the friction coefficients of DLC layers [12e14]. Therefore, depositing the DLC and metal-doped DLC coatings were the practical approach to control the frictional property of diamond related materials. Recently, Dwivedi et al. [15] have discovered that the film-substrate interface engineering could effectively control the friction and wear

104

J. Li et al. / Carbon 116 (2017) 103e112

behaviors of ultrathin carbon film, which were determined by the integrated framework of surface passivation, rehybridization, material transfer and tribolayer formation. Another new effective measure, sintering the PCD with a superhard cutting insert of chemical vapor deposition (CVD) diamond, was designed for manufacturing cutting and drilling tools, which greatly strengthened their abrasion resistance [16,17]. However, these modified technologies were confronted with a common issue that was high cost and high manufacturing difficulty. Annealing treatment, a conventional method with the advantage of convenience and high efficiency, has been widely used to improve the mechanical properties of engineering materials, especially for metallic materials. For diamond related materials, some previous efforts were devoted to the DLC and CVD diamond. Wang and Tokuta et al. [18,19] found that the pre-heat treatment above 500
C could cause the structural changes and the formation of graphite layer on DLC surface. The lower shearing resistance was obtained from the graphite layer between the sliding surfaces. Unfortunately, the super low friction coefficient for these diamond materials was always at the sacrifice of the wear resistance under dry conditions, which may be ascribed to the graphitization and reduction of hardness [18,20,21]. Khomich and Ralchenko et al. [22,23] have studied the effect of high annealing temperature on structures of CVD diamond films under vacuum condition (105 torr). It was showed that the allotrope content was determined by the annealing temperatures. Graphite and microcracks appeared at the annealing temperature above 1300 and 1575
C, respectively. However, the tribological properties of the annealed CVD diamond were not further explored. Yu et al. [24] recently indicated that the surface graphite induced by vacuum heat treatment could reduce the friction coefficient of CVD diamond, whereas the microcracks caused by vacuum heat treatment resulted in a serious wear. Until now, the approach to controllable tribological properties of sintered PDC through annealing treatment has not attracted the attention of researchers. It may attribute to previous studies on the thermal stability of PDC. As the previous references reported [25,26], some tiny spalling pits without chemical conversions appeared on PDC surfaces at the annealing temperature below 500
C. With further elevating the annealing temperature (600e800
C), the microcracks, surface graphitization, oxidation and extrusion of Co phase appeared. Wang et al. [27] heated the PCD layer in air and nitrogen atmosphere at 700
C for 10 min and then studied its machinability. It revealed that microcracks caused by thermal stress greatly reduced the wear resistance. Deng et al. [20] further verified that the microcracks, oxidation and extrusion of Co phase were detrimental to the wear resistance of PDC under the in situ high temperature conditions (600e700
C). Hence, it seemed that the tribological property of PDC was unlikely to be promoted by a pre-heat treatment, especially at high temperature. However, some contrary results have been achieved in present works. The tribological properties markedly depend on annealing temperature and the pre-heated PDC samples under a higher temperature (700 and 750
C) will have a perfect comprehensive tribological property, with a relatively low friction coefficient and nearly undetectable wear. This paper provided a systematic work on the tribological properties of PDC samples depending on the annealing temperatures in ambient air. Special efforts had been devoted to understand the effects of tribological surface characteristics on the tribological properties of annealed PDC samples. The mechanisms of the formation of transfer film, and the wear mechanisms of PDC samples and Si3N4 balls were particularly discussed. It would significantly instruct the engineers to regulate and control the tribological properties of PDC by the annealing treatment for industrial applications.

2. Experimental details 2.1. Characteristics of materials PDC was sintered at Zhongnan Diamond Co., Ltd. It consists of a cemented carbide substrate (WC-16wt%Co) and a sintered polycrystalline diamond (PCD) layer. The PDC, as shown in Fig. 1b, was cut into 1  1 cm2 samples by wire-electrode cutting. Fig. 1a illustrates that the total thickness of PDC sample is 3.7 mm with 3.2 mm of WC-Co substrate and 0.5 mm of PCD layer. The previous results [26] indicated that the PCD layer was mostly composed of diamond, Co and WC phase (Fig. 1c). Fig. 2a specifically shows that the big size diamond grains (BD grains, sizes of ~25 mm), medium size diamond grains (MD grains, sizes of 5e15 mm) and tiny diamond grains (TD grains, sizes of 0e5 mm) are compactly distributed on PDC surface. Moreover, some tiny holes and Co binder are enriched among those diamond grains, especially among the MD and TD grains. According to references [28,29], the diamond grains were bonded by carbon-carbon bonds, which insured the sufficient stability and strength of PCD layer. Surface topography of PDC and the corresponding EDS mapping images can be found in Fig. 2. They further confirm that the content of W is negligible. Moreover, the Si3N4 balls, with a diameter of 6 mm and surface roughness of 15e20 nm, are selected as counterparts in tribotests. 2.2. Annealing tests The PDC samples, ultrasonically rinsed with acetone and alcohol for 30 and 15 min, respectively, and then they were annealed at temperature from 200 to 750
C in a SX-8-10 muffle furnace for 30 min in ambient air. The temperature was detected by a thermocouple with a deviation of ±20
C. Then the annealed PDC samples were cooled with the process of air-cooling. 2.3. Ball-on-disc tribotests The tribotests were performed on a CSM ball-on-disc rotation tribometer in ambient air (relative humidity, 35 ± 3%). The Si3N4 ball was the upper sample and the PDC acted as a disc. Prior to the tribotests, both Si3N4 balls and annealed PDC samples were firstly ultrasonically rinsed with acetone for 30 min, and then with alcohol for 15 min. During the tribotest, the Si3N4 ball was fixed, while the PDC sample was rotated with the rotation radius of 2.5 mm and

(a)

(b)

3.2 mm WC

(c)

1 cm

0.5 mm PCD

Fig. 1. Optical images and composition of PDC: (a) cross section image of PDC, (b) image of PDC, (c) XRD pattern of PDC surface. (A colour version of this figure can be viewed online.)

J. Li et al. / Carbon 116 (2017) 103e112

(a)

105

(b)

T D grain

(d)

(c)

B D grain Co binder

Co

C

Tiny holes

W

M D grain 25 µm

25 µm

Fig. 2. SEM images of the PDC surface: (a) Morphology of the PDC surface, (b), (c) and (d) are the corresponding EDS mapping images of (a). (A colour version of this figure can be viewed online.)

rotation speed of 400 rpm corresponding to a liner velocity of ~105 mm/s. The normal load was 20 N and the corresponding initial Hertz contact pressure was ~1.89 GPa. All the tribotests were last for 30 min with the corresponding sliding distance of 188.5 m. In order to guarantee the reliability of all the data, each tribotest in this work was performed three times with the same tribological equipments in approximately similar relative humidity and temperature environment.

LabRAM HR Evolution spectrometer using a 514.5 nm line of Arþ laser to investigate the graphitization. The SSX-550 field emission scanning electron microscope (SEM) equipped with an energydispersive X-ray spectroscope (EDS) was used to detect the microstructures and chemical compositions of the PDC samples.

3. Experimental results 3.1. Surface characteristics

2.4. Microanalysis methods An optical microscope (OLYMPUS BX51M) with a Nikon camera was used to record the images of the PDC samples and the wear tracks of Si3N4 balls. The surface roughness and morphologies of PDC samples were measured by the white light interferometer (NanoMap-D). An automated D/max-2000 diffractometer was used to obtain the X-ray diffraction (XRD) patterns to study the phase transformation. Raman spectroscopy analysis were performed on a

(a) Air 25ć

(a1)

The surfaces of PDC samples displayed various geometrical morphologies and chemical conversions at the temperature from 200 to 750
C in ambient air. Previous researches [26] specifically showed that the PDC surfaces were damaged in ambient air including variations of morphology and chemical conversions, which was attributed to the thermal stress, oxidation and graphitization. At 200
C, the morphology of PDC surface was almost the same as that of the original PDC (Air 25
C PDC) surface (Fig. 3a).

Stress-induced TD grains

(b) Air600ć Ϩ

Tiny holes

200ć

ϩ

Tiny holes (a2)

25 µm

500ć

Spalling pits

(b1)

Ϫ

25 µm

5 µm

(d) Air 750ć

(c) Air 700ć

Oxidized holes Spalling pits

Spalling pits 25 µm

25 µm

Fig. 3. SEM images of the annealed PDC surfaces: (a) [26] air 25
C, where the insert (a1) is the image of air 200
C and (a2) is the image of air 500
C, (b) air 700
C, where the insert (b1) is the enlarge image of selected region Ⅱ, (c) air 700
C, (d) air 750
C. (A colour version of this figure can be viewed online.)

106

J. Li et al. / Carbon 116 (2017) 103e112 Table 1 Surface roughness (Ra) of the annealed PDC surfaces under air condition.

Fig. 4. Area ratios of spalling pits on annealed PDC surfaces. (A colour version of this figure can be viewed online.)

Temperature (
C)

25

200

300

400

500

600

700

750

Surface roughness (nm)

4.3

3.8

6.6

10.8

15.1

23.8

34.8

69.5

steady period, respectively. It can be observed that for all the tested samples, the friction coefficient curves evolve from a running-in period (~10 min). The running-in periods differ significantly with regard to the annealing temperatures, which is generally attributed to the variations of surface microstructures and chemical states induced by thermal effects. The stable friction coefficients of PDC samples are affected by the annealing temperatures. It is found that the air 25
C PDC sample has a stable friction coefficient of ~0.082, while the annealed PDC sample at 200
C gets the lowest stable friction coefficient of ~0.062. In addition, the mean friction coefficients rise from 0.062 to 0.091, when the annealing temperatures elevated from 200 to 750
C. It is noticed in Fig. 6c that considerable fluctuations of mean friction coefficients exist in the steady period for the PDC samples annealed at 400 and 500
C. It may be ascribed to the absence of transfer films between the tribological surfaces. 3.3. Wear properties

Fig. 5. XRD patterns of annealed PDC surfaces. (A colour version of this figure can be viewed online.)

Moreover, physical spalling pits had appeared on the PDC surface (no chemical conversion) with the annealing temperature increasing up to 500
C (Fig. 3a2). The quantity of spalling pits almost linearly depends on annealing temperature (200e500
C). A clear relationship between the area ratio of physical spalling pits and annealing temperature can be found in Fig. 4. At 600
C, many stress-induced TD grains appear on PDC surface (Fig. 3b). It might be related to the higher thermal stress caused by the different thermal expansion coefficients between diamond (3.2  106/K) and Co binder (14.4  106/K) [30]. The XRD patterns (Fig. 5) indicate that no visible new phase is found on the PDC surface at 600
C, and it presents a good thermal resistance. As the PDC is annealed at 700
C, large amounts of cobalt oxide (Co3O4, CoO) together with small amount of amorphous carbon and graphite are found, which implies that some chemical reactions have occurred. PDC treated at 700
C also has a more serious surface quality than that of 600
C. At 750
C, some tiny oxidized holes appear around the spalling pits, which lead to a coarser surface of PDC. Table 1 is given for the surface roughness (Ra, measured by the white light interferometer) of the annealed PDC surfaces and it shows that surface roughness is positively related to the annealing temperature, especially at 700 and 750
C. 3.2. Friction properties Fig. 6a presents the friction coefficient curves of the annealed PDC samples sliding against Si3N4 balls, and Fig. 6b and c shows the enlarged friction coefficient curves and mean friction coefficients in

3.3.1. Morphologies and wear rates of PDC samples The wear morphologies of annealed PDC samples are showed in Fig. 7. Many discontinuous wear tracks (seem like exfoliated pits) are formed on the PDC surfaces. It is noted that the PDC sample annealed at 200
C has a much wider wear track than that of air 25
C PDC sample, and the wear rates are 37.3  1012 and 23.0  1012 mm3/(N mm), respectively. Whereas, the wear rate of PDC samples decrease with the annealing temperatures from 200 to 500
C (Fig. 7bee), which can be further confirmed by the drop of the wear rate down to 3.2  1012 mm3/(N mm) for 500
C treated PDC in Fig. 8. At a higher annealing temperature of 600
C, a suddenly exacerbated wear track with wear rate of 40.1  1012 mm3/ (N mm) appears on PDC surface, which is explained by the exfoliated effect of MD grains on the sliding PDC surface. PDC samples annealed at 700 and 750
C with some chemical conversions (Fig. 5), oxidation (Fig. 5) and physical spalling pits (Fig. 3c and d) have the extremely low wear rates of 2.93  1012 and nearly 0 mm3/(N mm), respectively. It indicates that the annealing treatment can achieve the controllable wear properties of PDC. 3.3.2. Morphologies and wear rates of Si3N4 balls The characteristics (morphologies and sizes) of the wear tracks on Si3N4 balls depend on the annealing temperature of PDC samples. Fig. 9 shows that the transfer films (colorful attachments) are unevenly distributed on the wear tracks of Si3N4 balls. In addition, the transfer film area of the Si3N4 ball sliding against air 200
C PDC is larger than that of the Si3N4 ball sliding against air 25
C PDC and it decreases with the annealing temperature from 200
C to 500
C. When the annealing temperature of PDC comes to 600
C, 700
C and 750
C, enlarged areas of the transfer film is formed on the wear tracks. Many references [31e34] have also reported that the transfer films on wear tracks can reduce friction coefficients during the wear process. Si3N4 balls with lower hardness exhibit more enhanced wear than that of the PDC samples in many harsh environments [21,28,29]. It is clearly showed in Figs. 8 and 10 that the wear rates of Si3N4 balls are larger than those of PDC samples by one order of magnitude. The mean wear rates of Si3N4 balls from Fig. 9aeh are

J. Li et al. / Carbon 116 (2017) 103e112

(b)

(a)

107

(c) Steady period Considerable fluctuation

Running-in period Steady period

Fig. 6. Friction coefficient of annealed PDC samples sliding against Si3N4 balls: (a) friction coefficient curves in the whole sliding period, (b) the enlarged friction coefficient curves during the steady period in (a), (c) the mean friction coefficient during the steady period. (A colour version of this figure can be viewed online.)

Fig. 7. Three-dimensional surface topographies of wear tracks on annealed PDC surfaces. (A colour version of this figure can be viewed online.)

16.3, 22.0, 11.1, 8.69, 5.94, 14.1, 15.1 and 25.1  1011 mm3/(N mm), respectively. It is apparently found that the variation tendency of wear rates of Si3N4 balls is nearly similar to that of PDC samples (25e600
C). The wear rates of Si3N4 balls are controlled by abrasive action of the TD grains between tribological surfaces, and larger quantity of TD grains will lead to the enhanced wear of Si3N4 balls. As the annealing temperature goes up to 700 and 750
C, the Si3N4 balls present an anabatic wear rate (enhanced wear) in contrast with those of the counterpart PDC samples, which is ascribed to the cutting of the stiff surface MD and BD asperities.

3.4. Analysis of transfer films 3.4.1. SEM and EDS analysis In order to deeply understand the characteristics of the transfer films attained on the wear tracks, scanning electron microscopy (SEM) and energy dispersive X-ray spectrometry (EDS) were conducted to investigate their morphology and composition. Fig. 11 distinctly exhibits that there is some colorful plate like transfer film on the wear track of Si3N4 ball. Qin et al. [21] had systematically studied the characteristics of transfer film formed at different relative humidity conditions and the results indicated that it was a

108

J. Li et al. / Carbon 116 (2017) 103e112

Fig. 8. Wear rates of annealed PDC samples. (A colour version of this figure can be viewed online.)

Fig. 10. Wear rate of Si3N4 balls slid against various treated PDC samples. (A colour version of this figure can be viewed online.)

carbonaceous transfer film with the thickness of 10e500 nm, which could markedly reduce the friction coefficient. Fig. 12 is SEM image and the corresponding EDS mapping images of transfer film. It illustrates that the transfer film contains large amount of carbon, which comes from PDC surface. However, it is not clear whether the transfer film contains Si and N, because the thickness of the transfer film is sub-micrometer and the Si3N4 in subsurface can be detected.

carbon and graphite exist in the platelike carbonaceous transfer film.

3.4.2. Raman spectroscopy analysis Raman spectroscopy was used to further explore the chemical state of carbonaceous transfer films. Fig. 13 displays the Raman spectra of the transfer films. It shows that the characteristic peaks of Si3N4 exist in all the Raman curves. Taking consideration of the analytical depth of Raman spectroscopy (several micrometers), it indicates that Si3N4 phase may come from the Si3N4 substrate. The characteristic peaks marked as D- and G-peak at ~1360 and ~1580 cm1, respectively, correspond to a-C (non-diamond phase including disordered graphite and sp2-hybridized carbon phase) and sp2-C (a graphite structure with sp2 bond character) [35,36]. It is noted that all Raman curves of carbonaceous transfer films possess strong D- and G-peak, which proves that the disordered

4. Discussion 4.1. Mechanisms of friction properties Numerous studies [20,21,28,29] have emphasized that PDC samples sliding against various materials possess ultralow friction coefficients (0e0.2) both in scientific laboratory tribotests and industrial applications in ambient air. Transfer films formed on the worn surfaces of PDC samples and their counterparts during the running-in periods are believed as solid lubricants, which can obviously reduce the friction coefficient. In present experiments, the carbonaceous transfer films are found during running-in period and the detailed formation process is exhibited in Fig. 14. It is clearly showed that TD grains tend to locate at the regions surrounded by many defects of tiny holes among the BD grains (in Fig. 2a). Previous works [26] indicated that the TD grains were easy to exfoliate from PDC surface under high stress and high temperature conditions. Fig. 14a and b shows that the tribological surface

Fig. 9. Surface topographies of wear tracks on the Si3N4 balls slid against various treated PDC samples. (A colour version of this figure can be viewed online.)

J. Li et al. / Carbon 116 (2017) 103e112

109

Fig. 13. Raman spectra of transfer films, and the insert presents the Raman spectra of air 25
C and original Si3N4 ball. (A colour version of this figure can be viewed online.)

Fig. 11. Morphologies of wear track of Si3N4 balls and transfer films: (a) and (c) are the optical and SEM images of the whole wear track of Si3N4 ball, respectively, (b) and (d) are enlarged transfer film morphologies of the selected region Ⅰ of (a) and Ⅱ of (c), respectively. (A colour version of this figure can be viewed online.)

(a)

(b)

Si3N4

C

5 µm

(c)

N

Transfer film 5 µm 5 µm

(f)

Co

5 µm

(e)

O

5 µm

(d)

Si

5 µm

Fig. 12. SEM image and the corresponding EDS mapping images of transfer film on wear track of Si3N4 ball. (A colour version of this figure can be viewed online.)

generates the high initial Hertz contact pressure (~1.89 GPa) and high temperature produced by high speed rubbing [37e39]. The sliding Si3N4 ball with vibration characteristic will effortlessly result in the exfoliation of TD grains around the contact location (Fig. 14b). Moreover, these exfoliated TD grains are easily grinded between tribological surfaces during the running-in period and then realize the chemical conversion from diamond to disordered graphite, amorphous carbon phase and pure graphite [26], shown in Fig. 14c and d. Therefore, the carbonaceous transfer film is rapidly built up and the friction coefficient evolves into a steady solid lubrication state. The PDC sample annealed at 200
C, without any chemical conversion and variation of morphology, reveals the lowest friction coefficient of ~0.062, which is attributed to the relief of residual compressive stress. Many previous works have reported

Fig. 14. Formation mechanism of carbonaceous transfer films during running-in period. (A colour version of this figure can be viewed online.)

that residual compressive stresses (0.5e2.8 GPa) could exist in the PCD layer after sintering process [40e42]. Raman spectroscopy is reliably used to measure residual stress by determining the Raman shift and approximated as being biaxial to calculate the magnitudes of stress [43]. Fig. 15 shows that the residual compressive stresses of air 25
C and air 200
C PDC surface are 0.93 and 0.04 GPa, respectively, which are calculated by the formula as following [44]:

h



sb ðGPaÞ ¼ g0  g cm1

i. 1:62

(1)

where sb is the residual stress, g0 is the unstressed diamond wavenumber (1332 cm1) and g is the measured diamond wavenumber on the PDC surface. It should be mentioned that a higher residual compressive stress could improve the stability of TD grains on PDC surface during sliding tests. The air 200
C PDC surface without oxidization of TD grains [45] possesses a lower compressive stress (0.04 GPa), which will easily lead to the exfoliation of TD grains during the running-in period. The more quantity of the exfoliated TD grains between tribological surfaces, the more carbonaceous transfer films are formed on the wear tracks of Si3N4

110

J. Li et al. / Carbon 116 (2017) 103e112

surface roughness could cause a rise of friction coefficient. Hence, the low friction coefficient is dominated by the effects of carbonaceous transfer films and surface roughness. As the annealing temperatures go up to 700 and 750
C, the surface roughness rapidly increase and TD grains are rarely found on PDC surface, which may greatly contribute to a higher friction coefficient. Whereas, the steady and low friction coefficients (<0.1) are presented in Fig. 6b. It is ascribed to the mass of colorful attainments on the wear tracks of Si3N4 balls (Fig. 9g and h). Raman curves in Fig. 12 also confirm that the colorful attainments are carbonaceous transfer films, which may come from surface graphitization and markedly reduce the friction coefficients.

4.2. Wear mechanisms of annealed PDC

Fig. 15. Raman spectra of annealed PDC surfaces. (A colour version of this figure can be viewed online.)

ball. Hence, it creates a more unstable running-in period and eventually results in the lowest friction coefficient. At higher annealing temperatures from 300 to 500
C, large amount of TD grains have exfoliated and oxidized due to the thermal effect [26,45]. In addition, Fig. 4 shows that a positively dependent relationship presents between the number of exfoliated TD grains and the annealing temperature. It means that there is a decrease of the number of TD grains with the increasing temperatures from 300 to 500
C. Therefore, the sizes of carbonaceous transfer films decrease and the mean friction coefficients increase. When PDC samples are annealed at 600
C, many stress-induced TD grains with the diameters of 1e5 mm appear on the PDC surfaces, which play a similar role as TD grains and generate more carbonaceous transfer films. However, the PDC annealed at 600
C possesses a higher surface roughness of 23.8 nm than that (15.1 nm) of the PDC annealed at 500
C. Sha et al. [29] pointed out that a high

The PDC is synthesized by sintering microcrystalline diamond powder (nominal grain sizes of 2e20 mm) to a WC-Co substrate [46,47]. The bonding strength among the TD grains is weaker than other nano/microcrystalline diamond materials such as nano/ microcrystalline CVD diamond [48,49], nanocrystalline diamond aerogel [50] and aminated Nanodiamond-Epoxy Composites [51,52]. It is reported recently that some diamond grains with weak bonds on PDC surface are inclined to be pulled out in unstable running-in period especially in dry and vacuum environments [21,28,29]. In present works, more TD grains existed in tribological surface always result in larger wear rate of the annealed PDC. Reasonable explanation is related to exfoliation of MD grains caused by the effect of grinding and fluctuating of TD grains between the tribological surfaces. It is noted that PDC surface is composed of TD grains, MD grains and BD grains with corresponding diameters of 0e5, 5e15 and ~25 mm, respectively (Fig. 2a). In addition, the TD grains are more easily exfoliated into the contact location. Some of the grinded and fluctuated TD grains will result in the initiation of cracks around MD grains and then lead to the pulling out of MD grains (Fig. 16). As a result, the discontinuous exfoliated pits (seen in Fig. 16b, with diameters of 5e15 mm) are eventually formed on the surface of PDC samples. Basing on the above idea, it is reasonable to understand that the wear rates of the

(a)

TD grains MD grain 5 µm

(c)

(b)

Exfoliated pit 5 µm

Exfoliated pits

Cracks 25 µm

B D grain

Fig. 16. SEM images of PDC surfaces: (a) unworn air 25
C PDC surface, (b) wear track on air 25
C PDC surface, (c) wear track on air 600
C PDC surface. (A colour version of this figure can be viewed online.)

J. Li et al. / Carbon 116 (2017) 103e112

(a)

111

(c)

(b) Wear track Spalling pits

Region Ϩ Spalling pits

Region ϩ 0.5 mm

10 µm

10 µm

Fig. 17. SEM images of PDC surfaces: (a) SEM images of the slid air 750
C PDC surface, (b) unworn PDC surface (region Ⅰ), (c) wear track (region Ⅱ). (A colour version of this figure can be viewed online.)

PDC samples have an inverse dependence on the annealing temperatures (200e500
C). However, a large amount of stress-induced TD grains appear on PDC annealed at 600
C (Fig. 3b). These stressinduced TD grains may play a similar role as TD grains and result in the accelerating pulling out of MD grains during the running-in period (Fig. 16c). Contrast to the PDC annealed at 600
C, some chemical conversions and oxidizations occur on the PDC surfaces at higher temperature of 700 and 750
C, which establish much rougher PDC surfaces with more MD and BD asperities and fewer TD grains (Fig. 3d). The MD and BD grains with the mild graphitization (Fig. 5) are so hard that it is difficult to be worn by Si3N4 balls in absence of TD grains. Fig. 17 further shows that there is nearly no visible wear on the air 750
C PDC surface, which is consistent with the results in Figs. 7 and 8. Therefore, the exfoliation of MD grains caused by the grinding and fluctuation of TD grains is the main wear mechanism of the annealed PDC samples.

4.3. Wear mechanisms of Si3N4 balls Previous works have demonstrated that the wear process of Si3N4 ball sliding against PDC sample is nearly completed during the running-in period [21]. In order to understanding the wear mechanisms of Si3N4 balls in present tribotests, a reasonable interpretation combined with some evidences is given. It should be concerned that the TD grains, with a higher hardness of 30e40 GPa than that of Si3N4 balls (15e20 GPa) [21,28], will bring about many furrows on the wear tracks of Si3N4 balls (Fig. 9). It indicates that the abrasive wear occurs during sliding process. The SEM image of wear track on Si3N4 ball is shown in Fig. 18. It further presents that the furrows not only exist on the substrate of Si3N4 ball but also appear on the carbonaceous transfer film. Therefore, the wear mechanism of Si3N4 ball is abrasive wear, which has a strong

dependence on the quantity of TD grains (25e600
C). However, some doubts will be turned to the wear mechanisms of Si3N4 balls sliding against the PDC annealed at 700 and 750
C. Fig. 3c and d shows that the TD grains barely appear on the PDC surface because of the exfoliation caused by the thermal stress. As a result, it is impossible for TD grains to produce the abrasive wear on Si3N4 balls. The wear of Si3N4 balls will be attributed to the cutting effect of the ultrahard MD and BD asperities on PDC surface, which is directly confirmed by the cutting morphologies in Fig. 9 g and h. 5. Conclusions Tribological properties of the annealed PDC samples (25e750
C) sliding against Si3N4 balls were systematically explored to give some guidance for regulating and controlling the friction coefficient and wear rate by the simple and low cost annealing method. Carbonaceous transfer films can effectively reduce the friction coefficients of the annealed PDC (25e600
C), which indicates that the low temperature (200
C) annealing treatment can be used to improve the friction properties of PDC tools. With annealing temperature elevating from 700 to 750
C, the friction coefficients are dominantly controlled by surface roughness and carbonaceous transfer films. The wear rates of PDC greatly depend on the annealing temperatures, and it decreases from 200 to 500
C. Special attention should be paid to the wear rate of the PDC annealed at 750
C. It is surprised that the wear resistance is markedly improved to a nearly undetectable wear level, which can significantly instruct engineers to regulate and control the wear properties of PDC tools. What’s more, it is concluded that the exfoliation of MD grains is the main reason for the wear of the annealed PDC, and the wear of Si3N4 balls is attributed to the abrasion effect of TD grains (sliding against 25e600
C annealed PDC) and the cutting action of the MD/BD asperities (sliding against 700, 750
Cannealed PDC). Acknowledgments

Transfer films

Furrows

The authors would like to thank the National Natural Science Foundation of China (51375466, 41572359), International Science and Technology Cooperation Project of China (2011DFR50060), Fundamental Research Funds for the Central Universities (53200859604, 2652015077) and Beijing Nova Program for the financial support. The authors would like to thank professor Yusheng Li and Dr. Jianning Cai for assistance in revising this paper.

10 µm References Fig. 18. SEM image of wear track on Si3N4 ball. (A colour version of this figure can be viewed online.)

[1] T.N. Sexton, C.H. Cooley, Polycrystalline diamond thrust bearings for down-

112

J. Li et al. / Carbon 116 (2017) 103e112

hole oil and gas drilling tools, Wear 267 (5) (2009) 1041e1045. [2] C.W. Knuteson, T.N. Sexton, C.H. Cooley, Wear-in behaviour of polycrystalline diamond thrust bearings, Wear 271 (9) (2011) 2106e2110. [3] K.H. Park, A. Beal, D. Kim, P. Kwon, J. Lantrip, Tool wear in drilling of composite/titanium stacks using carbide and polycrystalline diamond tools, Wear 271 (11e12) (2011) 2826e2835. [4] J. Qian, C.E. McMurray, D.K. Mukhopadhyay, J.K. Wiggins, M.A. Vail, K.E. Bertagnolli, Polycrystalline diamond cutters sintered with magnesium carbonate in cubic anvil press, Int. J. Refract. Met. Hard Mater 31 (3) (2012) 71e75. [5] J. Hu, Y.K. Chou, R.G. Thomopson, J. Burgess, S. Street, Characterizations of nano-crystalline diamond coating cutting tools, Surf. Coat. Technol. 202 (4e7) (2007) 1113e1117. [6] K. Harano, T. Satoh, H. Sumiya, Cutting performance of nano-polycrystalline diamond, Diam. Relat. Mater 24 (24) (2012) 78e82. [7] A.R. Konicek, D.S. Grierson, P.U.P.A. Gilbert, W.G. Sawyer, A.V. Sumant, R.W. Carpick, Origin of ultralow friction and wear in ultrananocrystalline diamond, Phys. Rev. Lett. 100 (23) (2008) 1151e1156. [8] I. Neitzel, V. Mochalin, I. Knoke, G.R. Palmese, Y. Gogotsi, Mechanical properties of epoxy composites with high contents of nanodiamond, Compos. Sci. Technol. 71 (5) (2011) 710e716. [9] I. Neitzel, V. Mochalin, J.A. Bares, R.W. Carpick, A. Erdemir, Y. Gogotsi, Tribological properties of nanodiamond-epoxy composites, Tribol. Lett. 47 (2) (2012) 195e202. [10] T.B. Ma, L.F. Wang, Y.Z. Hu, X. Li, H. Wang, A shear localization mechanism for lubricity of amorphous carbon materials, Sci. Rep. 4 (1) (2014) 3662. [11] K. Bewilogua, D. Hofmann, History of diamond-like carbon films e from first experiments to worldwide applications, Surf. Coat. Technol. 242 (4) (2014) 214e225. [12] C. Corbella, E. Bertran, M.C. Polo, E. Pascual, J.L. Andujar, Structural effects of nanocomposite films of amorphous carbon and metal deposited by pulsed-DC reactive magnetron sputtering, Diam. Relat. Mater 16 (10) (2007) 1828e1834. [13] C. Liu, G.Q. Li, W.W. Chen, Z.X. Mu, C.W. Zhang, L. Wang, The study of doped DLC films by Ti ion implantation, Thin Solid Films 475 (1) (2005) 279e282. [14] F. Zhao, H.X. Li, L. Ji, Y.J. Wang, H.D. Zhou, J.M. Chen, Ti-DLC films with superior friction performance, Diam. Relat. Mater 19 (4) (2010) 342e349. [15] N. Dwivedi, R.J. Yeo, Z. Zhang, C. Dhand, S. Tripathy, C.S. Bhatia, Interface engineering and controlling the friction and wear of ultrathin carbon films: high sp3 versus high sp2 carbons, Adv. Funct. Mater 26 (10) (2016) 1526e1542. [16] R.K. Bogdanov, A.A. Shuzhenko, A.P. Zakora, A.M. Isonkin, V.G. Gardin, New superhard material for drilling tools, J. Superhard Mater 29 (1) (2007) 56e63. [17] G. Yan, W. Yue, D.Z. Meng, F. Lin, Z.Y. Wu, C.B. Wang, Wear performances and mechanisms of ultrahard polycrystalline diamond composite material grinded against granite, Int. J. Refract. Met. Hard Mater 54 (2016) 46e53. [18] L.L. Wang, X. Nie, X. Hu, Effect of thermal annealing on tribological and corrosion properties of DLC coatings, J. Mater. Eng. Perform. 22 (10) (2013) 3093e3100. [19] Y. Tokuta, M. Kawaguchi, A. Shimizu, S. Sasaki, Effects of pre-heat treatment on tribological properties of DLC film, Tribol. Lett. 49 (2) (2013) 341e349. [20] J.X. Deng, H. Zhang, Z. Wu, A.H. Liu, Friction and wear behavior of polycrystalline diamond at temperatures up to 700
C, Int. J. Refract. Met. Hard Mater 29 (5) (2011) 631e638. [21] W.B. Qin, W. Yue, C.B. Wang, Understanding integrated effects of humidity and interfacial transfer film formation on tribological behaviors of sintered polycrystalline diamond, RSC Adv. 5 (66) (2015) 53484e53496. [22] A.V. Khomich, V.G. Ralchenko, A.V. Vlasov, R.A. Khmelnitskiy, I.I. Vlasov, V.I. Konov, Effect of high temperature annealing on optical and thermal properties of CVD diamond, Diam. Relat. Mater 10 (5e7) (2001) 546e551. [23] V. Ralchenko, L. Nistor, E. Pleuler, A. Khomich, I. Vlasov, R. Khmelnitskii, Structure and properties of high-temperature annealed CVD diamond, Diam. Relat. Mater 12 (10) (2003) 1964e1970. [24] X. Yu, J. Ai, L. Yang, C.B. Wang, Exploring tribological behavior of diamond film by hot-filament chemical vapour deposition on tungsten carbide for lunar exploration, Vacuum 100 (2) (2014) 41e45. [25] S. Wang, H.T. Zhang, Study on mechanism of thermal damage on PCD compacts induced by induction heating, Mater. Sci. Technol. 13 (5) (2005) 492e495. [26] J.S. Li, W. Yue, C.B. Wang, Microstructures and thermal damage mechanisms of sintered polycrystalline diamond compact annealing under ambient air and vacuum conditions, Int. J. Refract. Met. Hard Mater 54 (2016) 138e147. [27] B.F. Wang, S.Y. Wang, Z. Tang, L.Q. Cheng, Mechanism of the effect of adhesive

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37] [38] [39] [40] [41]

[42]

[43]

[44] [45]

[46] [47]

[48]

[49] [50]

[51]

[52]

Co on the thermal stability of polycrystalline diamond compact, Min. Metall. Eng. 29 (5) (2009) 90e93. Y.H. Zhao, W. Yue, F. Lin, C.B. Wang, Z.Y. Wu, Friction and wear behaviors of polycrystalline diamond under vacuum conditions, Int. J. Refract. Met. Hard Mater 50 (2015) 43e52. X.H. Sha, W. Yue, Y.H. Zhao, F. Lin, C.B. Wang, Effect of sliding mating materials on vacuum tribological behaviors of sintered polycrystalline diamond, Int. J. Refract. Met. Hard Mater 54 (2016) 116e126. Z.C. Li, H.S. Jia, H.A. Ma, W. Guo, X.B. Liu, G.F. Huang, FEM analysis on the effect of cobalt content on thermal residual stress in polycrystalline diamond compact (PDC), Mech. Astron 55 (4) (2012) 639e643. Y. Liu, A. Erdemir, E.I. Meletis, An investigation of the relationship between graphitization and frictional behavior of DLC coatings, Surf. Coat. Technol. 86 (1996) 564e568. J.C. Sanchez-Lopez, A. Erdemir, C. Donnet, T.C. Rojas, Friction-induced structural transformations of diamond-like carbon coatings under various atmospheres, Surf. Coat. Technol. 163e164 (2) (2003) 444e450. J.D. Schall, G. Gao, J.A. Harrison, Effects of adhesion and transfer film formation on the tribology of self-mated DLC contacts, J. Phys. Chem. C 114 (12) (2009) 5321e5330. X.C. Chen, T. Kato, M. Nosaka, Origin of superlubricity in a-C: H:Si films: a relation to film bonding structure and environmental molecular characteristic, ACS Appl. Mater. Interfaces 6 (16) (2014) 13389e13405. R.M. Erasmus, J.D. Comins, V. Mofokeng, Z. Martin, Application of Raman spectroscopy to determine stress in polycrystalline diamond tools as a function of tool geometry and temperature, Diam. Relat. Mater 20 (7) (2011) 907e911. X. Yu, X.A. Zhao, Y.Y. Liu, H. Meng, X. Jiang, The effects of Ti carbonization on the nucleation and oriented growth of diamond films on tungsten carbide, ACS Appl. Mater. Interfaces 6 (7) (2014) 4669e4677. P.L. Tso, Y.G. Liu, Study on PCD machining, Inter. J. Mach. Tool. Manuf. 42 (3) (2002) 331e334. D. Belnap, A. Griffo, Homogeneous and structured PCD/WC-Co materials for drilling, Diam. Relat. Mater 13 (10) (2004) 1914e1922. M. Zeren, S. Karagozs, Sintering of polycrystalline diamond cutting tools, Mater. Des. 28 (3) (2007) 1055e1058. T.P. Lin, M. Hood, G.A. Cooper, R.H. Smith, Residual stresses in polycrystalline diamond compacts, J. Am. Ceram. Soc. 77 (6) (1994) 1562e1568. S.A. Catledge, Y.K. Vohra, R. Ladi, G. Rai, Micro-Raman stress investigations and X-ray diffraction analysis of polycrystalline diamond (PCD) tools, Diam. Relat. Mater 5 (10) (1996) 1159e1165. G.P. Xu, Q.W. Chen, Z.M. Yin, G. Xu, The effect of thickness of diamond layer on residual stresses in polycrystalline diamond compact, China J. High. Press. Phys. 23 (1) (2009) 24e30. H.S. Jia, H.G. Ma, X.P. Jia, Research on polycrystalline diamond compact (PDC) with low residual stress prepared using nickel-based additive, Int. J. Refract. Met. Hard Mater 29 (1) (2011) 64e67. M. Wieligor, T.W. Zerda, Surface stress distribution in diamond crystals in diamond-silicon carbide composites, Diam. Relat. Mater 17 (1) (2008) 84e89. S. Osswald, G. Yushin, V. Mochalin, S.O. Kucheyev, Y. Gogotsi, Control of sp2/ sp3 carbon ratio and surface chemistry of nanodiamond powders by selective oxidation in air, J. Am. Chem. Soc. 128 (35) (2006) 11635e11642. R.H. Wentorf, R.C. Devries, F.P. Bundy, Sintered superhard materials, Science 208 (4446) (1980) 873e880. R. Schouwenaars, V.H. Jacobo, A. Ortiz, Transition from normal to severe wear in PCD during high-speed cutting of a ductile material, Int. J. Refract. Met. Hard Mater 27 (2) (2009) 403e408. A. Nayak, H.D. Banerjee, Bonding characteristics and optical properties of amorphous carbon/diamond films deposited by an electron beam activated plasma CVD method, Phys. Status Solidi. 149(2) 629e635. V.N. Mochalin, O. Shenderova, D. Ho, Y. Gogotsi, The properties and applications of nanodiamonds, Nat. Nanotechnol. 7 (1) (2011) 11e23. P.J. Pauzauskie, J.C. Crowhurst, M.A. Worsley, T.A. Laurence, A.L. David Kilcoyne, Y. Wang, T.M. Willey, K.S. Visbeck, S.C. Fakra, W.J. Evans, J.M. Zaug, J.H. Satcher Jr., Synthesis and characterization of a nanocrystalline diamond aerogel, Proc. Natl. Acad. Sci. U. S. A. 108 (21) (2011) 8550e8553. V.N. Mochlin, I. Neitzel, B.J.M. Etzold, A. Peterson, G. Palmese, Y. Gogotsi, Covalent incorporation of aminated nanodiamond into an epoxy polymer network, ACS Nano 5 (9) (2011) 7494e7502. I. Neitzel, V.N. Mochalin, J. Niu, J. Cuadra, A. Kontsos, G.R. Palmese, Y. Gogotsi, Maximizing young’s modulus of aminated nanodiamond-epoxy composites measured in compression, Polymer 53 (25) (2012) 5965e5971.