An analysis of the tribological mechanism of GLC film in artificial seawater

Abstract

The tribological performances of the graphite-like carbon (GLC) films sliding against WC balls in distilled water (DW), artificial seawater (SW) and four types of saline solutions related to seawater were investigated. The GLC film was deposited by magnetron sputtering. The microstructure, mechanical properties and tribological performances of the GLC film were characterized by scanning electron microscopy (SEM), atomic force microscopy (AFM), Raman spectroscopy, nano-indentation and reciprocating ball-on-disk tribo-meter. The results showed that the smooth, dense and hard GLC film with significant sp²-hybridized carbon exhibited good tribological performance with low friction and wear not only in distilled water, but also in seawater. The tribological performance of the GLC film in seawater was closely related to the nature and constitution of seawater, as well as the nature and performance of the counterparts. Bulky wear debris with hard particles would generate a three-body wear regime, which increases the friction coefficient and wear rate of the GLC film slightly in seawater environment. In contrast, the divalent metal salts of seawater can decrease friction and wear between the two contact surfaces. The synergistic effect led to a relatively higher friction coefficient and wear rate of the GLC film against the WC counterpart in seawater than that in distilled water.

---

1. Introduction

Over the last few decades, the marine economy has been developing fast globally. In addition, the dependence of the marine economy on high-performance marine equipment, such as cargo ships, seawater pumps, underwater robots and hydraulic motors, have increased drastically.^1–3 Because the long-life and stable operation of the marine equipment used to rely on their moving parts, such as sealing rings, valve elements, water-lubricated bearings hydraulic system, gear, shaft, and propeller, advanced technique to reduce the friction coefficient and the wear loss of moving parts serviced in a seawater environment was emphasized in the modern marine industry.^4–6 Moreover, more and more people have focused on the application performance of materials in the ocean environment and developed new materials suitable for the ocean environment.⁷

Recently, the graphite-like carbon (GLC) film possessing an amorphous carbonaceous matrix with significant sp²-hybridized carbon has demonstrated high hardness, low residual stress, good adhesion force, low friction and wear performance in several environments, including ambient air, oil and water, having great potential to be used as the high-performance working surface of the moving parts operated in seawater.^8–19 For example, Yang et al. reported a GLC film named Graphit-iC consisting of very fine grains of graphite-like material but with some cross-bonding between the graphite-like layers (which causes the high hardness).¹⁰ Similar results were found by Stallard, who pointed out that Graphit-iC has a dense and amorphous matrix within fine regions of nano-crystalline graphite. More importantly, the Graphit-iC film exhibited high load bearing in combination with low friction and wear under three environmental conditions of air, water and oil.¹¹ Wang et al. reported the low friction and high load bearing of a GLC coating when sliding against GCr15 steel in air with a humidity of 70%.²⁰ Fujisawa et al. found that the tendency of a softer coating to process a greater sp² or graphite-like content provided more effective solid lubrication in a wet environment and therefore minimizing both wear and friction.²¹ Field et al. argued that the high hardness and lubricating transferred-layer formation led to the low friction and wear of the GLC film in ambient air, and the water molecules absorbed on the top surface of graphite weakened the bonds between the basal planes of its hexagonal structure and provided good frictional behavior in humidity or water environments.²² Wang et al. (State Key Laboratory of Solid Lubrication, LICP, CAS, China) demonstrated the nano-crystallites/amorphous matrix structure of the sputtering GLC film and its good friction-reduction and anti-wear functions to typical water-lubricated ceramics, including Si₃N₄, SiC and WC in distilled water.^23–25 Wang et al. (State Key Laboratory of Mechanics and Control of Mechanical Structures, NUAA, China) reported that the tribological performance of the amorphous carbon films with significant sp² carbon could be modified by Cr, which was also affected by different counterparts.^26,27 In addition, a suitable interlayer on the substrate surface could significantly improve the adhesion force of a GLC film. The most common metallic candidates for the adhesion interlayer were Ti, Cr and W.^28,29 Field et al. also showed that GLC-coated tool steel with a Cr interlayer exhibited high load-bearing capacity and excellent tribological performance in air, water and oil.¹¹

Although the high tribological performance with low friction and low wear of GLC film in distilled water was accepted; little is known about the tribological performance of the GLC film in a seawater environment. Therefore, this study performed a comparative study on GLC films in distilled water, artificial seawater and four solutions related to seawater constitutes. The GLC film was fabricated by magnetron sputtering. Metal Cr was pre-deposited on the substrate as an adhesive interlayer. Typical engineering ceramic WC was chosen as the sliding counterpart. The main object was to understand the tribological mechanism of the GLC film in seawater and probe the different effects of the seawater constitutes on the tribological performance of GLC film.

2. Experimental

2.1 Film preparation

Two types of substrates, including P(100) silicon wafers with dimension 30 mm × 20 mm × 0.625 mm and stainless steel 316L wafers with dimension 30 mm × 20 mm × 2 mm, were used. The roughness of the stainless steel 316L wafer was approximately Ra 50 nm. The former were used for the microscopy observations of the film microstructures and the latter were used for the mechanical and tribological tests. The depositing system was configured of three magnetron target positions, which focused on the substrate seat. The middle one fixed with the high pure graphite target (purity ≥ 99.95%), which was used to deposit the graphite-like amorphous carbon. The other two positions were twin targets installed Cr targets (purity ≥ 99.8%) for the deposition of metallic interlayer. The distances between cathode targets and fixed substrates were similar, which were approximately 100 mm. The graphite target was powered by a DC source. The twin Cr targets were powered by a mid-frequency (40 kHz) AC supply. The vacuum chamber was pumped using a molecular pumping system. The pressure in the chamber was controlled by vacuum gauges.

Prior to deposition, the substrates were cleaned ultrasonically in ethanol and acetone baths in succession and dried with a blower. The base pressure of the chamber prior to deposition was pumped to 1.0 × 10⁻³ Pa; subsequently, the deposition pressure 1.0 Pa was reached with a constant flow of Ar gas. When the pressure of the Ar atmosphere was reached, the substrates were DC sputter-cleaned for 15 min at a bias voltage −1000 V. The Cr layer was first deposited on the substrates followed by the GLC layers. The Cr interlayer was deposited with a target current 2.0 A at a bias voltage of −500 V for 10 min. The GLC layer was deposited at a graphite target current of 1.0 A at a bias voltage −400 V for 150 min.

2.2 Film characterization

The morphology of the GLC film was characterized by scanning electron microscopy (SEM, JSM-6701F, Japan). The surface roughness (Ra) and morphology of the GLC film was detected by atomic force microscopy (AFM, AIST-NT, USA) in contact mode. The scan area was 5 μm × 5 μm, the frequency was 1 Hz and the scan rate was 10 μm s⁻¹. The microstructures of the GLC film were also analyzed by Raman spectroscopy (LabRAM, HR800) with a 532 nm laser from 500 cm⁻¹ to 2750 cm⁻¹. The hardness and elastic modulus were measured using a nano-indenter (Nano Indenter, MTS Ltd, USA).

The tribological behaviors of the GLC film in different solutions were measured using a multifunctional tribo-meter (UMT-3 tribotester, Centre of Tribology, USA) according to the reciprocating ball-on-disk model. Commercial WC balls with diameter 3 mm were chosen as the sliding counterparts. The normal load was 2 N; the amplitude was 5 mm; the frequency was 5 Hz; and the test time was 30 min. Distilled water and artificial seawater were chosen as the test environments to detect the differences in the GLC film between the corresponding solutions. In addition, four saline solutions related different seawater constitutes were selected to analyze the tribological mechanism of the GLC film in seawater and the different effects of the different seawater constitutes on the GLC film. The artificial seawater was prepared according to the standard ASTM 1141-98, as shown in Table 1.³⁰ Four saline solutions included the first solution (NaCl), second solution (Na₂SO₄), third solution (CaCl₂, MgCl₂, and SrCl₂) and forth solution (KCl, NaHCO₃, KBr, H₃BO₃, and NaF), whose concentrations were consistent with seawater ASTM 1141-98, respectively. To simplify the expression, solutions 1, 2, 3 and 4 are abbreviated as “S1”, “S2”, “S3” and “S4”, respectively. The tests were run three times for each solution condition. The data error bars showed the confidence level. After the tribo-tests, the wear track depths of the GLC films were characterized by a contact step analytical instrument (Alpha Step-IQ). The specific wear rate was calculated from the volume loss using the following equation: W = V/(F × L), where V is the wear loss of the film in m³, F is the normal load in Newton and L is the total sliding distance in meters.³¹ The wear surfaces were examined by SEM, EDS and Raman spectroscopy.

Table 1 Chemical composition of the seawater and saline solutions

Constitutes	Concentration (g L^−1)
	Seawater	S1	S2	S3	S4
NaCl	24.53	24.53	—	—	—
Na2SO4	4.09	—	4.09	—	—
MgCl2	5.20	—	—	5.20	—
CaCl2	1.16	—	—	1.16	—
SrCl2	0.025	—	—	0.025	—
KCl	0.695	—	—	—	0.695
NaHCO3	0.201	—	—	—	0.201
KBr	0.101	—	—	—	0.101
H3BO3	0.027	—	—	—	0.027
NaF	0.003	—	—	—	0.003

---

3. Results and discussion

3.1 Microstructure of the GLC film

Cross-sectional and surface morphologies of the as-deposited GLC film are shown in Fig. 1. The cross-sectional morphology in Fig. 1(a) clearly reveals the two-layer microstructure of the as-deposited GLC film, including the adhesive Cr interlayer with a thickness of about 176 nm and amorphous carbon layer with a thickness of about 1324 nm. The total thickness of the as-deposited GLC film was approximately 1500 nm. In addition, Fig. 1(a) also demonstrates a dense film structure. As observed in Fig. 1(b), the surface morphology also has a dense film structure, which is composed of very fine grains on the nanoscale. The dense and fine microstructure generated a rather flat surface, which was demonstrated further by AFM observations, as shown in Fig. 1(c). The surface fluctuations were only limited to the nanometer range. The analyzed surface roughness was 6.5 nm. Thus, a dense and fine GLC film with smooth surface was fabricated successfully.

Fig. 1 Morphology of the as-deposited GLC film: (a) cross-sectional morphology in SEM observation; (b) surface morphology in SEM observation; (c) surface morphology in AFM observation.

Raman spectrum of the as-deposited GLC film is revealed in Fig. 2. Although there is still debate on the exact relationship between the atom vibration and Raman spectrum, Raman spectroscopy is one of the most widely and effectively used method to investigate the detailed bonding structure of carbonaceous materials.³² For nanocrystalline or amorphous carbon materials, the Raman spectrum typically includes a G peak centered around 1550 cm⁻¹ and a D peak centered at 1360 cm⁻¹.³³ Both G and D peaks are due to sp² carbon. The G mode has E_2g symmetry. This mode is due to the bond stretching of all pairs of sp² atoms in both rings and chains and does not require the presence of six-fold rings. The D peak is a breathing mode of A_1g symmetry involving phonons near the K zone boundary. This mode is forbidden in perfect graphite and only becomes active in the presence of disorder. The D mode is dispersive; it varies with the photon excitation energy, even when the G peak is not dispersive.^32–34 Based on the variations of the D and G bands, different amorphous carbon would possess unique features of the Raman spectrum, such as FWHM (full-width half-maximum), peak positions, and intensity ratio. Furthermore, the estimation of the sp³- and sp²-hybridized carbon bonds can be indirectly drawn from analyses of the Raman spectrum.³⁵ As shown in Fig. 2, the Raman spectrum of the as-deposited GLC film demonstrates a broad band approximately ranging from 890 to 1780 cm⁻¹. The D and G peaks could be positioned at about 1370 cm⁻¹ and 1526 cm⁻¹, respectively. Compared to other amorphous carbon films, the D band is relatively intense. The intensity ratio of the D and G peak (I_D/I_G) after peak fitting by Gaussian convolutional procedure was approximately 3.5. This value indicates a high content of sp² sites in the corresponding graphite-like carbon layer.^36,37

Fig. 2 Raman spectrum of the as-deposited GLC film.

3.2 Mechanical properties of the GLC film

The hardness and elastic modulus of the as-deposited GLC film is reveled in Fig. 3. To identify the valid hardness at 10% of the film thickness, the nano-indentation depth of about 200 nm was set first, which was much deeper than 10% of the film thickness. The hardness increased gradually as the indention depth increased to about 80 nm; subsequently, the hardness fluctuate around 16 GPa. In addition, the elastic modulus increased gradually as the indention depth was increased to approximately 40 nm, followed by fluctuating values ranging from 170 to 200 GPa. Owing to Fig. 1(a), the total thickness of the as-deposited GLC film was 1500 nm. Therefore, the valid values for the mechanical properties of the as-deposited GLC film should be 150 nm. The valid values for the hardness and elastic modulus were 16.8 and 172 GPa, respectively. These values were similar to previous studies and much higher than that reported by Huang et al.^24,36 The relative high hardness of the as-deposited GLC film with significant sp²-hybridized carbon might be due to the nano-crystallites/amorphous matrix structure, which was discussed deeply previously.²³ The load–displacement curve for the indention depth 150 nm is shown in Fig. 4 from which the plastic deformation of approximately 45 nm could be detected. Thus, the elastic recovery of 70% was calculated. Good elastic recovery suggested the high toughness of the as-deposited GLC film.

Fig. 3 Hardness and elastic modulus of the as-deposited GLC film.

Fig. 4 Load–displacement curve of the as-deposited GLC film.

3.3 Friction behaviors of the GLC film in different solutions

Fig. 5 shows typical friction curves of the as-deposited GLC film in different solutions. DW represents distilled water, and SW represents seawater. The meanings of S1, S2, S3 and S4 are stated in experimental section. All the friction coefficients in these solutions were below 0.1. Each friction curves appeared relatively smooth only with slight fluctuations, indicating a stable friction process. Low friction coefficients in combination with stable friction curves demonstrated good self-lubricating behavior in all these solutions, including not only distilled water, but also seawater or seawater constitute fluids.

Fig. 5 Friction curves of the GLC film in different solutions.

To search the small differences of the friction behaviors, the average friction coefficients of the as-deposited GLC film in these solutions were calculated, as shown in Fig. 6. Because of the small differences between the different friction coefficients, the vertical scale was set from 0.05 to 0.1. Based on Fig. 6, low friction coefficients ranged from 0.067 to 0.083 of the GLC in these six aqueous environments are observed clearly. The friction coefficient of the GLC film in seawater is slightly higher than that in distilled water. Different types of seawater play different roles in the friction coefficient. The friction coefficients of the as-deposited GLC film sliding against WC in S1, S3 and S4 were slightly lower than those in distilled water and seawater. In contrast, the GLC film exhibited higher friction coefficients in S2 than those in distilled water and seawater. Therefore, the high friction coefficient of the GLC film sliding against WC in seawater than that in distilled water arose mainly from the high friction coefficient in the seawater constitute, S2, which is the Na₂SO₄ solution.

Fig. 6 Friction coefficients of the GLC film in different solutions.

Moreover, according to the friction coefficients in Fig. 5 and 6, the GLC film not only exhibited low friction in distilled water, but also showed low friction in seawater or the seawater constitutes. The main reason to the good self-lubricating behavior of GLC film with a low friction in a water environment was attributed previously to the formation of a lubricating tribo-film at the friction contact interface in combination with the partial lubricating effect of the discontinuous water film.²³ The synthetic friction coefficient was described asμ = (1 − λ)μ_v + μ_s, where λ∈[0,1] is the ratio of solid–solid contact regions, μ_v is the water viscous friction coefficient, and μ_s is the solid friction coefficient.¹³ Owing to the rather limited differences for the viscosities of these water-based solutions in this study, the partial lubricating effects of the discontinuous fluid films to the water-lubricated micro-regions were similar. Therefore, the friction coefficients of the GLC film in these different solutions must be strongly dependent on the wear products, which would play different roles in the solid–solid contact regions on the micro-scale.

3.4 Wear behaviors of the GLC film in different solutions

The wear depths of the as-deposited GLC film in the different solutions were first gathered, as shown in Fig. 7. Each wear depth of the GLC film did not exceed 0.9 μm. Owing to Fig. 1, all the wear depths were less than the thickness of the carbonaceous layer (∼1324 nm), indicating that the GLC film survived in all these solutions. Comparatively, the average wear depth of GLC film in seawater (∼0.69 μm) was deeper than that in distilled water (∼0.39 μm). In specific, the average wear depth of the GLC film in seawater was the deepest among the values in these solutions. The averaged wear depths of the GLC films in S1, S2, S3 and S4 were approximately 5.2, 6.2, 5.3 and 4.8 μm, respectively. Although there were differences between the wear depths of GLC film in S1, S2, S3 and S4, each wear depth of the GLC film in these four solutions was deeper than that in distilled water. Different types of constitutes are believed to play different roles in the wear behavior of GLC film in seawater.

Fig. 7 Wear depths of the GLC film in different solutions.

In general, the low specific wear rate of GLC film in seawater suggests a good anti-wear protective effect on most of tribo-pairs serviced in seawater fluid for marine equipment. The calculated specific wear rates of the GLC film in the six solution environments are shown in Fig. 8. All the specific wear rates of the GLC film were in the magnitudes of 10⁻¹⁶ m³ N⁻¹ m⁻¹, meaning a low wear loss in all these solutions. In depth, the law of the specific wear rates was consisted with the law of the average wear depths. The wear rate of the GLC film in seawater (∼9.4 × 10⁻¹⁶ m³ N⁻¹ m⁻¹) was higher than that in distilled water (∼4.1 × 10⁻¹⁶ m³ N⁻¹ m⁻¹). In contrast, the wear rates of the GLC films in S1, S2, S3 and S4 were 6.9 × 10⁻¹⁶ m³ N⁻¹ m⁻¹, 8.0 × 10⁻¹⁶ m³ N⁻¹ m⁻¹, 6.6 × 10⁻¹⁶ m³ N⁻¹ m⁻¹ and 6.3 × 10⁻¹⁶ m³ N⁻¹ m⁻¹, respectively. The wear rates of the GLC films in these four solutions were higher than those in distilled water. The similar laws for averaged wear depths and specific wear rates suggest that the wear loss of GLC film was closely related to the wear depth. In addition, it appears that the wear of GLC film could be aggravated by each type of seawater constitute solution, resulting in a higher wear loss of the GLC film in seawater than that in distilled water. This is because the corrosion effect of seawater was much stronger than distilled water. Although the amorphous carbon matrix demonstrated good anti-corrosion performance and the as-deposited GLC film exhibited a low friction coefficient, the seawater molecules would have a harsher role than the distilled water molecules to the micro-cracks of wear surface.^38–40 That might be one of the reasons for the relatively severe wear damage of the GLC film in seawater than that in distilled water because the wear rate in each seawater constitute solution (S1, S2, S3 or S4) was higher than that in distilled water.

Fig. 8 Specific wear rates of the GLC film in different solutions.

For further analysis, the wear behaviors of the GLC film, Fig. 9 displays the wear tracks of GLC film in different solutions. Overall, there were no significant differences among the widths of these wear tracks. This might be because the different wear losses were highly dependent on the wear depths in these different solutions. Compared to Fig. 9(a) and (b), many more products were generated beside the wear track of the GLC film in seawater than in distilled water. In addition, plenty of wear products with different features could also be observed at the wear surfaces of the GLC film in S1, S2, S3 and S4. The wear products in S1 included the bulk and fine particles and the bulk particles might be aggregated by fine ones. However, wear products in S2 mainly included bulk particles, even though there were a few fine ones. The size of the wear products in S3 was moderate and the wear products in S4 appeared fine.

Fig. 9 Wear tracks of the GLC film in different solutions: (a) DW; (b) SW; (c) S1; (d) S2; (e) S3; (f) S4.

The compositions of the wear debris in different solutions were further identified by EDS, as shown in Fig. 10. The peak intensities were strongly affected by the collection times. The peak signals suggested the existence of the corresponding elements. The pattern in Fig. 10(a) demonstrated that the main composition of the wear debris in distilled water was WC or with some carbonaceous materials. Fig. 10(b) shows the complex composition of the wear debris in seawater due to the diversity of the seawater constitutes. The appearances of Na and Cl signals besides C, O and W signals in Fig. 10(c) confirmed the proposition of NaCl on the wear surface in S1. The signals in Fig. 10(d) suggested that the wear debris in S2 might include C, O, Na, W, S and Co. Co must come from the binding phase of the cement carbide of the WC matrix. The signals in Fig. 10(e) indicated Mg and Ca existed in the wear debris in S3. Almost no elements related to the constitutes of S4 could be detected, as observed in Fig. 10(f), which could be due to the rather limited content of the constitutes in the corresponding solutions. Some unmarked peaks were interference signals, such as Ar peak (at 2.9 eV) and Cr peak (at 5.4 eV). Ar must come from plasma bombardment during the deposition process. Cr could be assigned to the interlayer between the carbon layer and substrate.

Fig. 10 EDS patterns of the wear debris in different solutions: (a) DW; (b) SW; (c) S1; (d) S2; (e) S3; and (f) S4.

The wear scars of the coupled WC balls in different solutions are shown in Fig. 11. Compared to Fig. 11(a) and (b), more products were generated inside and outside wear scars of WC in seawater than that in distilled water. In specific, there were so much wear products that no clear edge of the wear scar of WC could be observed in seawater. Similar to the wear surfaces of the GLC film, abundant wear products with different features could be observed at the wear surfaces of WC in S1, S2, S3 and S4. Although the wear products on the wear scar in S1 included bulk and fine particles, the wear debris on the wear scars in S2, S3 and S4 were similar. The shape of products in S1 was consistent with crystal form of NaCl, which is the major constitute of S1. This suggests that the formation of bulk products at the wear surfaces in S1 were due mainly to the water evaporation. In addition, the transferred tribo-layers were observed clearly on wear scars in all the six water-based solutions.

Fig. 11 Wear scars of the WC balls in different solutions: (a) DW; (b) SW; (c) S1; (d) S2; (e) S3; and (f) S4.

Comparing the wear surfaces and wear scars, it could be found that the wear debris on the wear track of the GLC film was bulky and wear grooves on the wear scar of WC was rough. This suggests that the bulky wear debris was the key to the relatively higher friction in S2 or seawater, which generated severe abrasion to the friction contact surfaces. In Fig. 10, wear debris mainly contained C, W, O and some elements related to the solution constitutes. According to the references, corrosion or corrosion-wear of WC-based cement carbide with Co as the binder material was initiated from the dissolution of the Co binder phase, and the WC hard phase was almost inert in most of the aqueous environments, including seawater.^41,42 Therefore, the hard WC particles would generate wear debris between the friction contact surfaces as a result of the decreased toughness of the WC matrix due to the dissolution of Co phases. Because bulky debris, including hard WC particles, would generate higher friction resistance than fine wear debris to the friction contact faces, the higher friction coefficient of the GLC film in S2 than that in distilled water, S1, S3 or S4 can be understood. The severe abrasive wear could also be deduced from the coarse plough grooves on the wear surface of the WC ball in S2, as shown in Fig. 11(d). A potential reason for the abundant WC particles in S2 might be the fast dissolution of Co in a SO₄²⁺ solution. Although there were bulky products on the wear surface in S1, the main composition formed during water evaporation was NaCl. Thus, the friction behavior of the GLC film might be affected slightly. Another attractive phenomenon was the lower friction coefficient in S3 than in distilled water. S3 mainly contained divalent metal salts of CaCl₂, MgCl₂ and SrCl₂. In contrast to the other seawater constitutes, the divalent metal salts would generate lubricating wear products due to the chemical reactions shown as follows:⁴³

Mg²⁺ + 2H₂O⁻ → Mg(OH)₂ + 2H⁺ (1)

HCO₃²⁻ → CO²⁻ + 2H⁺ (2)

Ca²⁺ + CO₃²⁻ → CaCO₃ (3)

Wang et al. performed careful analyses and discussion of the tribo-chemical reaction of divalent metal salts in water and argued that both friction and wear would decrease when Mg²⁺ and Ca²⁺ exist in the aqueous medium. Therefore, it could be proposed that the bulky wear debris with abundant WC particles due to the Na₂SO₄ solution might increase the friction, while the divalent metal salts would decrease the friction for the GLC film sliding against WC in aqueous environments. Combined with the slight effects of S1 and S4, the synergistic effect resulted in a relatively higher friction coefficient of the GLC film in seawater than that in distilled water, even though the friction coefficients of GLC film in all the aqueous environments were rather low.

As for the wear rate, different seawater constitutes made different contributions. The greatest contribution to the wear loss of the GLC film in seawater was Na₂SO₄ among these different seawater constitutes. This might be closely related to the severe three-body abrasive wear regime due to the abundant hard WC particles inside the wear debris. The bulky wear debris with abundant hard particles would work as the rigorous third body for the friction contact faces. As shown in Fig. 9(d), the wear products that mainly contained WC hard particles on the wear surface of the GLC film in S2 was bulky, while the wear products on the wear surface of the GLC film in other solutions were fine. Although some rather bulky products consistent with NaCl were also detected (Fig. 9(c)), they were ascribed to the water evaporation. In addition, the soft NaCl particles would enhance limitedly the abrasion to the friction contact surfaces. This point is also evidenced in Fig. 11, which illustrates the roughest plough grooves on the wear scar of the WC ball in S2 than other solutions. Based on the friction and wear differences, it appeared that the tribological performance of the GLC film in seawater might be closely related to the nature and constitutes of seawater, as well as to the nature and performance of the coupled counterparts.

Finally, the low friction and wear loss might be related to the graphitization effect or tribo-films formed on the contact surfaces during the cycle sliding process. The Raman spectra of the transferred tribo-films on the wear scars of WC balls were acquired, which are shown in Fig. 12. The broad bands of the Raman spectra ranging from approximately 1000 to 1800 cm⁻¹ suggested that the main content of transferred tribo-film formed on the wear surface of the WC counterpart was a carbonaceous feature for each solution condition. The character of the amorphous matrix with deep graphitization for the carbonaceous tribo-film could be due to the intense D band and weak G band. According to calculation, the I_D/I_G ratio of the GLC film sliding against WC in DW, SW, S1, S2, S3 and S4 were about 1.61, 1.37, 1.64, 1.35, 1.41 and 1.53, respectively. In general, the higher I_D/I_G ratio, the lower friction coefficient and higher wear loss. For example, the I_D/I_G ratio of the GLC film in distilled water was higher than that in seawater. At the same time, the friction coefficient in distilled water was lower than that in seawater. Conversely, the wear loss in distilled water was lower than that in seawater, this might be related to the corrosion of seawater. The I_D/I_G ratio of GLC film sliding against WC in S1 was higher than in the other solutions, whereas the friction coefficient of S3 was lower than S1 in Fig. 6; this is because the divalent metal salts would generate lubricating wear products. This phenomenon suggests that the synthetic friction regime of the GLC film was not only worked in distilled water, but also worked in seawater. However, it was difficult to identify the relationship between the carbonaceous tribo-films and the friction differences according to the similar features of the Raman spectra acquired from these wear surface of the mating WC balls. The deep reason attributed to the wear debris, which was another type of solid product on the friction contact surfaces besides tribo-films.

Fig. 12 Raman spectra of the transferred tribo-films on the wear scars of the WC balls.

3.5 Tribological mechanism of GLC film in seawater

According to the abovementioned analysis, the tribological mechanism of the GLC film in seawater is illustrated in Fig. 13. The mixed friction regime due to the poor lubricating effects of the water-based fluid would divide the countered surfaces into solid–solid contacted regions and water-lubricating regions on the micro-scale in seawater. The solid-lubricating effect dependent on the graphite-like carbonaceous matrix in combination with the partial fluid-lubricating effect by the discontinuous seawater film were the main factors for the low friction and wear of the GLC film in seawater; moreover, the fine wear products due to the divalent metal salts of seawater could also decrease the friction coefficient and wear rate of the GLC film. However, the wear products with hard particles would result in a typical three-body wear model of the GLC film in seawater. As a result, the friction coefficient wear relatively higher than that in distilled water due to the enhanced abrasion. In addition, each type of seawater constitutes accelerated the wear of the GLC film in seawater, resulting in relatively more severe wear damage to the GLC film in seawater than in distilled water. The enhanced abrasion between two contacted faces of the GLC film and WC-based cement carbide with the Co binder phases by bulky and hard third body due to the Na₂SO₄ constitute made a great contribution to the increased wear of the GLC film in seawater.

Fig. 13 Tribological mechanism of the GLC film in seawater.

4. Conclusion

The GLC film exhibited good tribological performance with low friction and wear not only in distilled water, but also in seawater. The low friction and wear behaviors of the GLC film in seawater provide a novel approach to achieve long-life and stable operation of tribo-pairs services in seawater for marine equipment. The tribological performance of the GLC film in seawater was closely related to the nature and constitutes of seawater and the nature and performance of the counterparts. Abundant wear products with hard particles induced by the complex seawater constitutes between the friction and wear interfaces would result in a typical three-body wear model of the GLC film in seawater. The relatively higher friction coefficient of the GLC film against the WC counterpart in seawater was attributed mainly to the enhanced abrasion by the bulky wear products with hard particles due to the seawater constitute, Na₂SO₄, even though the divalent metal salts in seawater could decrease the friction resistance between the two contact surfaces. The relatively higher wear rate of the GLC film in seawater than that in distilled water was attributed to the accumulation effects of the different seawater constituents. The bulky wear debris with hard particles due to the Na₂SO₄ constituent of seawater had greatest effect on the increase in wear rate of the GLC film sliding against the WC counterpart in the seawater environment.

Acknowledgements

This study is supported by the National Natural Science Foundation of China (Grant no. 51202261 & 51475449) and the National Basic Research Program of China (973 Program) (Grant no. 2013CB632302). We would like to acknowledge them for the financial support.

References

1. J. Wang, F. Yan and Q. Xue, Wear, 2009, 267, 1634–1641.
2. S. L. Nie, G. H. Huang and Y. P. Li, Tribol. Int., 2006, 39, 1342–1354.
3. X. Wang and A. Yamaguchi, Tribol. Int., 2002, 35, 425–433.
4. H. Hirani and S. S. Goilkar, Wear, 2009, 266, 1141–1154.
5. R. P. C. Costa, F. P. Marciano, D. A. Lima-Oliveira, E. J. Corat and V. J. Trava, Surf. Sci., 2011, 605, 783–787.
6. R. J. K. Wood, Wear, 2006, 261, 1012–1023.
7. J. Z. Wang, F. Y. Yan and Q. J. Xue, Tribol. Lett., 2009, 35, 85–95.
8. L. Zhang, F. Wang, L. Qiang, K. Gao, B. Zhang and J. Zhang, RSC Adv., 2015, 5, 9635–9649.
9. M. Cui, J. Pu, J. Liang, L. Wang, G. Zhang and Q. Xue, RSC Adv., 2015, 5, 104829–104840.
10. S. Yang, D. Camino, A. H. S. Jones and D. G. Teer, Surf. Coat. Technol., 2000, 124, 110–116.
11. J. Stallard, D. Mercs, M. Jarratt, D. G. Teer and P. H. Shipway, Surf. Coat. Technol., 2004, 177–178, 545–551.
12. Y. X. Wang, L. P. Wang, J. L. Li, J. M. Chen and Q. J. Xue, Tribol. Int., 2013, 60, 147–155.
13. Y. X. Wang, L. P. Wang, S. C. Wang, G. Zhang, R. J. K. Wood and Q. J. Xue, Tribol. Lett., 2010, 40, 301–310.
14. Y. X. Wang, L. P. Wang and Q. J. Xue, Appl. Surf. Sci., 2011, 257, 4370–4376.
15. J. X. Huang, S. H. Wan, L. P. Wang and Q. J. Xue, Tribol. Lett., 2015, 57, 10–18.
16. J. X. Huang, L. P. Wang, B. Li, S. H. Wan and Q. J. Xue, ACS Appl. Mater. Interfaces, 2015, 7, 2772–2783.
17. Q. Ding, L. P. Wang, L. T. Hu, T. C. Hu and Y. Wang, Wear, 2012, 274–275, 43–49.
18. X. Y. Guan and L. P. Wang, Tribol. Lett., 2012, 47, 67–78.
19. X. Y. Guan, Z. B. Lu and L. P. Wang, Tribol. Lett., 2011, 44, 315–325.
20. Y. Wang, H. Li, L. Ji, F. Zhao, X. Liu, Q. Kong, Y. Wang, W. Quan, H. Zhou and J. Chen, J. Phys. D: Appl. Phys., 2010, 43, 505401–505407.
21. N. Fujisawa, M. V. Swain, N. L. James, J. C. Woodard, R. N. Tarrant and D. R. McKenzie, Tribol. Int., 2003, 36, 873–882.
22. S. K. Field, M. Jarratt and D. G. Teer, Tribol. Int., 2004, 37, 949–956.
23. Y. Wang, L. Wan, S. C. Wang, G. Zhang, R. J. K. Wood and Q. Xue, Tribol. Lett., 2010, 40, 301–310.
24. Y. Wang, L. Wang and Q. Xue, Surf. Coat. Technol., 2011, 205, 2770–2777.
25. Y. Wang, J. Pu, J. Wang, J. Li, J. Chen and Q. Xue, Appl. Surf. Sci., 2014, 311, 816–824.
26. Q. Wang, F. Zhou, X. Ding, Z. Zhou, C. Wang, W. Zhang, L. K. Li and S. T. Lee, Tribol. Int., 2013, 67, 104–115.
27. Q. Wang, F. Zhou, X. Ding, Z. Zhou, C. Wang, W. Zhang, L. K. Li and S. T. Lee, Wear, 2013, 303, 354–360.
28. G. Capote, L. F. Bonetti, L. V. Santos, V. J. Trava-Airoldi and E. J. Corat, Thin Solid Films, 2008, 516, 4011–4017.
29. Y. S. Li, Y. Tang, Q. Yang, C. Xiao and A. Hirose, Appl. Surf. Sci., 2010, 256, 7653–7657.
30. J. Wang, F. Yan and Q. Xue, Chin. Sci. Bull., 2009, 54, 4541–4548.
31. J. Qi, L. Wang, Y. Wang, J. Pu, F. Yan and Q. Xue, Wear, 2011, 271, 899–910.
32. P. K. Chu and L. Li, Mater. Chem. Phys., 2006, 96, 253–277.
33. A. C. Ferrari and J. Robertson, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61, 14096–14107.
34. A. C. Ferrari and J. Robertson, Phys. Rev. B: Condens. Matter Mater. Phys., 2004, 64, 075414.
35. J. Robertson, Mater. Sci. Eng., R, 2002, 37, 129–281.
36. M. Huang, X. Zhang, P. Ke and A. Wang, Appl. Surf. Sci., 2013, 283, 321–326.
37. G. A. Viana and F. C. Marques, Vacuum, 2015, 112, 17–24.
38. W. Q. Bai, X. L. Wang, C. D. Gu and J. P. Tu, Thin Solid Films, 2015, 584, 214–221.
39. N. W. Khun and E. Liu, Thin Solid Films, 2009, 517, 4762–4766.
40. A. M. M. Santos, R. J. C. Batista, L. A. M. Martins, M. Ilha, M. Q. Vieira, D. R. Miquita, F. C. R. Guma, I. L. Müller and T. M. Manhabosco, Corros. Sci., 2014, 82, 297–303.
41. S. Hochstrasser, Y. Mueller, C. Latkoczy, S. Virtanen and P. Schmutz, Corros. Sci., 2007, 49, 2002–2020.
42. A. K. Basak, J. P. Celis, P. Ponthiaux, F. Wenger, M. Vardavoulias and P. Matteazzi, Mater. Sci. Eng., A, 2012, 558, 377–385.
43. J. Wang, F. Yan and Q. Xue, Wear, 2009, 267, 1634–1641.

---