Fabrication of Cu-Doped Diamond-like Carbon Film for Improving Sealing Performance of Hydraulic Cylinder of Shearers

Abstract

During shearer operation, the piston rod is susceptible to wear from the invasion of pollutants, thus ruining the sealing ring in the hydraulic cylinder. This work attempts to conduct a systematic investigation of Cu-doped diamond-like carbon (Cu-DLC) film to improve the seal performance. The failure process of the cylinder was analyzed, and relevant parameters were determined. Several Cu-DLC films were deposited on the substrate of the piston rod in a multi-ion beam-assisted system, and their structures and combined tribological performances were investigated. The hardness of the film ranges from 27.6 GPa to 14.8 GPa, and the internal stress ranges from 3500 MPa to 1750 MPa. The steady-state frictional coefficient of the film ranges from 0.04 to 0.15; the wear rate decreases first and then increases, and it reaches its lowest (5.0 × 10⁻⁹ mm³/N·m) at 9.2 at.% content. a:C-Cu_9.2% film presents optimal combined tribological performances in this experiment. The modification mechanism of Cu-DLC film for the seal performance may come from the synergistic effects of (i) the contact force and friction-heat-induced film graphitization, (ii) Cu doping improves the toughness of the film and acts as a solid lubricant, and (iii) the transfer layer plays a role in self-lubrication.

1. Introduction

Coveted clean power, such as photovoltaic and wind power, remains up in the air, and the shortage of electricity in hot summers has prompted attention to the reliable mining of coal for thermal power [1]. A hydraulic cylinder is a pivotal height control device, and its sealing performance directly affects the operational stability and efficiency of the shearer [2]. Invasion of extrinsic pollutants during shearer operation tends to induce wear of vulnerable friction pair of hydraulic cylinders, and the induced seal failure may endanger both the shearer operation and operator safety [3]. Film modification is a preferred method to enhance the seal performance of vulnerable components [4]. Carbon-based materials like diamond-like carbon (DLC) films are attractive in many tribological applications. Some progress has made in creating carbon-based films using tribocatalysis for a long duration [5]. Sang T.P. et al. did some good work on using catalyst nanoparticles to create DLC films in situ [6]. Tribo-induced catalytically active oxide surfaces enable the formation of durable and high-performance carbon-based tribofilms [7]. In situ engineered catalytically active surfaces are constructed for tribocatalysis with layered double hydroxide nanoparticles [8]. The strategy of catalyst nanoparticles still cannot deal with the seal issue of vulnerable friction pair of shearers due to rough and complex service conditions. Herein, we propose Cu-doped DLC (Cu-DLC) to overcome the drawback.

Unlike carbide metals, Cu does not react with carbon and can directly embed into DLC matrix as ductile metal particles to improve the toughness and then mechanical and tribological performances of DLC film [9,10]. Carbide-containing metals of diamond-like carbon (DLC) film like W- or Cr-doped ones are commonly used but are subject to early failure due to poor toughness [11,12]. As a non-carbide-forming metal, Cu may bring hope to replace its carbide-forming counterparts due to the low solubility of carbon in Cu, low cost, and availability [13]. Some researchers have conducted studies on the effects of Cu doping on the mechanical or tribological performances of DLC films [9,10,13,14,15]. Sushil Kumar et al. investigated the mechanical performances of Cu-DLC film produced by PECVD (plasma-enhanced chemical vapor deposition) and indicated a reduction in stress and an improvement in the transport performance of DLC film [9]. Nanostructured amorphous carbon–copper composite films were also formed by PECVD, and it was found that the nanohardness values of the films varied in the range of 0.2–3 GPa [13]. Also, the effects of Cu doping on tribological performances of Cu-DLC or Cu/DLC films via magnetron sputtering were investigated to improve the graphitization of wear debris on counterparts [10], corrosion resistance on mild steel [14], and surface performance of electrical contact materials [15]. In contrast, the prepared film in this work has to meet the requirements of good combined tribological performances, including both high hardness and a low frictional coefficient and wear rate for the seal issue in rigid service environments.

The service conditions of shearers are rather complex and rigid and are involved in interactions of working loads and media like gravel [16]. Systematic research is rare in the literature to date on the fabrication of Cu-DLC film to improve the sealability of hydraulic cylinders for reliable shearer operation. Given the seal issue under service conditions, a systematic investigation is expected to have three progressive links: (1) Failure analysis and film solution: determining the component to be worn and clarifying performance requirements of the film under service conditions. (2) Film fabrication and verification: fabrication of Cu-DLC film and verification of the film requirements. (3) A film modification mechanism to reveal the improvement mechanism of Cu-DLC film applicable to the seal issue.

The piston rod is vulnerable to wear due to the intrusion of pollutants during shearer operation, and thus, it may ruin the sealing ring in the hydraulic cylinder. In this work, we endeavor to fabricate Cu-DLC film to improve the sealing performance of the hydraulic cylinder of a shearer under rigid service conditions. Failure of the hydraulic cylinder is analyzed. Cu-DLC film with different Cu contents is deposited on the piston rod substrate with a multi-ion beam-assisted system. The structure of the film is characterized, and the combined tribological performances are evaluated. This allows us to propose a modification mechanism for Cu-DLC film.

2. Performance Requirements and Preparation of Thin Films

2.1. Performance Requirements

2.1.1. Analysis of the Structure and Working Principle of the Hydraulic Cylinder

Figure 1 shows the structure diagram of a hydraulic cylinder. The cylinder plays an important role in raising and lowering the cutting transmission part of the shearer drum in the fore and aft motion. As shown in Figure 1, the cylinder is composed of guide ring 1, seal ring 2, piston rod 3, rod chamber 4, cylinder block 5, piston 6, and head port 7. Piston 6 is driven by hydraulic oil, and piston rod 3 and sealing ring 2 rub against each other at the sealing component. The cylinder withstands load pressure by controlling the flow of hydraulic oil, and the flow speed and rate of the fluid flow determine the reciprocating frequency and the speed of piston rod 3 and cylinder block 5. In the case of drum raising, high-pressure hydraulic oil in head port 7 pushes the piston rod out. In the event of drum lowering, the high-pressure hydraulic oil in rod chamber 4 pushes the piston rod back.

2.1.2. Cylinder Failure Analysis

The process of hydraulic cylinder sealing failure during shearer operation can be divided into three stages. Stage I: The sealing ring is an important component in preventing the invasion of extrinsic pollutants by utilizing its elastic deformation. The retractable deck can prevent large-sized pollutants from entering the cylinder. However, there is a possibility that these pollutants may still enter the cylinder or mix with the hydraulic oil due to the movement of the piston rod and stay in gaps between the piston rod and sealing ring, which can cause issues with the seal performance. Stage II: The sealing ring expands when hydraulic oil enters the cylinder, increasing the sealing area at both ends of the piston and resulting in an increase in friction force. Solid pollutants have a higher hardness than the rubber sealing ring and are likely to be embedded in the cylinder under high internal pressure, leading to wear. Stage III: The pollutants embedded in the sealing ring serve as an abrading agent, repeatedly grinding against the piston rod during its reciprocating movement. This induces a wear scar formed on the piston rod. The abrasion worsens over time and leads to seal failure. In this way, the oil leakage occurs in the cylinder and eventually leads to the failure of the cylinder, as well as a shearer accident.

2.1.3. Film Performance Parameters

Accordingly, the fabrication of Cu-DLC film on the piston substrate is recommended for improving the seal performance. The film should have good combined tribological performance, i.e., high hardness, low internal stress, and low friction coefficient. The high hardness, over 7 GPa of solid pollutants, is needed to avoid the formation of wear scars on the piston. The low internal stress may enable the film to have good toughness and help resist early film failure. The low friction coefficient, which is less than 0.2, may enable the film to provide lubrication and resist the attachment of pollutants on the piston.

2.2. Film Preparation

The substrate consisted of two wafers. Si wafer was used for structural and mechanical analyses; 40CrNiMoA (raw material of piston rod, AISI4340) (Beijing Chemical Workstation, Beijing, China) was used for tribological analysis. Pretreatment was conducted for the substrate and included two steps: (1) Immersed in an ultrasonic bath with acetone and ethanol for 20 min. (2) Dried with a stream of nitrogen prior to deposition.

Cu-DLC film was deposited on the substrate using a multi-ion beam-assisted deposition (IBAD) system (SP9060, PowerTech, Beijing, China) [17]. The deposition system consisted of four Kaufman ion sources to give different ion energies. The four ion sources were: (i) a high-energy source for ion implantation before sputtering, (ii) two middle-energy sources for sputtering targets of copper (Cu) and graphite (C), respectively, and (iii) a low-energy source for ion bombardment during deposition. The vacuum chamber was evacuated to 2.0 × 10⁻⁴ Pa, and deposition pressure was set to 2.5 × 10⁻² Pa. Prior to deposition, Ar⁺ in an ion source was used with a voltage of 10 kV and current of 20 mA to remove possible surface oxides from the substrate. A 0.2 μm thick layer of Cu was deposited on the substrate with a Cu target at 1100 eV and 40 mA. 0.8 μm thick layer (a:C-Cu_x%) was then deposited with co-sputtering of C and Cu targets under conditions of (i) C target operated at 1200 eV and 50 mA, (ii) with constant energy of 900 eV. Cu sputtering current was, respectively, set to 0, 18, 35, 50, 70, and 90 mA for adjustment of Cu contents in the film. Six samples of Cu-DLC films (approximately 1 μm thick) were prepared, whose Cu sputtering current and atomic content (at.%) are listed in

Table 1. Atomic percentage content and sputtering current of Cu for six samples of a:C-Cu_x%.

Sample	Film	Cu Sputtering Current (mA)	Cu (at.%)
C0	a:C-Cu0%	0	0
C1	a:C-Cu6.8%	18	6.8
C2	a:C-Cu9.2%	35	9.2
C3	a:C-Cu15.4%	50	15.4
C4	a:C-Cu23.7%	70	23.7
C5	a:C-Cu32.6%	90	32.6

.

2.3. Characterization

The morphology and elemental composition were investigated using scanning electron microscope (SEM, JSM-6301F, JEOL, Tokyo, Japan) with an energy-dispersive spectroscopic (EDS) detector. Analysis of EDS spectra for quantification of elemental composition was performed with INCA software (Version 7.3). The phase structure was evaluated by X-ray diffraction (XRD, Model XD-3, Rigaku, Tokyo, Japan). The bonding structure was analyzed using Raman spectroscopy (LabRAM HR Evolution, HORIBA, Tokyo, Japan); Gaussian curves were fitted using commercially available OriginLab software (Origin, Version 2021, OriginLab corporation, Northampton, MA, USA) to identify peak positions, and integral peak areas were derived to evaluate intensity ratio I_G/I_D. The elements and valence changes contained were analyzed by X-ray photoelectron spectroscopy (XPS, Escalab 250Xi, ThermoFisher, Waltham, MA, USA). The film thickness and relevant morphologies were observed by 3D white-light interference profiler (Nano Map-D, Aep, Columbus, OH, USA).

As for the mechanical performance of the film, hardness was estimated using a nano-indenter (MTSXP, MTS, Eden Prairie, MN, USA) with continuous rigidity measurement test mode under conditions of maximum indentation load of 2 mN and dwell time of 5 s. Each measurement was conducted five times on different positions on the film, and the average value was used as the measurement value. The intrinsic stress in the film was analyzed by measuring the thicknesses and curvature radius of the Si wafer (φ 150 mm); the individual values of the intrinsic stress were then indirectly obtained by processing the relevant measurements according to the Stoney Equation below:

$$\sigma ={\displaystyle \frac{1}{6R}}{\displaystyle \frac{E}{(1-v)}}{\displaystyle \frac{d_0^2}{d}},$$

(1)

where E, v, d₀ are Young’s modulus, Poisson ratio, and thickness of Si wafer, respectively, and d is the film thickness. Radius R of curvature of the reference wafer before deposition was measured and taken as R₁. After deposition, the R was taken as the relative wafer radius of curvature and was calculated by R = R₁ R₁/(R₁ − R₁), in which R₁ and R₂ are the radii of curvatures before and after depositions, respectively.

Tribological tests were performed using reciprocating friction and wear tester (MFT-4000, Huahui, Lanzhou, China) under condition of (i) the coated sample was fixed, and a plastic ring was used as a counterpart; (ii) the sample surface was conducted a reciprocating movement under a load of 5 N; (iii) the amplitude, the reciprocating frequency, and the duration was 30 mm, 10 Hz, and 15–60 min, respectively; (iv) the wear scars were observed using an optical microscope (OM, BX51M, OLYMPUS, Tokyo, Japan) with an additional digicam. The values of wear rate were calculated by measuring the worn cross-sectional area on the film at 20 equally spaced positions along the worn trace.

To ensure credibility, the values measured for each sample were averaged over six measurements and were accurate within ±5% margin of error.

3. Results and Discussions

3.1. Structures and Mechanical Properties of Cu-DLC Film

3.1.1. Structures of Cu-DLC Film

Figure 2 shows XRD spectra of Cu-DLC film with six Cu contents. Without Cu doping, only an amorphous carbon peak can be seen for the pure DLC film (a:C-Cu_0%). With Cu doping, Cu-DLC film exhibits two characteristic peaks of face-centered cubic polycrystalline Cu, the crystallographic peak of Cu (111) around 2θ = 43.3° while the second peak corresponds to Cu (200) near 2θ = 50.4° [18]. With an increase in Cu doping content, the intensity of Cu diffraction peaks increases. Cu doping may induce a competitive growth between Cu grains at Cu (111) and Cu (200) peaks [18]. The degree of growth intensity of Cu (111) is always higher than that of Cu (200) and suggests that doped Cu grains may have preferred orientation at the crystallographic plane (111) [19]. On the other hand, the diffraction peak of Cu (111) widens with a decrease in Cu content, which may be attributed to the presence of Cu grains with smaller sizes [20]. In this way, the microscopic structure of Cu-DLC film is composed of a compound of polycrystalline copper and amorphous carbon. The size of Cu crystals may vary with the doping content of Cu crystals, which in turn can affect the film’s performance.

3.1.2. Hardness of Cu-DLC Film

Figure 3 shows the hardness values of Cu-DLC film with six Cu contents. As shown in Figure 3, the hardness value ranges from 14.8 GPa to 27.6 GPa and decreases as the doping content of Cu increases. The film hardness is much higher than that of the piston substrate (7 GPa) and is higher than the maximum working stress of the cylinder (10 GPa). Such a film provides the piston with superior abrasion resistance to the pollutants.

The hardness of a pure DLC film without doping is 27.6 GPa. The hardness of Cu-DLC film decreases to 24.6 GPa when Cu content is 9.2 at.%, but the decrease in hardness is not significant. The hardness of Cu-DLC film continues to drop to its minimum of 14.8 GPa when doped Cu content reaches its maximum content of 32.6 at.%. This means that adjusting Cu content can change the film’s hardness.

Figure 4 shows a stress variation curve for Cu-DLC film. The internal stress of Cu-DLC film is lower than that of pure DLC film without Cu doping. This suggests that the internal stress can be adjusted by introducing copper doping. The internal stress decreases rapidly from 3500 MPa to 1750 MPa when Cu content increases from 0 at.% to 9.2 at.%. When Cu content continues to increase, the internal stress decreases slowly. When the content of Cu is 32.6 at.%, the internal stress decreases to a minimum of 1250 MPa. The presence of soft and flexible Cu crystallites embedded in the amorphous carbon matrix may release strains and reduce film stress [21]. The reduction in internal stress may prevent the film from peeling during the movement of the components [22,23].

The mechanical results show that Cu-DLC film meets the hardness requirements of actual working conditions. The hardness value of the film is in the range of 14.8–27.6 GPa, and it drops with a rise in the doping content of Cu. The film hardness is higher than the maximum working stress (10 GPa) of the cylinder and is much higher than that of the piston substrate (7 GPa). Additionally, an appropriate doping content can result in a film with both high hardness and low internal stress.

3.2. Tribological Performance of Cu-DLC Film

3.2.1. Friction Coefficient and Wear Rate

The invasion of extrinsic pollutants leads to a deterioration in the seal performance of the cylinder. Typically, performance deterioration is associated with the seal ring and piston rod. During the reciprocating motion, pollutants attached to the piston rod can transfer to the seal ring. The abrasive wear between the seal ring and piston rod can potentially cause seal invalidation.

Friction curves are shown in Figure S1 for Cu-DLC film with six Cu contents. The pure DLC (a:C-Cu_0%) film has a higher coefficient of friction (COF) than Cu-DLC film, suggesting that Cu doping can improve the tribological performance of Cu-DLC film. The steady-state COF of Cu-DLC film after 45 min of reciprocal sliding is shown in Figure 5. COF of Cu-DLC film fluctuates between 0.04 and 0.15 with different Cu contents, and it is lower than that of undoped DLC (0.18). With an increase in Cu content, COF value initially decreases from the maximum value of 0.18 for a:C-Cu_0% to a value of 0.09 for a:C-Cu_6.8%. It then reaches a minimum value of 0.04 when the Cu content is increased to 9.2 at.% for a:C-Cu_9.2%, followed by a gradual rise to 0.07 for a:C-Cu_15.4%. As Cu content is increased to 23.7 at.%, the value of COF reaches 0.1, and then it rises quickly to 0.15 for the a:C-Cu_32.6%. The friction coefficient is involved with Cu doping and the surface roughness of the film, and a smooth surface may result in a low COF value [16,24]. Among them, the COF of Cu-DLC film with Cu content of 9.2 at.% (a:C-Cu_9.2%) is the lowest, as low as 0.04. In this experiment, all friction coefficients of Cu-DLC film are much lower than 0.2, indicating that the film can play a good lubricating role.

Figure 6 shows a curve of the wear rate for Cu-DLC film. The wear rate initially decreases and then increases with the increase in Cu content, reaching its lowest point (5.0 × 10⁻⁹ mm³/N·m) at a content of 9.2 at.%. As a soft and ductile metal, Cu grains embedded in the amorphous carbon network of DLC can reduce the brittleness of DLC film. Doping Cu may also provide a buffer space, allowing stress concentration in the carbon matrix, thereby improving its tribological performances [25]. In this way, doping Cu may improve the wear resistance of DLC film by means of improving toughness, reducing stress concentration, and delaying early crack initiation.

Suitable Cu doping content may range from 5 at.% to 15 at.% and facilitate low COF (Figure 5) and wear rate (Figure 6), as well as good mechanical properties (Figure 3 and Figure 4). A comparison of the results (Figure 3, Figure 4, Figure 5 and Figure 6) exhibits that Cu-DLC film with 9.2 at.% Cu has optimal mechanical and tribological performances. a:C-Cu_9.2% film shares the least wear rate at approximately 5.0 × 10⁻⁹ mm³/N·m (Figure 6), the lowest COF at 0.04 (Figure 5), a hardness of 25.6 GPa (just lower than the lowest counterpart with 6.8 at.% Cu), and an intrinsic stress of 1750 MPa.

Differing from doping metals like Cr and W, which tend to bond strongly with amorphous carbon matrix, the presence of nanosized Cu in the nanocomposite films has a tendency to make the carbon inert and abruptly reduce the interatomic forces between crystallite carbon. The formation of such a moderated bond between the nanocrystallite and the matrix may facilitate grain–matrix interface sliding and increase the ductility of the film [26]. Subsequently, it facilitates the mutual multiple shifts of nanocrystallites when external stress is applied. Such shifts result in deformation either at grain boundaries or at Cu-DLC interfaces [10]. As a result, diffusion of suitable content and size of nanosized Cu crystallites into a-C matrix can improve the combined tribological performance of Cu-DLC film.

Figure 7 shows optical microscope images of wear scars for a:C-Cu_9.2% film at different sliding times. It can be seen from the comparison between Figure 7a and b that the wear marks and debris of a: C-Cu_9.2% film become obvious with an increase in sliding time. On the edges of the wear marks, there is visible debris accumulation, along with the formation of a transfer layer. Figure 7c shows that the wear of the film increases, and traces of furrows become visible. Abrasive particle accumulation can be observed at the edge, and the transfer layer is noticeable. The wear morphology of the corresponding part in Figure 7c is shown in Figure 7d. Since the hardness of the counterpart is lower than that of Cu-DLC film, its wear rate is higher than that of the film.

Figure 8 shows SEM and local EDS mapping images of wear scar for a:C-Cu_9.2% film. The SEM image of the wear scar in Figure 8a shows that the Cu-DLC film grinded against the ring forms a transfer layer in addition to debris. Figure 8b is a local enlargement of Figure 8a, and its EDS mapping images are shown in Figure 8c–f. EDS results of Figure 8c–f show that the transfer layer contains elements of C, Cu, Cr, and Fe. The oxygen is not observed on the wear tracks. The presence of oxygen may be involved with the base pressure of the vacuum chamber during deposition [27]. A high base pressure like 0.1 Pa may induce the presence of oxygen, and a low base pressure like 2.0 × 10⁻⁴ Pa in this experiment may induce the absence of oxygen [9]. Carbon and copper elements come from the Cu-DLC film, while chromium and iron elements come from the steel substrate. It is concluded that the transfer layer containing copper is the main reason for the decrease in the friction coefficient of the film. Among them, a:C-Cu_9.2% film has the best tribological performances.

3.2.2. Raman and XPS Analyses of Cu-DLC Film

Figure 9 compares Raman spectra of undoped DLC (a:C-Cu_0%) film and a:C-Cu_9.2% film under a load of 5 N. In Figure 9, four Raman spectra include one spectrum of undoped DLC (Figure 9a) and three spectra of a:C-Cu_9.2% film, corresponding to deposited film (no wear) (Figure 9b), the wear track on worn film (Figure 9c) and the wear debris (Figure 9d).

In Figure 9a, Raman spectra of undoped DLC films show two D and G characteristic peaks of 1395 cm⁻¹ and 1500 cm⁻¹, respectively. A comparison between Figure 9a,b indicates that peak G shifts from 1500 cm⁻¹ (undoped DLC film) to 1550 cm⁻¹ (no wear film) and peak D shifts from 1395 cm⁻¹ (undoped DLC film) to 1380 cm⁻¹ (no wear film). The shift of G-peak to a higher wave number and D-peak to a lower wave number means an increase in sp²-C content and a decrease in sp³-C bond content [28]. I_G/I_D ratio increases from 1.3 for undoped DLC film to 1.4 for no-wear Cu-DLC film, and sp³-C content decreases from 41.7% for undoped DLC film to 35.8% for no-wear film. Such a change in carbon bond structure may affect the performances of Cu-DLC films. The reason for this phenomenon is that the heat generated by reciprocal friction between the film and the corresponding part increases the temperature of the contact area [29,30]. After conducting the friction and wear test, the G-peak position deviates from 1550 cm⁻¹ (no wear film) to 1585 cm⁻¹ (debris) and 1590 cm⁻¹ (wear mark). I_G/I_D ratio increases from 1.4 (no wear film) to 1.9 (wear mark) and 2.3 (wear debris). This reflects the change in the distribution and content of sp²-C and sp³-C bonds [31].

During the experiment, a portion of sp³-C bond in Cu-DLC film is transformed into sp²-C bond and moves to the film surface under the synergistic effect of friction heat and contact force. The carbon concentration decreases, which further leads to the conversion of sp³-C bond to sp²-C bond. On the other hand, Cu particles can alter the angle of the C-C covalent bond and the distribution of carbon atoms [32]. In the reciprocating motion, the Cu formed by noncarbides deforms plastically and diffuses to the high-temperature area, where it is uniformly distributed in the film. Such Cu can fill more d orbitals and convert the sp³-C bond into a sp²-C hybrid bond [33].

XPS is conducted to obtain further structural information for a:C-Cu_9.2% film. Two different core level spectra for C 1s and Cu 2p are shown in Figure 10a,b, respectively. C 1s spectrum shows a sharp peak at 283.8 eV. Compared with the standard result (where the C 1s peak is obtained at 284.6 eV) [34], a nominal shifting of the C 1s peak (~0.8 eV) towards the lower binding energy side can be observed. This shift may be due to Cu particles because its incorporation can enhance the graphite-like sp²-C bonding [9]. Cu 2p spectrum is also presented in Figure 10b in the binding energy range of 925 to 965 eV and illuminates the incorporation of Cu particles in the Cu-DLC film. Moreover, two peaks of pure Cu can be observed at 931.8 eV and 951.4 eV, and the peak position difference of 19.6 eV is specific to metal Cu [35].

As a result, doping the appropriate amount of Cu into DLC film can enhance its toughness and effectively improve the tribological performances of piston rods. In this experiment, a:C-Cu_9.2% film can significantly improve the tribological performances of the piston rod. It thus provides the basis for revealing the improving mechanism of Cu-DLC film for the sealing performance of hydraulic cylinders.

3.3. Improvement Mechanism of Cu-DLC Film for the Seal Performance

Figure 11 schematically presents the improvement mechanism of Cu-DLC film for sealing the performance of hydraulic cylinders. Figure 11a shows a schematic drawing of the sealing ring and piston rod deposited with Cu-DLC film. Figure 11b shows a section view of the friction portion between the sealing ring and the piston rod. Double arrows and black blocks indicate the reciprocating motion of the piston rod and solid particles embedded in the sealing ring, respectively. Figure 11c shows a microscopic diagram of the contact surfaces for the friction portion in Figure 11b and contains the solid particles, transfer layer, graphite layer (black ellipse), Cu-DLC film layer, and AISI4340 substrate, from top to bottom. Figure 11d further illustrates the schema of the transfer layer shown in Figure 11c, and the composition of the transfer layer includes copper atoms (copper-colored balls), sp³-C (blue balls), sp²-C (golden balls), iron atoms (grey balls), and chromium atoms (purple balls), from top to bottom.

Given the experimental data above, three possible factors affecting the sealing performance of Cu-DLC film are (1) the appearance of graphitization; (2) the doping of Cu; and (3) the generation of the transfer layer. This allows us to suppose that the improvement mechanism of Cu-DLC film may be a synergistic effect of the following three mechanisms.

• Contact force and friction heat induce graphitization of the film surface.

In Figure 11c, the accumulation of friction heat may result in high temperatures in localized micro-contact areas. Under dynamic action, the metastable sp³-C bond transforms into sp²-C bond with high thermal stability at high temperatures (about 110 °C), which is consistent with Raman results shown in Figure 9. Under the influence of contact force and thermal diffusion, the sp²-C diffuses onto the surface of Cu-DLC film and forms aggregates, resulting in the formation of a top graphitization layer (black ellipse) on the film. As shown in Figure 11c, it can be observed that sp²-C aggregates in the contact area, forming the top graphitization layer. The top graphitization layer can play the role of solid lubrication for the friction pair of the piston rod–seal ring, thereby reducing the friction coefficient.

2.

Cu doping improves the film toughness and acts as a solid lubricant.

From Figure 11c,d, the action of internal stress resulted in the plastic deformation of Cu in the film, and the strain energy is released. This can avoid the initiation and propagation of early microcracks, thereby improving the toughness of the film. Under the joint action of contact force and friction heat, plastic deformation occurs in Cu, and some Cu grains diffuse and aggregate to the surface of the film. This part of Cu grains with low shear stress acts as a solid lubricant, absorbing the deformation energy in the contact zone and reducing the friction coefficient. As a result, the sealing performance of Cu-DLC film can be improved.

3.

The transfer layer plays a role of self-lubrication and long duration.

According to Figure 11c, the transfer layer is formed between the film and solid content under the action of friction heat and contact force. This layer can avoid direct contact between two friction surfaces, which has a self-lubricating and long-term protection effect. From Figure 9, it can be seen that due to the accumulation of friction heat, the sp³-C bond is transformed into sp²-C bond, which diffuses on the surface of the film and forms graphitization. Graphite is produced and accumulated between two contact surfaces and forms a transfer layer with Cu, Fe, and Cr. The constitution of the transfer layer prevents immediate contact between solid content and the film. When the force is applied, solid particles are squeezed into the transfer layer. Local stress concentration causes the plastic deformation of Cu, and local strain energy can be released. So, the self-lubricating performance of the transfer layer makes the coated piston rod possess a significantly lower wear rate (5 × 10⁻⁹ mm³/N·m, Figure 6) and a predictable long service time.

In summary, the transfer layer, soft Cu, and graphitized top layer play the role of solid lubricant for Cu-DLC film, improving the wear resistance of the hydraulic cylinder piston rod and the sealing performance of the sealing ring, thus extending the service life of the friction ring.

4. Conclusions

Sealing failure of hydraulic cylinders results from the wear of the piston rod. Cu-DLC film is fabricated to improve the sealing performance of the cylinder. The structure, harness, and internal stress of Cu-DLC film with different Cu contents are investigated, as well as the wear rate and frictional coefficient, so as to find the Cu-DLC film with optimum Cu content. A modification mechanism of Cu-DLC film is also proposed. The main conclusions of this work are as follows:

• Dust particles attached to the piston rod enter the cylinder, and the piston rod tilts to rub the sealing ring under the radial force, resulting in the failure of the piston rod. The particle hardness is about 7 GPa, and the highest contact force of the piston rod is 10 GPa. Cu-DLC film with Cu contents is then deposited on 40CrNiMoA in a multi-ion beam-assisted system.
• The degree of growth intensity of Cu (111) is always higher than that of Cu (200), and doped Cu grains may have a preferred orientation on the crystallographic plane (111). With an increase in Cu content, the hardness of the films decreases in a range from 14.8 GPa to 27.6 GPa, and the internal stress drops from 3500 MPa to 1750 MPa to avoid the film spalling.
• The friction coefficient fluctuates between 0.04 and 0.15, and the wear rate ranges from 4.7 × 10⁻⁹ mm³/N·m to 7.5 × 10⁻⁹ mm³/N·m. a:C-Cu_9.2% film has both the lowest frictional coefficient (0.04) and wear rate (5.0 × 10⁻⁹ mm³/N·m). Doping of Cu in DLC film results in G-peak shifts from 1500 cm⁻¹ (a:C-Cu_0%) to 1550 cm⁻¹ (a:C-Cu_9.2%) and D-peak shifts from 1395 cm⁻¹ (a:C-Cu_0%) to 1380 cm⁻¹ (a:C-Cu_9.2%). As for a:C-Cu_9.2% film, G-peak position deviates from 1550 cm⁻¹ (no wear film) to 1585 cm⁻¹ (debris) and 1590 cm⁻¹ (wear track). An increase in I_G/I_D ratio from 1.4 to 2.3 suggests the transition of sp³-C to sp²-C bond, and the transition may lower the friction coefficient and the wear rate.
• Improving the mechanism of Cu-DLC film for the seal performance may be a synergistic effect of (i) contact force and friction heat inducing the graphitization of the film surface; (ii) Cu doping improving the film toughness and acting as a solid lubricant; and (iii) the transfer layer playing a role of self-lubrication and offering a long duration.