Introduction

In periodic table, carbon has been regarded as one of the most important elements. It has the maximum number of allotropes, and more than 90% of chemicals contain carbon. Carbon-derived solid materials show unique properties such as super high hardness and thermal conductivity like in diamond or excellent lubricity and softness like in graphite. In last several years, carbon has also been used in the synthesis of diamond, diamond-like carbon (DLC) and carbon nitride coatings which have widely been used in engineering applications to control friction and wear because of their unique mechanical and tribological properties (Ref 1).

DLC coatings are metastable amorphous materials that have many properties similar to that of diamond. It contains sp, sp 2 and sp 3 hybridized carbon. These coatings are scientifically very fascinating and extremely important for numerous industrial applications due to their wide range of exceptional physical, chemical, mechanical and tribological properties (Ref 2,3,4,5). They are resistant to corrosive and oxidative attacks in acidic and saline media because of their excellent chemical inertness. While diamond is crystalline with complete carbon to carbon (C-C) sp 3 bonding, DLC is amorphous with varying degree of diamond-like properties that is highly dependent on the ratio of diamond sp 3 to graphitic sp 2 C-C bonds. However, the disadvantage of the DLC coating is poor adhesion to metal and metal alloys substrates due to its high internal stress which can be overcome by using interlayers and dopants (Ref 6).

In recent years, deposition of many metal-doped DLC coatings has been attempted to overcome poor adhesion and high residual stress of pure DLC coating (Ref 7,8,9). Nanocluster can be formed when metal atoms are incorporated into the DLC matrix or metal atoms are bonded with carbon atoms. It has been observed that metal-doped DLC coatings are formed with carbide-forming metals like Ti, W and Cr and non-carbide-forming metals like Cu, Ag and Pt. These transition metals modify the structure and help in improving mechanical and tribological properties of DLC coatings because of their d orbital that tends to overlap with p orbital of carbon (Ref 10,11,12,13,14,15,16,17). Among these dopant metals, Cr-doped DLC coating exhibits impressive tribological properties in comparison with undoped DLC coating because of their high adhesion and low stress. Research works on chromium interlayered and chromium-doped DLC coatings have been carried out by many groups, and they have investigated their thermal, mechanical, tribological, magnetic and electrical properties (Ref 18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36). However, detailed studies on electrochemical behavior on chromium-doped DLC coatings are limited in the literature. Wu et al. have studied corrosion characteristics of Cr-doped DLC coatings (Ref 37). But, detailed corrosion studies with electrochemical impedance spectroscopy (EIS) and equivalent circuit have not been attempted. In the present work, we report the development of Cr-doped DLC coatings using combined plasma-enhanced chemical vapor deposition (PECVD) and magnetron sputtering techniques. Furthermore, the effects of doping of chromium with DLC on the corrosion and wear behavior are also discussed.

Experimental Details

Sample Preparation

Three different substrates were used in this study for studying different characterizations and properties. Silicon wafers (100) were cut into appropriate sizes and cleaned in acetone ultrasonically for x-ray photoelectron spectroscopy (XPS), nanohardness testing and Raman spectroscopy studies. Glass slides were cut into pieces and cleaned, and a portion was masked for thickness measurement by profilometer. Stainless steel (202) sheets were cut into 2 cm × 2 cm pieces and were grounded by silicon carbide emery papers of different grit sizes to smoothen their surfaces. After grinding substrates were polished to mirror-like finish with 0.3 μm alumina powder mixed in distilled water. The polished substrates were then ultrasonically cleaned for 15 min in acetone. These substrates were used for corrosion and wear studies.

Deposition of DLC and Cr-Doped DLC Coatings

The deposition chamber was pumped down to base vacuum of 5 × 10−6 mbar by a turbo-molecular pump backed with a rotary pump. The chamber and gas line were purged with process gases after reaching the base pressure. The source gases such as C2H2 and Ar were introduced into the chamber by a shower-head-type distributer, and flow rates were controlled by mass flow controllers (MKS Instruments, USA) and a throttle valve. The plasma was generated by 50 W RF power at 13.56 MHz frequency by inductive coupling. The samples were etched in hydrogen plasma for 10 min before deposition. Pure DLC was deposited with 100% of C2H2 plasma, and Cr-doped DLC was deposited with plasma of Ar and C2H2. Cr was sputtered from 99.95% pure Cr target using magnetron sputtering powered by a symmetric bipolar pulsed DC power supply from MAGPULS Stromversorgungs Systeme GmbH, Germany. Target parameters for sputter deposition of coatings with different Cr concentrations were 300-350 V bias voltage, 0.6 A bias current, 0.6 kW bias power and 50 kHz bias frequency. Pulse on and off times were 7 and 3 µs in both positive and negative channels. Substrates were biased with −110 V DC voltage for the deposition of all coatings. Gas composition and other parameters for the deposition of pure DLC and Cr-doped DLC coatings are summarized in Table 1.

Table 1 Pressure, gas composition, time, Cr concentrations and thicknesses of pure DLC and Cr-doped DLC coatings
Full size table

Characterization

The coating thickness was measured by NanoMap500LS profilometer. Surface morphology of the coatings was obtained by FESEM with Carl Zeiss SUPRA 40 VP. Surface topography was investigated by AFM with CSEM Instruments (Model SSI) operated in contact mode. Raman spectra were collected using DILOR-JOBIN-YVON SPEX made LABRAM 010A with He-Ne laser source having wavelength of 632.8 nm. XPS of Cr-doped DLC coatings were recorded with a SPECS spectrometer, Germany, using non-monochromatic AlKα radiation (1486.6 eV) as an x-ray source operated at 150 W (12 kV, 12.5 mA). The binding energies reported here were referenced with C1s peak at 284.6 eV. All the individual spectra were recorded with a pass energy and step increment of 40 and 0.05 eV, respectively. For XPS recording, sample was mounted on a sample holder and placed into a load-lock chamber with an ultrahigh vacuum (UHV) of 8 × 10−8 mbar for 5 h in order to desorb any volatile species present on the surface. After 5 h, sample was transferred into the analyzing chamber with UHV of 5 × 10−10 mbar. Depth profiling studies were performed by sputtering the coatings with focused Ar+ ion beam using IQE12/38 ion gun by applying energy of 2 keV with Ar gas pressure of 5 × 10−7 mbar for 5 and 10 min. The measured current during depth profiling studies by Ar+ ion sputtering was 1 µA. C1s and Cr2p core level spectra were used to calculate the elemental concentrations of Cr-doped DLC coatings and were also curve-fitted with Gaussian-Lorentzian peaks to identify various carbon and chromium species after subtracting a Shirley background.

Properties

Nanohardness of substrate, pure DLC and Cr-DLC coatings was measured by using Nano Hardness Tester (CSM Instruments) with a Berkovich diamond indenter at 2 mN load.

Electrochemical studies of the substrate, pure DLC and Cr-doped DLC samples were carried out using CH 604D electrochemical workstation (CH Instruments, USA). A conventional three-electrode glass cell was used to conduct the electrochemical studies in 3.5% NaCl solution at room temperature. The sample, Pt foil and saturated calomel electrode (SCE) were used as the working electrode, counter electrode and reference electrode, respectively. The reference electrode was kept very close to the surface of the working electrode. The sample with 0.8 cm2 area was immersed in 3.5% NaCl solution for 1 h to establish the open circuit potential (EOCP). Electrochemical measurement has resulted in Tafel plot which is represented as log i versus potential plot. The corrosion potential (E corr) and corrosion current density (i corr) are obtained from the Tafel plot. The corrosion current is evaluated employing Stern-Geary equation (Ref 38).

$$ i_{\text{corr}} = \frac{{\beta_{a } \times \beta_{c} }}{{2.3 R_{\text{p}} \left( {\beta_{a} + \beta_{c} } \right)}} $$
(1)

where β a and β c are the Tafel slopes related to anodic and cathodic part of Tafel plot and R p is polarization resistance. EIS measurement was taken in the frequency range from 10 mHz to 100 kHz. The applied alternating sinusoidal potential was 10 mV on the E OCP. After each experiment, the impedance data were displayed as Bode plots. The Bode plot is a plot of log|Z| versus log f and −phase angle (θ) versus log f where |Z| and f are absolute impedance and frequency, respectively. The acquired data were curve-fitted and analyzed using ZSimpwin program (Princeton Applied Research, USA) to obtain suitable equivalent circuit parameters. The fitting quality was checked by the χ 2 value. After EIS measurements, potentiodynamic polarization studies were carried out in a potential range of 200 mV below and above the OCP value with a scan rate of 1 mV s−1.

Wear studies of bare substrate and 3.8% Cr-doped DLC-coated substrate were carried out in a reciprocating wear tester (Model CM 9084 DuCom) as per ASTM G133-02 standards (Ref 39). However, these tests are not in full compliance with the provisions of Test Method G 133, Procedure A, because the normal forces in these experiments were 2, 5 and 7 N instead of 25 N as prescribed by the standard. Also, the stroke length was 10 mm and an alumina ball of 6 mm diameter was used as the counter surface. Further, the experiments were carried out at a frequency of 100 Hz and each experiment duration was 20 min. Relative humidity during tribology measurements was around 65%. After the experiments, wear profiles of the samples were examined by a profilometer. Considering ASTM G133-02 method the wear losses were calculated from the cross-sectional areas of the respective wear profiles (Ref 40).

Results and Discussion

Composition and Thickness

In our previous study, it has been found that pure DLC deposited on Ti alloy substrate by PECVD does not adhere to the substrate without an interlayer (Ref 40). Undoped DLC starts to peel off from the substrate after some time. However, we have also reported that Zr- and Mo-doped DLC coatings with graded Zr/Mo composition improves the adhesion to the titanium substrates (Ref 40,41,42). Table 1 shows the Cr concentrations (at.%) obtained from XPS studies. The concentrations of Cr in the coatings vary from 0.5 to 17.6 at.% depending on the process gas composition and the pulse on/off times of the bipolar pulsed power supply. The thicknesses of the films are obtained from profilometry by masking a portion of the glass substrate. The thicknesses of pure and Cr-doped DLC coatings range from 150 to 370 nm. To obtain different Cr concentrations different deposition parameters have been considered while depositing the coatings.

FESEM Studies

FESEM images of the pure DLC and Cr-doped DLC coatings with higher magnifications are shown in Fig. 1(a), (b), (c), (d) and (e) for different Cr concentrations. The coatings show a compact and dense microstructure. Coatings with lower Cr concentrations show smooth amorphous surface morphologies similar to pure DLC coating as observed in Fig. 1(a), (b) and (c), whereas coatings with higher Cr concentrations contain some nanoroughness as shown in Fig. 1(d) and (e). These results are consistent with our previous studies for zirconium-doped DLC (Ref 40). However, no localized delamination of the coatings is observed in all samples.

Fig. 1
figure 1

FESEM images of Cr-doped DLC coatings with different Cr concentrations: (a) pure DLC (b) 0.5%, (c) 3.8%, (d) 10.9% and (e) 17.6%

Full size image

AFM Studies

Figure 2 shows surface topographies of the silicon substrate, pure DLC coating and DLC coatings with different Cr concentrations. AFM images reveal that all the coatings possess smooth surface. However, nanoscale roughness on coated samples has been observed. The average roughness (R a) values of silicon substrate, pure DLC, 0.5% Cr-DLC, 3.8% Cr-DLC, 10.9% Cr-DLC and 17.6% Cr-DLC coatings are 0.44, 0.69, 0.51, 0.36, 1.38 and 0.81 nm, respectively. It is observed that 10.9% Cr-doped DLC coating shows higher roughness value of 1.38 nm compared to all other coatings that is similar to features in FESEM images.

Fig. 2
figure 2

AFM images of (a) substrate, (b) pure DLC, (c) 0.5% Cr-DLC, (d) 3.8% Cr-DLC, (e) 10.9% Cr-DLC and (f) 17.6% Cr-DLC coatings

Full size image

Raman Spectroscopy Studies

The Raman spectra of different Cr-doped DLC and pure DLC coatings on silicon are shown in Fig. 3. For estimating I D /I G ratios of the coatings Raman spectra have been curve-fitted with Gaussian-Lorentzian peaks. Normally, two peaks are used to fit Raman spectra of DLC coatings in the range from 1100 to 1700 cm−1. However, it has been suggested that four Gaussian peaks are required for good fitting of the Raman spectra of DLC (Ref 33, 43, 44). Raman spectra of Cr-doped DLC are fitted with four peaks (D1, D2 and G1, G2). These four peaks are consigned around ~1350 cm−1 for D1 peak, which corresponds to microcrystalline or nanocrystalline graphite, and peak around ~1180 cm−1 for D2 peak, which is nanocrystalline diamond. The peak around ~1490 cm−1 for G1 peak, is related to disordered graphite, and ~1580 cm−1 for G2 peak, corresponds to ordered graphite (Ref 43). These peaks are characteristic of DLC structure (Ref 33, 43,44,45,46). The ratio of the amplitudes of the D1 and D2 peaks and the G1 and G2 peaks is used to estimate the I D /I G ratio. The ratio has been calculated on these peak intensities using D = D1 + D2 and G = G1 + G2 (Ref 43). The I D /I G values represent the degree of disorder in the carbon network. The I D /I G ratios lie in between 1.35 and 0.6 for pure DLC and different Cr-doped DLC coatings and are listed in Table 2. The I D /I G ratios of pure DLC and 0.5% Cr-DLC coatings are more or less similar, and it decreases with an increase in Cr concentrations up to 10.9% Cr. However, the ratio increases with an increase in Cr concentration to 17.6% and therefore suggests that the disorder in the material increases with increasing metal concentration in the DLC coating (Ref 46).

Fig. 3
figure 3

Raman spectra of (a) pure DLC, (b) 0.5% Cr-DLC, (c) 3.8% Cr-DLC, (d) 10.9% Cr-DLC and (e) 17.6% Cr-DLC coatings

Full size image
Table 2 I D /I G ratios, sp 3/sp 2 ratios, nanohardness, Young’s modulus and indentation depth of substrate, pure DLC and Cr-doped DLC samples
Full size table

XPS Studies

Detailed XPS characterization has been carried out to understand the surface nature of pure DLC and Cr-doped DLC coatings. C1s core level spectral envelopes of pure DLC and different Cr-DLC coatings are broad and asymmetrical, indicating the presence of several carbon components in the coatings which are curve-fitted into several component peaks. Figure 4 displays curve-fitted C1s core level spectra of pure DLC and different Cr-DLC coatings. Accordingly, peaks at 283.1, 284.5, 285.3, 287.7 and 289.4 eV observed in the coatings are attributed to carbide, sp 2 carbon (C=C), sp 3 carbon (C-C), C=O and CO 2−3 species, respectively (Ref 47,48,49,50). Ratios of sp 3/sp 2 in all coatings have been calculated from the areas under the respective peaks that is summarized in Table 2. It is observed that sp 3/sp 2 ratio decreases with an increase in Cr concentrations. At very high Cr concentration it increases again. This observation agrees well with the Raman spectroscopy studies.

Fig. 4
figure 4

C1s core level spectra of (a) pure DLC, (b) 0.5% Cr-DLC, (c) 3.8% Cr-DLC, (d) 10.9% Cr-DLC and (e) 17.6% Cr-DLC coatings

Full size image

Figure 5 shows Cr2p core level spectra of Cr-doped DLC coatings. It can be seen from the figure that spectral natures of Cr2p are broad, indicating that Cr is present in various oxidation states in Cr-doped DLC coatings. Cr2p core level spectra of Cr-DLC coatings are resolved into two sets of spin-orbit doublets. Curve-fitted Cr2p core level spectra of 3.8 and 17.9% Cr-doped DLC coatings are also presented in Fig. 5. Accordingly, Cr2p3/2,1/2 peaks observed at 574.4 and 583.8 eV in the curve-fitted spectra are ascribed to Cr carbide species, whereas peaks at 576.7 and 585.9 eV as observed correspond to oxidized Cr3+ species. All these values of Cr carbide and Cr3+ species are close to the values reported in the literature (Ref 51,52,53). Comparing the intensities of Cr carbide and Cr3+ related peaks in the curve-fitted spectrum it has to be noted that around 44% of total Cr is in Cr carbide form and the rest of Cr is present as Cr3+ species in 3.8% Cr-DLC coating. On the other hand, 75% Cr carbide and 25% oxidized Cr are observed in 17.6% Cr-DLC coating. It is noticed that Cr is present mainly in oxidized state in the coatings contained with lower amount of Cr, whereas it is in carbide form in higher Cr content coatings. Relative surface concentrations of Cr with respect to carbon have been evaluated from XPS studies and are given in Table 1. It is important to mention that the initial decrease in sp 3/sp 2 ratio is due to the Cr doping which can be attributed to the catalyst effect of Cr atoms on the formation of sp 2 sites. As sp 2 carbon has lower binding energy compared to sp 3 carbon, it prefers to bond with Cr to form Cr carbide phase. Because of this, a significant numbers of sp 2 carbon are used to form Cr carbide phase resulting in the decrease in sp 2 fraction of carbon which in turn increases sp 3/sp 2 ratio in DLC coatings with higher amount of Cr (Ref 19). Similar type of increase in sp 3/sp 2 ratio is reported for Mo-DLC coating (Ref 54). Thus, results obtained from XPS studies are also corroborated with Raman spectroscopy studies discussed in previous section.

Fig. 5
figure 5

Top: Cr2p core level spectra of (a) 0.5% Cr-DLC, (b) 3.8% Cr-DLC, (c) 10.9% Cr-DLC and (d) 17.6% Cr-DLC coatings. Bottom: curve-fitted Cr2p core level spectra of (a) 3.8% Cr-DLC and (b) 17.6% Cr-DLC coatings

Full size image

To understand the composition of in the subsurface regions, depth profiling studies of the coatings using XPS have also been carried out. Cr2p core level spectra of 3.8 and 17.6% Cr-DLC coatings after depth profiling experiments are shown in Fig. 6. It has been observed that Cr is mainly present in carbide form after successive sputtering in both the coatings. It is also important to note that concentrations of Cr increase after first sputtering and are observed to be same after second successive sputtering in both the coatings.

Fig. 6
figure 6

Cr2p core level spectra of 3.8% Cr-DLC and 17.6% Cr-DLC coatings with different stages of sputtering: (a) as-deposited, (b) after 5 min sputtering and (c) after 10 min sputtering

Full size image

Nanohardness Studies

Nanohardness measurements have been taken on uncoated silicon substrate, pure DLC and Cr-doped DLC-coated silicon samples at 2 mN loads. Figure 7 shows variation of hardness and Young’s modulus with Cr concentrations. The average hardnesses along with Young’s modulus of the coatings are summarized in Table 2. It can be found that uncoated silicon substrate has hardness of 13 ± 1.2 GPa with indentation depth of 135 nm. The hardness values of pure DLC and Cr-DLC coatings are varied from 15 to 22 GPa. In the case of pure DLC coating, hardness and depth of indentation are 21.1 ± 1.5 GPa and 86 nm, respectively. Indentation depth of Cr-doped DLC-coated samples is in the range from 86 to 97 nm that is given in Table 2. According to the table, no clear dependence of hardness on Cr concentrations is observed. This is possibly linked with the change in intrinsic strain during inclusion of metal phase in DLC coating which is observed in Ag impregnation in DLC coating (Ref 55). Previous works have also reported the softening of DLC coatings with metal doping (Ref 56,57,58,59). In the present study, 3.8% Cr-doped DLC shows better hardness (22.6 ± 0.5 GPa) with indentation depth of 89 nm compared to that of other doping concentrations and its Young’s modulus is 190 ± 1.1 GPa. Hardness and Young’s modulus values of 13.5 and 165 GPa are observed in Cr-DLC coating obtained with linear ion beam technique in the work of Dai et al. (Ref 18). They have also reported hardness and Young’s modulus values of 12 and 155 GPa in Cr-DLC coating obtained by hybrid beam technique (Ref 21). Wu et al. have reported hardness value of 26 GPa in Cr-DLC coating deposited with high-power pulsed magnetron sputtering (Ref 37).

Fig. 7
figure 7

Variation of hardness and Young’s modulus in Cr-doped DLC coatings

Full size image

Electrochemical Studies

In Fig. 8, potentiodynamic polarization curves of the substrate, pure DLC and different Cr-doped DLC-coated substrates in 3.5% NaCl solution are presented. The corrosion current density (i corr) has been obtained by extrapolation of the anodic and cathodic branches of the polarization curves to the corrosion potential using Eq 1. Table 3 summarizes the corrosion current density and corrosion potential for the substrate, pure DLC and Cr-doped DLC-coated substrates. The observed corrosion current density (i corr) for substrate is 0.0919 μA cm−2 which is higher than the pure DLC and different Cr-doped DLC-coated substrates. This indicates that coatings increase the corrosion resistance of the substrate. It is also shown in Fig. 8 that the current for formation of a passive layer is lower for the pure DLC and Cr-doped DLC-coated samples as compared to that of substrate. This shows that the pure DLC and Cr-doped DLC coatings have better passivation compared to that of the substrate. The E corr values are shifted to much nobler value for pure DLC and different Cr-doped DLC samples in comparison with the substrate value of −0.236 V. The values of E corr given in Table 3 show that Cr-doped DLC coatings possess good resistance for corrosion. An increase in the concentration of Cr doping results in slight change in corrosion resistance. Pure DLC and 0.5% Cr-DLC coatings show higher corrosion resistance compared to that of 3.8, 10.9 and 17.6% Cr-DLC coatings. This can be due to the fact that DLC has higher corrosion resistance.

Fig. 8
figure 8

Potentiodynamic polarization curves of substrate, pure and Cr-doped DLC-coated substrates

Full size image
Table 3 Results of potentiodynamic polarization studies of substrate, pure DLC and Cr-doped DLC samples
Full size table

In Fig. 9, electrochemical impedance spectroscopy results in 3.5% NaCl solution are displayed in the form of Nyquist and Bode plots. The Nyquist plots in Fig. 9(a) are semicircular with the substrate, pure DLC and Cr-doped DLC coatings wherein Cr-doped DLC coatings have larger diameters in higher and lower frequencies. The limiting impedance at the high-frequency end corresponds to solution resistance (Rs). In Bode phase angle plot in Fig. 9(b), for substrate, the phase angle changes rapidly from −0° to −20° in the high-frequency range (10 to 100 kHz). In the low-frequency range from 0.01 to 100 Hz, the phase angle remains nearly constant around −78° that is lower than −90°, the value for an ideal capacitor. In the same frequency range, a linear relationship between log |Z| and log f is noticed with a slope close to one (Fig. 9c). In the case of pure DLC, there is a rapid change in the phase angle from −50° to −40° in the high-frequency range (10 to 100 kHz); however, at low-frequency range (0.01 Hz to 100 Hz), the phase angle remains nearly constant around −80° that is lower than −90°, the value for an ideal capacitor. In the case of Cr-doped DLC coatings, as shown in Fig. 9(b), the phase angle in the frequency range from 0.1 Hz to 100 Hz is −84° and slightly decreases in the lower-frequency range.

Fig. 9
figure 9

(a) Nyquist plots, (b) Bode phase angle plots, (c) Bode plots and (d) equivalent circuit diagram used for fitting EIS data for substrate, pure and Cr-doped DLC coatings

Full size image

The equivalent circuit (EC) model employed for fitting the EIS data of the substrate, pure DLC and Cr-doped DLC-coated samples is presented in Fig. 9(d). The equivalent electrochemical circuit consists of resistances and constant phase elements (CPE). In the fitting procedure, the non-ideal capacitive response of the oxide layers is entailed employing a constant phase element instead of a pure capacitance. This CPE is because of the difference in the relaxation times as a result of different degrees of surface inhomogeneity. The impedance with the capacitance is defined as Z CPE = [Q (jω n)]−1, where Q, j, ω and n are the pseudo-capacitance or non-ideal capacitance, imaginary function (\( \sqrt { - 1} \)), angular frequency and the deviation from the ideal behavior of a pure capacitor, respectively. When n = 1, the system behaves like a pure capacitor and Q = C where C represents capacitance. In the EC, the resistive components R s , R 1 and R 2 are related to the solution resistance, outer porous layer resistance and inner layer resistance, respectively. Q 1 is the capacitance of the outer layer, and Q 2 is the capacitance of the inner layer (Ref 40). In the case of coated sample, the equivalent circuit is same as that of the substrate with the outer porous oxide layer replaced by roughness and/or DLC coating with pinholes which play the role of pores in the oxide. The values of the electrical parameters obtained by fitting the impedance data of the substrate, pure DLC and Cr-doped DLC samples are given in Table 4. For the substrate, the resistance of outer porous layer (R 1) is observed to be 25 Ω cm2. High corrosion resistance (R 2) of 1.6 × 106 Ω cm2 observed on the substrate is associated with inner oxide layer. The Q 2 component is related to the capacitance of this inner barrier layer. The value of n for the constant phase element representing the outer layer is 0.87 and that representing the inner layer is 0.93. Table 4 shows high R2 values for pure DLC and Cr-doped DLC coatings, indicating that they possess good corrosion resistance. The barrier layer (DLC/Cr-doped DLC layer) resistance of coated samples is higher compared to that of substrate one. This can be due to DLC and doping of DLC with metal. In addition to the above, it is well known that doping of DLC with metals improves its adhesion to substrate and also improves their tribological properties.

Table 4 Electrochemical impedance parameters obtained by fitting equivalent circuit model for substrate and Cr-doped DLC samples
Full size table

Wear Studies

Among different Cr-doped DLC coatings, 3.8% Cr-doped DLC has resulted in good hardness and better corrosion resistance. Hence, reciprocating wear studies have been carried out for 3.8% Cr-doped DLC coating and compared with the bare substrate at different loads. The evaluation of coefficient of friction (COF) on the bare stainless steel substrate and 3.8% Cr-doped DLC coating for different loads in a linear reciprocating wear test is shown in Fig. 10. COF graph of the substrate has two regions. In first region, the COF has low values of 0.4-0.6 and increases gradually, especially after 5 min at 2, 5 and 7 N loads. The COF for the 3.8% Cr-doped DLC is 0.15-0.16 initially and does not change significantly with load. Dai et al. have reported COF value of 0.3 in Cr-DLC coating deposited by hybrid beam technique which is lower than magnesium alloy substrate value of 0.4 (Ref 19). They have also observed COF value of 0.15 in DLC coating on silicon wafer with low Cr concentration deposited by hybrid beam technique (Ref 21). Wu et al. have reported COF values of 0.09-0.22 in Cr-DLC coatings deposited by high-power pulsed magnetron sputtering (Ref 37).

Fig. 10
figure 10

Plots of coefficient of friction vs. time on substrate and 3.8% Cr-doped DLC-coated substrate at (a) 2 N, (b) 5 N and (c) 7 N loads

Full size image

Figure 11 and 12 represent the 3D and 2D profiles of wear tracks for 2, 5 and 7 N loads on substrate and 3.8% Cr-doped DLC coatings, respectively. It can be noticed from the figures that the wear tracks on the substrate have deep grooves, whereas the wear tracks on the 3.8% Cr-doped DLC appear less deep. The average wear loss given in Table 5 shows that 3.8% Cr-doped DLC reduces the wear loss. In the case of substrate, the wear loss is 0.010 to 0.016 mm3 at 2 to 5 N loads, respectively. Wear loss doubles to 0.022 mm3 when increasing the load from 2 N to 7 N. In case of 3.8% Cr-doped DLC, the wear loss is negligible (0.00014 mm3) at 2 N load and slightly increases to 0.0008 and 0.0012 mm3 at 5 and 7 N, respectively, and the effect of increasing load on wear loss is also not significant. It has been observed that the wear loss of the coated substrate is less than that of the substrate and wear marks are also found to be shallower. These observations indicate a change in the wear mechanism from abrasive wear with high COF and high wear loss on the substrate to a mild adhesive wear on the 3.8% Cr-doped DLC-coated substrate, particularly at low loads.

Fig. 11
figure 11

3D profiles of wear tracks at 2, 5 and 7 N loads on substrate and 3.8% Cr-doped DLC-coated substrate

Full size image
Fig. 12
figure 12

2D profiles of wear tracks at 2, 5 and 7 N loads on substrate and 3.8% Cr-doped DLC-coated substrate

Full size image
Table 5 Coefficient of friction (COF) and wear loss for substrate and 3.8% Cr-doped DLC samples
Full size table

Conclusions

Cr-doped DLC coatings have been deposited on stainless steel and silicon substrates with different chromium concentrations. I D /I G ratio decreases with Cr doping up to 10.6%, and then it increases at 17.6% Cr, indicating that disorder gets enhanced in the DLC coating with very high Cr concentration. XPS studies demonstrate the presence of Cr carbide and Cr3+ species in the Cr-doped DLC coatings. At lower Cr contents Cr3+ dominates, whereas carbide species is more at higher Cr concentrations. However, carbide species are mainly present in the coatings after successive sputtering. DLC coating with Cr concentration of 3.8% shows good hardness, corrosion resistance and wear resistance properties. It shows hardness value of 22.6 GPa. Potentiodynamic polarization studies demonstrate that pure DLC and Cr-doped DLC coatings show better corrosion and passivation behavior compared to that of the substrate. Wear studies reveal a low friction coefficient of 0.15 and an order of magnitude lower wear rate of the Cr-doped DLC as compared to uncoated one.