Experimental studies on wear and corrosion resistance of pulse electrodeposited Ni-TiO2 nanocomposite coatings on AISI 304 stainless steel

Nickel-nano titanium oxide (Ni-TiO2) composite coatings were developed on stainless steel 304 substrate using pulsed electro-deposition technique. Experiments were carried out employing L27 orthogonal array, taking current density, frequency, and duty cycle as prime parameters. Coating characterization was done using FESEM with EDAX and XRD. Applying the Response Surface Methodology optimization method, the optimized values of the parameters were identified as duty cycle, frequency and current density with values of 24%, 10 Hz and 0.6 A cm−2 respectively. Vickers micro-hardness, surface roughness, wear resistance and corrosion resistance were measured initially for the as-coated sample. The sample that yielded the optimum parameters was then heat treated for an hour at 400 °C in a closed furnace and quenched in SAE 40 grade oil. Test results of post heat-treated sample produced maximum micro-hardness of 446.45Hv, reduced surface roughness of 0.292 μm, superior wear rate of 3.20E−07 mm3 N−1-m−1 and corrosion protection efficiency of 94.52% of Ni-TiO2 coating at low frequency, along with enhanced wear and corrosion resistance as compared with the as-coated sample.

different factors manipulated for obtaining surfaces with higher volume of particles in the coatings [23]. For the optimization study in this work, experimental variables with corresponding values have been selected, based on the literature [24] to produce maximum inclusion of nano particles in the composite coating, and to get enhanced surface properties. The Ni-TiO 2 composite coating shall prove as an advantageous and a better coating as against hard chromium coatings, owing to better wear resistance [25] and least surface roughness conducive for plain bearing applications [26].
In this study, the Ni-TiO 2 nanocomposite coating is produced by pulsed electro-deposition technique. From the literature survey, it is inferred that only very few research works have analyzed the electrodeposited Ni-TiO 2 nanocomposite coating samples that are further surface hardened by oil quenching. The main aim of this in situ experimental work is to analyze the micro-hardness, surface roughness and properties like wear and corrosion resistance, by studying the Ni-TiO 2 composite coated samples that are heat treated at 400°C [11,27] for one hour in a closed furnace with rapid oil quenching in SAE 40 grade oil [28]. By using Box-Behnken design in Response Surface Methodology (RSM) [12,29]; the quadratic models were developed with the help of Minitab software for optimizing the process parameters.
Section 2 describes the experimental details like samples preparation and procedure. Section 3 details the experimental design for optimization and the optimization output. Section 4 unfolds a detailed discussion on the results obtained, regarding the analyses of microstructure, micro-hardness, surface roughness, wear and corrosion resistance for the as-coated and heat treated samples. Section 5 draws the conclusion of the present studies.

Experimental details
2.1. Samples preparation Samples are prepared using stainless steel 304 as the base material. Initially 27 numbers of stainless steel 304 work pieces are taken for the coating process. Electro-deposition coatings of Ni-nano TiO 2 were prepared in 300 ml of nickel sulphate bath solution in a glass beaker. Chemicals used for preparing the bath where purchased from Merck brand. Pure nickel plate was used as anode and stainless steel 304 as cathode. A distance of 3 cm [30,31] is maintained between anode and cathode in the bath. The composition of the bath and operating conditions are given in table 1.
The schematic diagram in figure 1 explains the setup for the coating process. The sample preparation for coating process is illustrated in figure 2. The procedure adopted for coating is illustrated in figure 3. The optimum voltage was found using trial and error approach and the coating time was calculated using Canning handbook surface finishing technology [32].

Experimental procedure
Surface morphologies of the coatings were examined using Field Emission Scanning Electron Microscope (FESEM) coupled with Energy Dispersive X-Ray Spectroscopy (EDAX). All chemical composition values are quoted in weight percentage.
The micro-hardness of the nanocomposite coated surface using the sample size of 10 mm diameter and 10 mm height was measured by Vickers micro-hardness instrument with a load of 50 g and five indentations [18,32] on the surface. The measured values have been tabulated for both as-coated samples and heat treated sample. The X-Ray Diffraction (XRD: Philips X'pert Pro) analysis was performed on both samples having the dimensions of 10 mm diameter and 2 mm thickness with Cu K a radiation. The surface roughness was measured using surface profilometer and three readings have been recorded and the values were averaged. Wear test was conducted on coated samples as per ASTM G99 standard using the pin on disk tribometer with an applied load of 20 N, sliding distance of 1000 m, track diameter of 60 mm, and sliding time of 15 min The test was carried out in room temperature under dry condition. A multi-channel electrochemical workstation (CHI-650C, USA) was employed to study the corrosion behavior of as-coated and heat treated samples in 3.5% NaCl solution [19,33] using three-cell electrode and platinum foil electrode. Saturated calomel is the reference electrode. SS304, prepared as per ASTM G106-89 standard, and coated and exposed in 10 mm diameter area, is the working electrode. The open circuit potential (OCP) were performed in the -300 mV and +300 mV range in NaCl medium, followed by electrochemical impedance spectroscopy (EIS) measurements. EIS testing was performed at the frequency range of 100 kHz to-0.1 Hz with amplitude of 10 mV at the OCP.
Potentiodynamic polarization curve tests were carried out over the range from -300 mV below the OCP to +300 mV above the OCP at a scanning rate of 0.5 mV s −1 . The criteria of reactivity of the sample were chosen as the corrosion current that was extrapolated at the corrosion potential using cathodic Tafel extrapolation.   The following formulae were used in pulse plating [23] Crystallite dimensions were calculated according to equation (5) by using Debye-Scherrer formula [10] with full width at half maximum (FWHM) values of X-ray diffraction. Here, D is mean crystallite size, λ is X-ray wavelength (0.15418 nm), β is the corrected peak width at half maximum intensity (FWHM) and θ is Bragg diffraction angle.
Using the optimum parameters as obtained from optimization, five numbers of samples were further coated and heated for one hour in furnace at 400°C, the recrystallization temperature of nickel [27] and suddenly quenched [28] in SAE 40 grade oil for improving the hardness by rapid cooling. The micro-hardness and surface roughness values of these five samples were measured and the average value is determined.

Process parameter optimization
3.1. Experimental design RSM is a widely used technique to optimize process parameters for more number of variables [12]. By using three independent variables, namely, frequency (f), duty cycle (DC) and current density (CD) at three levels, the Box-Behnken design was used in RSM with 27 experiments. Micro-hardness and surface roughness as output responses with optimized data the wear resistance and corrosion resistance was found. The experimental variables and the corresponding levels are indicated in table 1. The design matrix along with the experimental values is exhibited in table 2.

Effect of process parameters
Experiments were conducted and the quadratic models (6) and (7) were developed by using the experimental inputs in second order polynomial equations and the multi-objective optimization equation was also formed with the help of Minitab software.
The empirical relations of micro-hardness and average surface roughness, with the frequency (f), duty cycle (D), current density (CD) are expressed as second-order polynomial (regression) equations for framing the response surface using [30]. The model presents high determination coefficients R 2 of 95.82% for microhardness and of 97.18% for surface roughness of the responses. The responses of micro-hardness and average surface roughness were evaluated using the Analysis of Variance (ANOVA) tables 3 and 4.
The regression equation in un-coded units for the microhardness relationship with frequency (F), duty cycle (DC) and current density (CD) the coating parameters were expressed in the second order polynomial equation (9).
As inferred from the ANOVA table 3, the quadratic term frequency had the most influencing parameter contributing 39.11%, followed by the linear term current density of 14.07% impact on the hardness value. The other terms interaction shows that the frequency-duty cycle, frequency-current density, linear term duty cycle, current density, quadratic term duty cycle contributing the hardness with values of 10.78, 8.33, 7.77, 6.90 and 6.84% respectively. High current density and low frequency has formed smaller crystal sizes of 17.86 nm in the coating as found using XRD and this provides increased hardness in the coatings. As per the hall-petch effect, the micro-hardness value was increased with lower frequency and higher amount of current density [13]. The model provides high determination coefficient values of R 2 and adjusted R 2 as 95.82% and 93.61% respectively that are higher than the predicted R 2 value of 90.61%.
The regression equation is expressed, as before, as the second order polynomial equation (7).
The most influencing factors from the ANOVA table 4 shows that the linear term frequency, the interaction term frequency-current density, duty cycle-current density and quadratic term frequency giving 30.27, 29.43, 14.75 and 11.34% respectively for the average surface roughness value. The less influencing factors are from the interaction terms frequency-duty cycle, quadratic term current density, linear term current density and duty

Multiple response prediction for both micro-hardness and surface roughness
From the multi-objective optimization plot shown in figure 4, by giving equal weightage to both outputs with composite desirability of 0.99, the micro-hardness and average surface roughness values predicted the optimized parameters. By using the optimized parameters, conformation test were conducted with frequency, duty cycle and peak current density of 10 Hz, 14% and 0.26 A cm −2 as input parameters which produced the output responses of micro-hardness 319.15 Hv and surface roughness of 0.323 μm.

Microstructure analysis
The experimental results of the coating in this present study are compared with the results of earlier studies performed on Ni-TiO 2 nanocomposite coating. The maximum inclusion TiO 2 in the coating was achieved at frequency 10 Hz [13]. Using the EDAX spectra of Ni-TiO 2 nanocomposite coating, the elements Ni−86.94%, Ti-7.52% and O-5.53% in weight percentage are present on the surface. From the FESEM image (figures 5(a)-(d)), it is found that the surface morphology of the titanium oxide was densely absorbed at the substrate surface, formed in cauliflower shape [34,35], and as worms like structure that are visible at 200 nm and 100 nm magnification. The presence of nickel is high due to the higher current density.

XRD analysis
The XRD plot, in figure 6, shows that amorphous structure was formed and characteristic oxide content was not shown in the as-coated surface. The sample that is heat treated at 400°C and oil quenched also shows the same pattern. However the difference observed is that the heat treated sample matched well with JCPDS-3-65-5537 due to the provision of required thermal energy, but in the case of as-coated sample there is a slight shift observed at 2θ angle of 42.5 degree due to the close packing of nanocomposites in the coating, forming the plane of 110 [31]. However both peaks of the coated surface show Ni-Titanium phase (Ni-Ti) presence in the coating, and crystal size has increased with the applied thermal energy. There is no presence of oxygen in the coating as ascertained by EDAX performed to find the oxygen presence along with titanium. The full width at half maximum (FWHM) value for the heat treated surface is higher when compared with as-coated surface. Hence, the crystallite size is also increased in heat treated surface. The average crystal size of the as-coated sample was calculated using Debye-Scherrer formula (equation (5)) as 17.86 nm. This clearly indicates that the formation of crystal size is smaller due to the addition of nucleation sites in Ni electrodeposition coatings [35]. The inclusion of nickel in the coatings was reduced due to inclusion of TiO 2 in bath, so smaller sized crystals have been formed. The heat treated surface produced average crystallite size of 23.83 nm and the size was increased due to thermal treatment [27].The data from EDAX analysis shows a TiO 2 presence of 13.05% by weight in the composite coatings.

Micro-hardness analysis
The micro-hardness values of the bare, as-coated Ni-TiO 2 SS304 samples and heat treated Ni-TiO 2 SS304 samples were assessed by using Vickers micro-hardness instrument. Due to low frequency and low duty cycle, the maximum hardness was achieved with maximum inclusion of TiO 2 particles and longer off-time in the pulse plating [13]. The as-coated work piece has shown improved hardness than the bare surface. The increase in hardness is attributed to the following three factors: (i) particle strengthening, (ii) dispersion strengthening and (iii) grain refinement [13]. Due to structural refinement of the grains and smaller crystallite size in the coating, the hardness value has been increased. The absorption of TiO 2 increases the nucleation site, delays grain growth, so falling in the grain size and subsequently increasing micro-hardness of the as-coated surface [31].
The micro-hardness value of bare surface is 250 Hv, and after the Ni-TiO 2 deposition over the substrate the micro-hardness value is increased to 319.15 Hv. In addition, it was discovered that the microhardness increased after post-heat treatment and oil quenching, rising from 319.15 Hv to a high of 446.45 Hv. Due to rapid cooling in oil medium, the particles are packed tightly giving high hardness and also better wear resistant surface.

Surface roughness analysis
In pulsed electrodeposited Ni-TiO 2 nanocomposite coating, the surface roughness is mainly dependent on the current density and the coating thickness [36]. The surface roughness values were measured and averaged value of three readings at three different locations were determined. The bare, as-coated, and heat treated surfaces produced average surface roughness values of 0.480, 0.336 and 0.292 μm. The inclusion of second phase element TiO 2 nanoparticles had major influences on the roughness values. The average surface roughness value has been decreased for the heat treated sample from 0.336 to 0.292 μm, due to grain refinement that formed the smooth surface; hence the wear resistance was much increased to protect the surface [35]. Since surface roughness value for plain bearing applications must be less than 0.4 μm [26]. On comparison, the value obtained in our experiment for heat treated sample (0.292 μm) is lesser nearly by 13.09% than the as-coated surface. Thus it can be clearly inferred that heat treating the coated part can be a good procedure for plain bearing applications.

Wear analysis
The morphology of wear tracks for the as-coated and heat treated samples are shown in figures 7(a)-(d) and Tribological properties are highlighted in table 5. The hardness and surface deterioration due to wear of the material had the inverse relationship as per Archards law [36]. The worn surface clearly shows that deeper grooves are visible in the as-coated worn surface indicating more amount of material and lesser wear resistance [37,38]. However, the heat treated worn surfaces are having lighter grooves and minimum amount of material has been removed during the test. The weight loss of the as-coated surface is 0.00438 grams and that of the heat  treated surface is 0.00032 grams (lesser by 0.00406 grams). This lesser weight loss is due to the plastic deformation and high wear resistance to the applied load.
Due to the grain refinement, heat treated sample had greater wear resistance than as-coated and bare surface. The average coefficient of friction value of as-coated surface and heat treated surfaces are 0.4509 and 0.2075. The lower the friction coefficient value, the higher is the hardness produced, with smooth surface in heat treated surface. The decrease in the value of friction coefficient is due to plastic resistance of the heat treated surface and reduced area of contact between the mating surfaces. Heat-treated surfaces had lower rates of wear, which led to higher wear resistance being seen from the wear data and making them a viable alternative to hard chromium plating [25].

Potentiodynamic polarization
The potentiodynamic polarization and electrochemical impedance spectroscopy (EIS) data for bare surface, ascoated and heat treated surfaces are given in the table 6. Corrosion parameters like corrosion potential (E corr ), corrosion current density (i corr ), linear polarization (R), corrosion rate (r corr ), and corrosion protection efficiency were calculated from the Tafel plot given in figure 8(a). From seen from the Tafel plot, the curves are formed smoothly for the as-coated and heat treat samples indicating that the surfaces are highly passive to corrosion attack. The corrosion surface protection efficiency is 95.55% and 94.52% for as-coated and heat treated surfaces and this indicates very less porosity in the coated surface. The anti-corrosion property of the coatings is increased due to free from cracks, and the uniformity of the coating is improved with decreased grain size in coating to form passivation [35]. The corrosion current density of the as-coated sample is 10 −6 A cm −2 ; hence the ascoated Ni-TiO 2 surface can be a vital replacement for chromium coatings because of its range from 10 -6 to 10 -7 A cm −2 and lesser microhardness value than the chromium coating [39].

Electrochemical impedance spectroscopy
Since the pores are getting reduced, the corrosion resistance has increased [33,36] for the as-coated and heat treated surfaces, and grains are refined well due to morphological changes in heat treated sample. As indicated by the surface roughness values, smooth surface is formed in the heat treated sample and this will reduce the corrosion rate by dropping the corrosion sites in it [18]. The R ct values of heat treated surface as against bare surface decline, (7000 to 45), which clearly indicates that the semi-circle diameter for bare surface is higher and more protection was given in 3.5% NaCl solution. From the Nyquist plot figure 8(b), the heat treated surface and as-coated surface had least diameter and medium semi-circle while comparing with bare surfaces, and hence both surfaces have good electrical conductivity and thermal conductivity properties than the bare surface. Due to less corrosion resistance of heat treated sample due to enhanced ion transfer between the coated surface and the corrosion medium [41]. However, from the present experimental results, both as-coated and heat treated surfaces had excellent corrosion resistance and their corresponding values are given in table 7.

Conclusion
The nanocomposite Ni-TiO 2 coating has a mixture of cauliflower and worms morphology, as per FESEM images analysis, which confirms uniform distribution of nano TiO 2 particles with the nickel matrix. The EDAX data suggests a maximum inclusion of 13.05% of titanium oxide by weight percentage in the coating.
The microhardness and Corrosion protection efficiencies of the heat treated surfaces were enhanced by 78.58% and 94.55%. Consequently, the surface roughness and the average co-efficient of friction values were reduced by 13.09% and 75.73% than the bare surface.
The above results clearly suggest that Ni-TiO 2 coating of parts and post heat treating them can be the right process for many appropriate industrial applications-like plain bearing parts. Ni-TiO 2 coating and heat treatment can also serve as a viable replacement for hard chromium coating.