Effect of Al2TiO5 powder coating on the tribological, corrosion and mechanical properties of AISI 316L stainless steel

The aim of the current research is to examine the tribological, corrosion and mechanical properties of AISI 316L stainless steel without and with Al2TiO5 surface coating. AISI 316L is selected for the study, owing to its extensive usage in power plant and marine members that are usually subjected to wear, fatigue and corrosion either separately or in a combinatorial mode from mild to severe intensities. Al2TiO5 coating is provided to components to improve their tribological, corrosion and mechanical properties. Being a ceramic material, Al2TiO5 coating is expected to improve the properties under consideration. The coated specimens are analyzed by considering two factors viz. ‘speed of rotation of job’ and ‘axial speed of the spray gun’, applying Taguchi L4 array. Coating of Al2TiO5 on AISI 316L substrate increased the corrosion resistance, coefficient of friction and micro-hardness, however the wear rate and fatigue life decreased. Twenty times reduction in wear rate is recorded with the coating of Al2TiO5 on the base material when compared to the uncoated counterpart. The wear rate has also decreased by 16% with the increase in coating thickness from 300 to 375 μm. The fatigue life of the coated specimens reduced by around 12% while their corrosion resistance increased by 20% when compared to the uncoated specimens.


Abstract
The aim of the current research is to examine the tribological, corrosion and mechanical properties of AISI 316L stainless steel without and with Al 2 TiO 5 surface coating. AISI 316L is selected for the study, owing to its extensive usage in power plant and marine members that are usually subjected to wear, fatigue and corrosion either separately or in a combinatorial mode from mild to severe intensities. Al 2 TiO 5 coating is provided to components to improve their tribological, corrosion and mechanical properties. Being a ceramic material, Al 2 TiO 5 coating is expected to improve the properties under consideration. The coated specimens are analyzed by considering two factors viz. 'speed of rotation of job' and 'axial speed of the spray gun', applying Taguchi L 4 array. Coating of Al 2 TiO 5 on AISI 316L substrate increased the corrosion resistance, coefficient of friction and micro-hardness, however the wear rate and fatigue life decreased. Twenty times reduction in wear rate is recorded with the coating of Al 2 TiO 5 on the base material when compared to the uncoated counterpart. The wear rate has also decreased by 16% with the increase in coating thickness from 300 to 375 μm. The fatigue life of the coated specimens reduced by around 12% while their corrosion resistance increased by 20% when compared to the uncoated specimens.

Introduction
Components in power plants and marine applications are usually subjected to wear, fatigue and corrosion either separately or in a combinatorial mode from mild to severe behaviour. The biomass fuels release high amounts of corrosive agents effecting parts of power plants while the marine repository gears undergo through all these conditions. To achieve strength and protection against these degradations, surface coatings can be considered as one of the solutions. The micro-abrasion wear coefficient is higher than the abrasion-corrosion wear coefficient for AISI 316L steel. The friction coefficient is also observed to be lower under the same test conditions [1]. When the stainless steel bodies work against each other, pitting and spalling occur which cause crack opening that leads to surface fatigue issues [2,3]. Seasonal variation in temperature and variations due to changes in machine operating variables cause problems in chemical processing plants [4]. The materials with NiAl coating used in fast breeder reactor undergo flow induced variation and experience micro-scale sliding between each other leading to fatigue induced wear [5].
The chromium electroplating and thermal spray coating of WC on ferritic stainless steel increases the hardness and corrosion resistance but affects the fatigue strength [6]. Coating of Zirconia by HVOF method on AISI 316L substrate abridged its fatigue life but resistance to corrosion increased considerably. Zirconia coating enabled to the increase in pitting potential of the substrate from 914 mV to 1009 mV; however the fatigue life was decreased from 11220 cycles, in case of uncoated specimen to 11010 cycles for the coated counterpart [7]. Coating of Al 2 O 3 -TiO 2 increased the corrosion stability, especially for the HVOF-sprayed coatings [8]. It is an evident fact that hardness of steels can be improved substantially by providing Al 2 TiO 5 coating. SKD61 steel uncoated specimen had a Vickers micro hardness value of 252.5 whereas for the coated counterpart the value increased to 816.3 [9]. The substrate material has lower coefficient of friction due to its 'heterogeneous lamella structure'; however the coated specimens exhibited higher coefficient of friction due to the coarse grained microstructure owing to the presence of oxidation particles [10]. Effects of processing conditions on the mechanical properties of distinct thermal spray coatings obtained at distinct process parameters were presented for various materials [11]. The effect of processing conditions like pressure, temperature, powder particle morphology and post-process treatments are observed on the tribological and mechanical properties of cold sprayed coatings [12].
In the current research, high velocity oxy-fuel coating of Aluminium Titanate (Al 2 TiO 5 ) on AISI 316L austenitic stainless steel is undertaken to investigate its influence on tribological, corrosion and mechanical properties of the substrate material. It is a known fact that Al 2 TiO 5 coating inhibits the severe corrosion that occurs in marine parts but the comprehensive evaluation of tribological, corrosion and mechanical properties is scarce till date and thus the current study is embarked. The design of experiments is carried out employing Taguchi methods and TOPSIS algorithm is applied to analyse the results [13,14]. Taguchi method is a simple and efficient quality improvement method to derive optimal conditions. It also enables the users to identify the control factors influencing the quality characteristics. It eliminates the numerous experimental samples especially when the design variables are high in number; the orthogonal array (OA) study as enunciated in Taguchi method caters this need. The TOPSIS algorithm chooses the alternative of the shortest Euclidean distance from the ideal positive solution and largest distance from the ideal negative solution. This method assumes that the criterion is monotonically increasing or decreasing and normalization is carried out in multivariate conditions.

Materials and methods
2.1. Material AISI 316L austenitic stainless steel rods of 14 mm diameter are undertaken for the current investigation. AISI 316L is more popularly used in chemical industries, power plants, marine components etc as it has superior resistance to corrosion and greater strength. The chemical composition is presented in table 1 and the mechanical properties are presented in table 2, for AISI 316L.

Aluminium titanate (Al 2 TiO 5 )
Aluminium Titanate (Al 2 TiO 5 ) ceramic is the stoichiometric solid solution of Al 2 O 3 aluminium oxide and TiO 2 titanium dioxide. Al 2 TiO 5 is taken in powder form of about 200 microns size for the purpose of coating on AISI 316L substrate materials. It has a lower coefficient of thermal expansion, lower elastic modulus and excellent thermal shock resistance when compared to the substrate material. The Modulus of Elasticity of Al 2 TiO 5 is 20 GPa which is around 1/9th of the substrate material and this indicates the brittle nature of the material. Being a ceramic material, the hardness of Al 2 TiO 5 is higher than that of the substrate material. The power coating of Aluminium Titanate would be more feasible and crack free due to its thermal coefficient of expansion to a tune of 1/15th of AISI 316L.

HVOF coating
The powder material is melted by HVOF spray gun and immediately coated on the substrate material. Coating material is sent via the orifice of the spray gun and oxygen mixed fuel gas travels through the conduit of the spray gun. Subsequent to adequate heating at the tip of the spray gun, Al 2 TiO 5 is sprayed on to the specimen in molten form such that it gets coated on the job. The machine employed for coating has the provision to vary the speed of rotation of job and the axial speed of the spray gun. The coating equipment is shown in figure 1 and the parameters maintained during coating of Al 2 TiO 5 on the specimens are enlisted in table 3.
The factors chosen for the experimentation along with their levels are presented in table 3 and the Taguchi L 4 orthogonal array is mentioned in table 4.

Tribometer
To evaluate the 'wear rate' and 'coefficient of friction', the specimens are tested on a pin-on-disc type Tribometer (Magnum make). Both uncoated and coated samples are made to a size of 12 diameter and 30 mm length.  Experiments are conducted for 10,000 cycles for all specimens in ambient air as per ASTM G99 standards. Other test parameters are maintained as per the manufacturer's catalogue for metals and alloys i.e. 100 rpm disc speed and a normal load of 10N on pin. The machine employed for testing is shown in the figure 2.

Fatigue testing machine
Fatigue testing is performed using a plug-and-play 25 KN Servo Hydraulic Nano UTM of ITW-BISS make ( figure 3). The machine has the provision to apply cyclic loads and the fatigue life in cycles in retrieved from the data logger of the machine. The specimens are fabricated as per the ASTM E606 standard and the testing is conducted at a completely reversible stress level of 250 MPa. The gauge length available in the specimen is 15 mm and the drawing of the specimen is presented in figure 4.

Hardness
Micro-hardness testing was done by means of a digital micro-hardness tester of Olympus make, at a loading of 100 grams applied for 10 s. The hardness of the specimens is measured multiple times for consistency.

Corrosion testing
Pitting corrosion tests for the uncoated and Al 2 TiO 5 coated counterparts were carried out using a potentiostat. The electrolyte employed for corrosion testing is 0.5M of H 2 SO 4 +0.5M of HCl. The potential where the current shoots up suddenly after the passive region was considered as 'pitting potential'. Higher the pitting potential value exhibited, higher is the pitting corrosion resistance.

Scanning electron microscope
The fractured features of fatigue specimens after failure were examined using a Scanning electron microscope at various magnification levels. Ultrasonic cleaning and degassing of the samples was performed before placing in the SEM chamber.
2.9. TOPSIS method TOPSIS algorithm is employed to perform the analysis of the results. It is a multi-criterion decision analysis resting on the idea that the alternatives must be at the shortest possible distance from the positive ideal solution and at a longest distance from negative ideal solution. With diversified characteristics of the criterion, the normalization is carried out enabling a trade off between poor and good results. TOPSIS algorithm is presented as a flowchart in figure 5. AISI 316L stainless steel is very widely used in marine applications where members are subjected to wear with sand, salt and other particles. The members are subjected to low cycle fatigue due to ebbs and flows in sea tidal waves along with corrosion due to salt water. Taking these into consideration, TOPSIS algorithm is applied with a weightage of 30% to each of the responses viz wear rate, fatigue life and pitting potential while a weightage of 5% each is given to micro-hardness and coefficient of friction.

Results and analysis
Uncoated and coated specimens of AISI 316L are tested for wear rate, coefficient of friction, hardness, fatigue life and corrosion resistance. The results are presented in tables 5 and 6. For the sake of consistency, each experiment is repeated multiple times and the average of three such tests is taken for further analysis. The wear rate of the uncoated specimens is higher than the coated specimens while the hardness is in the reverse trend. The corrosion resistance of the coated specimens in terms of pitting potential is higher than the uncoated specimens while the fatigue life is in the reverse trend.
With the coating at various speeds and feeds of machine, the coat thickness has varied in the range of 300-375 μm. Lowest thickness is observed when the parameters are at highest level while thickness is highest when parameters are at lowest level. The wear rate is in direct proportion with coat thickness. The fatigue life of the base material without coating is 11253 cycles while the coated samples had a fatigue life in the range of 9856-10895 cycles. A similar trend is recorded with the fatigue life for AISI 316L with ZrO 2 [7].
The corrosion resistance measured in terms of pitting potential is in the range of 1038-1092 mV. The coat thickness and the pitting potential are found to be in the monotonically increasing function and similar trend was observed for AISI 316L with a coating of ZrO 2 [7].
The coefficient of friction on the coated samples is in the range of 0.35-0.55 and this is higher than that of the substrate material. The surface has become rough due to the formation of oxides during coating and this caused the coefficient of friction to shoot up. The coating of ZrO 2 films on AISI 316L by Sol-gel method resulted in a maximum coefficient of friction of 0.59, following a similar trend [15].
By employing TOPSIS method, all the responses under consideration are normalized based on the weights with a view to formulate a realistic model (table 7(a)). It enables a trade-off with the poor results in one set of criterion and a good result in another set [11]. The closeness coefficients computed are illustrated in table 7(b).
By using the concept of Eucledian distance for each alternative, the ideal positive and ideal negative solutions are identified. Then the relative proximity index of each alternative to the ideal solution called as closeness coefficient is computed by the ratio of the positive solution to the sum of ideal solutions (table 7(b)). The closeness coefficients are analyzed by standard Analysis of means (table 8) and Analysis of variance (table 9) based on the methodology suggested by Taguchi [16][17][18][19][20][21][22].
The average response for each combination of control factors is given in Analysis of means (table 8). The delta identifies the size of effect by taking the difference between highest and lowest value of the averages for a factor and subsequently the rank identifies which factor effects more. The factor with highest delta value is given rank 1 and the other one is given rank 2.   Table 6. Results of the coated specimens. Run

Micro-hardness (MH) (H V )
Coat thickness (CT) (μm) The analysis of variance gives out the relation between response and controlling factors along with the variation between samples and within samples. The total deviations are computed by squaring the difference between each alternative from the mean called as sum of squares (SS). The degrees of freedom (DoF) is the difference the number of levels are unity since it is measuring about the overall mean, while for total degrees of freedom it is the difference between total number of samples and unity.

Results and discussion
The coating is done by passing on the Aluminium Titanate through the annular space of the HVOF gun. The powder gets melted at the tip with the aid of neutral flame of Oxy-fuel gas welding torch. The molten metal gets carried away by high velocity flame and gets deposited on to the substrate. The TiO 2 in the powder intended to promote corrosion resistance primarily along with other properties on the surface, but inhibits the wettability on the substrate. However, the Al 2 O 3 in the powder promotes the adhesion property on the substrate and helps in forming a highly resistive layer [23].
The thickness of the coating in all the experimental runs is measured and is found to be between 300 and 375 microns. Lowest thickness is formed on the specimen where the run conditions are at highest level while the thickness is largest when the conditions are at the lowest level. With the increase in thickness of coating from 300 to 375 microns, the wear rate reduced by 16% and the hardness of the surface increased by 21%. This can be attributed to the presence of Aluminium Titanate (Al 2 TiO 5 ) coating which is a ceramic material possessing high hardness and wear resistance. The range of fatigue life reduction between the coated and the uncoated specimens is about 1400 cycles and this can be attributed to high heat inputs incurred in the process of coating. The temperature is in the range of 2500°C-3000°C and the Aluminium Titanate gets melted and sprayed fast on to the substrate maintained at room temperature. The micro-pores on the coated surface of the specimen became the sites for crack formation and their subsequent rapid propagation lead to fatigue life reduction. The coefficient of thermal expansion of aluminium, stainless steel and titanium are different and hence microporosity is caused as exhibited in the SEM micrograph ( figure 6). The coefficient of friction has thus increased due to the change of surface morphology on the stainless steel substrate. Also Al 2 TiO 5 coating contains aluminium that exhibits higher coefficient of friction compared to other metals [24]. The corrosion resistance in terms of pitting potential has increased due to layer thickness of the coating owing to the coating parameters at their lowest level. With the coating of Aluminium Titanate on AISI 316L substrate, the corrosion resistance increased by 20% compared to the parent material. The variation of the pitting potential among the coated specimens is about 5% and this can be attributed to the presence of microcracks due to the oxide layer formation during the HVOF coating process. From the ANOM (table 8) and ANOVA (table 9) of closeness coefficients, it is evident that the speed of rotation of job played a major role (Rank 1) with 85.66% contribution while the axial speed of the spray gun played a secondary role with a contribution of 12.7%. The current study enables to predict the optimal values of the controlling factors f 1 and f 2 . The factors at their low level show optimality condition with higher average of closeness coefficients (table 8). Higher-thebetter criterion of Taguchi method is applied for the closeness coefficient and optimal parameter combination is elicited. The factors at their lower levels turned out to be the optimal condition and the optimal value (predicted) is found using the relation The optimal value of the closeness coefficient at the optimal parameter combination is 0.8909 and the optimal results of the coated specimens are found to be of the Run 1 in table 6. From the results, it is clearly evident that the hardness, corrosion resistance and coefficient of friction increased while the fatigue life and wear rate decreased due to the HVOF coating.

Conclusions
• Coating of Al 2 TiO 5 on AISI 316L steel increased its corrosion resistance, coefficient of friction and microhardness; however the wear rate and fatigue life decreased due to the same.
• Twenty times reduction in wear rate is recorded with the coating of Al 2 TiO 5 on the base material when compared to the uncoated counterpart. The wear rate has also decreased by 16% with the increase in coating thickness from 300 to 375 μm.
• The fatigue life of the coated specimens reduced by around 12% while their corrosion resistance increased by 20% when compared to the uncoated specimens. The reduction of Fatigue life can be attributed to the formation of micro-cracks due to high heat inputs during the HVOF coating process. • The optimal parameter combination for the multi-response domain is identified at the low levels of the factors. The speed of rotation of job played a major role with a contribution of 85.66% while the axial speed of the spray gun played a secondary role with a contribution of 12.7% in the current multivariate analysis.