Liquid Phase Surface Nitriding of Al-5052

Liquid Phase Surface Nitriding of Al-5052 Abstract: Liquid phase surface nitriding of Al-5052 was performed using the heat of a TIG (tungsten inert gas) torch in a gas shielding which was a mixture of argon and nitrogen. The feasibility of obtaining nitride compounds at various TIG processing parameters and nitrogen contents in the shielding gas were studied. The presence of AlN phase being formed during surface nitriding was proved by X-ray diffraction analysis. Scanning electron microscopy equipped with energy dispersive X-ray spectroscopy (EDS) analyzer was carried out to study the morphology and chemical composition of the nitride phase. The microhardness test was also performed on cross sections of treated layers. This measurement demonstrated that the surface hardness increased from 52 HV for the untreated aluminum alloy to as high as 1411 HV for the nitrided sample due to the formation of AlN phase in the treated layer. It was also found that, variation of nitrogen contents in the shielding gas has little effect on th e formation of AlN phase and its properties. It was also noticed that liquid phase surface nitriding reduced the wear rate to less than quarter of that of the untreated substrate. Introduction Liquid phase surface engineering including surface melting, alloying, and formation of composite layers on aluminum alloys have been studied and applied for more than three decades. High-energy sources such as laser and electron beam, as well as other heat sources like tungsten inert gas (TIG) process have been used for these treatments [1–3]. In order to improve the wear resistance, formation of hard nitride layers via liquid phase surface engineering on nitride former alloys like titanium and iron in atmospheres containing nitrogen have also been studied by a number of researchers [4–11]. Aluminum alloys like titanium are strong nitride former. Attempts have been made to form nitride compounds on aluminum and its alloys to enhance their wear resistant [12–16]. The majority of researchers have used plasma nitriding technique. The main disadvantage of plasma nitriding is formation of rather thin AlN layers, which are not suitable, and useful while high load bearing ability is required [12,13,17–19]. Some researchers have tried to form aluminum nitride via liquid phase surface engineering of aluminum using laser beam [14,20–24]. Sicard et al. [22] obtained thin nitride layers on aluminum based substrate by liquid phase laser nitriding. Carpene et al. [23] studied laser nitriding of pure iron and aluminum in nitrogen atmosphere using a pulsed nanosecond Excimer laser. Their study revealed that approximately all the phases predicted by the Fe-N phase diagram was observed in the case of liquid phase iron nitriding, while in aluminum, only AlN was formed . There are only a couple of works on liquid phase surface nitriding of aluminum using electric arc in atmospheres of argon and nitrogen [15,16]. Hioki et al. [15] introduced an aluminum nitriding method by heating aluminum in a mixture gas of argon and nitrogen using the heat of a TIG torch. By this treatment, a dense layer of aluminum nitride was formed on the surface of aluminum so that it improved the wear resistance of aluminum. Zheng et al. [16] reported an improvement in the microhardness and wear resistance of 1050 aluminum by nitrogen arc discharge at atmospheric pressure. The nitride formation mechanism via liquid phase surface treatment has not been completely realized. According to some researches [16,20,21], the plasma formation by the electric arc or laser irradiation on the substrate surface under nitrogen atmosphere allows ionization of nitrogen and penetration to some depth and then according to Al+N → AlN reaction, nitride layers grow in the melt pool. It has been reported that if the proportion of nitrogen gas exceeds 50% by weight, the paucity of argon gas might result in unfavorable effects on generation and stability of the electric arc [15]. Therefore, it is preferred that the shielding gas to be diluted by argon gas. In this study, TIG surface nitriding of Al-5052 in ambient nitrogen atmosphere will be carried out to investigate the effects of various TIG processing parameters such as current and travel speed as well as nitrogen contents on the formation of AlN on Al-5052 alloy. Subsequently, the hardness and wear resistance of the treated surfaces were studied. Experimental AA5052 aluminum plates with dimensions of 100 mm Ãâ€" 80 mm Ãâ€" 10 mm were used as the substrate. Prior to surface nitriding, their surfaces were sandpapered with 120 paper grit SiC and then cleaned with acetone. TIG surface treatment was carried out using a MERKLE TIG 200 AC/DC unit in alternative-current (AC) mode as a heat-generator. A coaxial argon gas flow was adjusted at a fixed amount of 9 l/min and high purity nitrogen gas (at flow rates of 3, 4, and 5 l/min) was blown into the molten pool to provide shielding. Tungsten electrodes with diameter of 2.4 mm and a constant distance of 2 mm from the specimens’ surfaces were used for all experiments. Surface melting trials were conducted to optimize the TIG processing parameters (Table 1). The effects of volume percentage of added nitrogen to the shielding gas and TIG processing parameters on the properties of the fabricated layers were studied. On the whole, liquid phase surface nitriding was performed under two different series of processing parameters. In the first series, surface nitriding was performed in a constant mixture of argon and nitrogen gas atmosphere at various TIG processing parameters and in the second series the mix tures of argon and nitrogen gas shielding were changed while other TIG operating parameters were kept constant (Table 2). The voltage of TIG process was kept at a constant value of 15 V, the current varied from 75 to 150 A, and the travel speed differed from 50 to 200 mm/min. The heat input for each test was calculated using Eq. 1 [25]. Heat input (kJ/cm) = (0.48 Ãâ€" voltage Ãâ€" current)/(Travel speed) (1) The nitrided layers were characterized and analyzed by optical microscope (OM) and scanning electron microscope (Model:Camscan MV2300) equipped with an EDS analyzer. The samples used for microanalysis were polished metallographically to get smooth surfaces and then were etched with Kellers reagent for 15–30 s. The nitrided layers were also analyzed using a Philips X’Pert Pro X-ray diffractometer equipped with a Ni filter, Cu KÃŽ ± source operating at 40 kV and 30 mA. The cross-sectional hardness of the surface treated layer was measured by a MicroMet microhardness testers-Vickers with an applied load of 100-200 g and holding time of 15 s. The given values of hardness were average values taking from three to five measurement points at the same depth. The wear rates of the samples at room temperature and humidity of 45% were also evaluated by measuring the weight loss, using a pin-on-disc wear test machine. The cylindrical pins with a diameter of 4.9 mm were wire-cut from the untreated AA5052, surface melted and surface nitrided samples for the wear tests. A quench-tempered steel (AISI 52100) disc with a diameter of 37 mm and hardness of 59 HRC was chosen as the counter face. The testing parameters were 20N load, 0.3 mm/s sliding speed, and 250, 500, 750 and 1000 m sliding distance on a radius of 12.5 mm from the center of the disc. 3. Results and Discussions 3.1 Surface melting Fig. 1 shows a typical cross sectional view of a crack and porosity-free surface melted specimen achieved at a heat input of 2.16 kJ/cm (current of 100 A and travel speed of 200 mm/min). This figure also shows that the optical macrostructure of the cross section of the surface melted specimen is composed of three distinctive structures: Area 1 is the unchanged structure of the base metal. Area 2 with columnar structure, which is formed due to the high heat transfer rates because of rapid solidification and high thermal gradient between the melted zone and the base metal. Area 3 with equiaxed structure, which is emerged due to heat transfer rates during the melting process. 3.2 Surface nitriding: Effects of various TIG processing parameters Liquid phase surface nitriding was carried out under various TIG processing parameters in a constant mixture of nitrogen–argon shielding gases. Surface nitriding caused the formation of gray colored tracks, with 0.6–1.6 mm thickness and 3–6 mm width, indicating composition changes and possibly formation of aluminum nitride in the treated layer. A couple of other works have also reported similar observations [16,21]. Fig. 2a and b shows the effect of heat input on the depth and width of the treated zone. The depth and width of treated zone proportionally increased with increasing heat input. In addition, the change in gradient due to increased heat input is the same in both graphs. Fig. 3a and b shows the surface treated zone achieved at the minimum (N-1) and maximum (N-4) heat input used in this work, when the mixture of nitrogen–argon shielding gas was remained constant. In the sample with maximum heat input, the treated layer is larger and contains crack s, which are due to the formation of hard aluminum nitride and high temperature gradient. The rough nature of the treated layer is due the metal evaporation as result of high heat input. EDS analysis from the marked areas (Fig. 3c and d) reveals aluminum and nitrogen percentages for N-1 and N-4 specimens. Nitrogen content in the sample with maximum heat input (27.22 at%) was much lower than the nitrogen content in the sample with minimal heat input (40.41 at%). Increasing heat input results in dissolved nitrogen in the larger melting pool of aluminum and there would be less excess nitrogen. 3.3 Surface nitriding: Effects of shielding gas Surface nitriding was also processed at various volume percentage of nitrogen in the shielding gas when the other TIG processing parameters were kept constant. Fig. 4 shows the low magnification cross sectional SEM micrograph of sample N-5 (see Table 2), near its surface. According to this figure, the nitrided layers are consisted of two morphologies, one is dendritic morphology, and other is lamellar morphology dispersed between the dendrites. Similar morphologies are reported by other researchers [16]. Fig. 5 shows the high magnification SEM micrograph of sample N-5 that reveals these two morphologies, separately. Detailed SEM-EDS spot analysis (Fig. 5c and d) suggests that in lamellar structure, the dark areas are aluminum nitride and the bright bands are aluminum. The EDS analysis results disclosed that the atomic percentage of nitrogen and aluminum are almost 1:1 at the dark areas. X-ray diffraction patterns of the nitrided zone also confirmed peaks corresponding to hexagonal Al N phase (Fig. 6). Other researchers also reported formation of hexagonal AlN phase [12,16]. AlN phase was seen at 2ÃŽ ¸ = 33.216, 36.041, 37.917, 49.816, 59.350, 66.054, 69.731, 71.440, 72.629, 81.090 and 94.844 (PDF no. 25-1133). There are other peaks in the XRD patterns at 2ÃŽ ¸ = 38.473, 44.740, 65.135, 78.230, 82.438, and 99.081, that correspond to face centered cubic aluminum structure. Fig. 7 shows the high magnification cross sectional SEM micrographs of samples that were surface nitrided under various volume percentages of nitrogen in shielding gas. The morphologies of the samples treated with different nitrogen flow rates (3, 4 and 5 l/min) were similar to each other and by changing the nitrogen flow rate, the atomic percentage of nitrogen in the dark bands did not changed, significantly (Table 3). However, it seems that the thickness of the dark bands increased with increasing the volume percentages of nitrogen gas shielding. Meanwhile, bright areas increased that indicate d aluminum content changed by reducing the amount of nitrogen in the gas mixture. 3.4 Microhardness Table 4 compares the average microhardness of a number of surface melted samples with that of the base metal. In surface melted samples, the hardness of the surface melted layer reduces by increasing of the heat input applied. Nevertheless, minimum hardness of the surface melted samples (81 HV) was much higher than that of the base aluminum substrate (52 HV). Table 5 shows the average microhardness of a number of surface nitrided samples being compared with that of the base metal. The increase in hardness was due to the formation of AlN phase in the nitrided zone. In surface nitrided samples by decreasing the heat input, the size of the treated layer decreased and hence, the solidification rate increased, resulting in a finer grained structure. In addition, by reducing the heat input, nitriding pool shrinks and nitrogen content rises from about 27 to 40 %at and therefore the relative amount of nitride compounds is increased in the layer. The hardness of the nitrided sample achieved a t higher heat input (HV 1109) is lower than that of the sample nitride at lower heat input (1411HV), which is close to the hardness value reported by others [13,16]. Furthermore, various volume percentages of nitrogen in the gas shielding had an insignificant effect on the average hardness of the nitrided zone. Fig. 8 illustrates the microhardness profiles along the depth of the layer achieved for surface melted (M-1) and nitrided (N-5) samples. These profiles indicate that the microhardness of the surface nitrided sample is much higher than that of the surface melted sample. The hardness for the surface melted (M-1) sample reached 148 HV and for the surface nitrided (N-5) sample increased by up to 1411 HV. Besides, the hardness of the nitrided layer along treated zone does not vary significantly and abruptly reduce to the hardness of the base material that can be an indication of in depth diffusion of nitrogen in the surface treated zone. 3.5. Wear resistance The weight losses of the worn untreated substrate, surface melted and surface nitrided samples against sliding distance are shown in Fig. 9. As it is shown, the weight loss after 1000 m sliding reduced from 4.2 mg to 2.9 and 1.2 mg, respectively, for the untreated aluminum, surface melted (M-1) sample (with highest hardness among surface melted samples) and surface nitrided (N-5) sample (with highest hardness among surface nitrided samples). According to these results, the wear rates of the surface melted and nitrided aluminum are about 69% and 28% of that of the untreated aluminum. Besides friction coefficients varied from 0.81 to 0.71 and 0.54 for untreated substrate, surface melted and surface nitrided samples respectively. It is obvious that grain refinement and presence of hard aluminum nitride phases have a positive effect on the wear resistance of the melted and nitrided samples. Similar improvement in wear resistance by formation of nitride layer has also been reported in oth er studies [11,13,14,16]. Fig. 10 shows the scanning electron micrographs of the worn surfaces of the untreated, surface melted and surface nitrided samples after 1000 m sliding distances. Plastic deformation including deep grooves on the worn surface of the untreated sample are signs of abrasive wear mechanism. After surface melting processing, the surface hardness increased and the intensity of the grooves reduced. Fig. 10c shows the worn surface of the surface nitrided sample, which has the lowest wear rate among the samples. Worn surface of the nitrided sample are smoother than those of the other samples that is due to presence of AlN phases in the nitrided layers and its higher hardness. Conclusions Liquid phase surface nitriding of Al-5052 by TIG process in ambient nitrogen atmosphere resulted in the formation of nitrided layers containing hexagonal AlN hard phase with two morphologies of dendritic and lamellar. Increasing the heat input resulted in reduced nitrogen content in the nitrided layer and thus AlN content reduced in the layer. The hardness of the surface nitrided layer reached to 1411 HV as compared to 52 HV for untreated aluminum alloy. Meanwhile, various volume percentages of nitrogen in the gas shielding had no significant effect on the morphology and the average hardness of the nitrided layers. Liquid phase surface nitriding resulted in the formation of AlN phase and hence increased the hardness of the treated layer that in turn reduced the wear rate to less than quarter of that of the untreated aluminum.