HAZ Phase Transformation and Thermal Damage for Laser Remanufacturing a High-Strength Stainless Steel

2.1. The effects of solid state phase transformation on the specific volume of martensite stainless steel

In the case that no phase transformation occurs for some generic solid material, the temperature rises will increase as the internal energy rises, so is the volume. On the contrary, the temperature and the volume decrease as the internal energy fall.

The specimens of the thermal expansion coefficient measurement were shown as Figure 1. The dilatometry experiment was processed in DIL801 dilatometer (Figure 2). In the temperature range (−150~1500°C), the maximum heating rate was 50°C/min and the maximum cooling rate was 25°C/min at low temperatures. In addition, the maximum heating rate rise to 100°C/min and the maximum cooling rate reached to 180°C/min as well. The errors of linear expansion coefficient was set as 0.03*10–6/°C, the gradient of temperature was 10°C/min. Using nitrogen as carrier gas, the flow rate was set as 20 ml/°C. The ambient temperature was 25°C and humidity was 35%.

The radial strain of the specimens for the thermal expansion coefficient experiments were used as measurement results. The strain was decomposed into two parts in the Ti (Range from room temperature to Ms): One part were the strain variations of martensite and austenite as temperature changes. The rest were caused by the volume effect of martensite phase transformation. It could be described as follows:

dealed as:

where ΔεMRTare the intercept of the extension line of austenite section during the cooling period in the temperature-strain curves and βM0are the strain caused by the volume effect of solid state phase transformation.

For FV520B high-strength stainless steel, the free expansion curve was shown as Figure 3. The critical temperature of transformation from pearlite to austenite (Ac1) was 600°C and the complete transformation temperature of austenite (Ac3) was 900°C. The starting temperature of the martensite phase transformation (Ms) was 130°C.

Shown as Figure 4, in the temperature between 130 and 600°C, the strain of martensite was 0.0057 and that of austenite was 0.088. The expansion coefficient of austenite was 18.72 and that of martensite was 12.13. In high temperature, the expansion coefficient of austenite was 21.5 and that of martensite reached up to nearly 19 at temperature over 600°C.

Figure 5 shows the temperature range of martensite transformation during the free dilatometry test. In the martensite phase transformation period, the strain caused by volume effect during phase transformation was 0.0067249 and the real strain should be 0.00328 considering the thermal contraction. Ideally, the material volume increased by 0.97% compared with that at Ms. In addition, the material volume at 320°C was equivalent to that at room temperature.

For materials in the temperature between M_s and A_c line, it could be martensite, austenite or dual-phase structure. The lower was the temperature, the higher was the density contrast of the two phase states. Consequently, the volume effect during phase transformation was more significant and the volume changes increased as the temperature drop. For the solid state phase transformation of the common steel, the transformation temperature of pearlite was the highest, next was bainite and that of martensite was the lowest [9]. In general, the volume effect of these phase transformations were corresponding to the transformation temperatures.

It had been noted above that superhigh heating rate and cooling rate had magnificent impact on the physical parameters of material during the laser remanufacturing. Studies indicates that the heating rate could reached over 500°C by methods of computer numerical simulation and temperature measurements. The simulation experiments were designed and conducted to study the effect of heating rate on the solid state phase transformation. Limited to the instrument condition, the heating rates were set as 0.25, 1, 5, 10, 30, 80 and 100°C/s. The results revealed that the rise of heating rates of each thermal cycles significantly increased Ac1 and Ac3 of FV520B (as shown in Figure 6). The expansion coefficient of martensite at 640°C increased apparently as the heating rate increased from 5 to 30°C/s. In general, heating rates had a great effect on the kinetics parameters of the austenite phase transformation.

2.2. The thermal cycles of laser cladding FV520B considering the effect of solid state phase transformation

In recent years, many studies were concerned about the effects of solid state phase transformation on the temperature and stress field and the mechanisms were revealed by experimental methods. However, the theoretical analysis were not enough in conditions of the complex projects like laser remanufacturing, which involved process control with many parameters. Therefore, computer numerical simulation was considered as a practical method to obtain precise, comprehensive and quantitative results, which had engineering meanings.

All the involved parameters were listed below: laser power, scanning rate, laser spot size and lapping rate were 1.5KW, 10 mm/s, 3 mm, 50% respectively. Using the technique of deactivate and reactivate element, the cladding material were activated by passes and layers. By building the coupling thermo-mechanical model, the evolution of the temperature field during the cladding was revealed. This work focused on the short and varying thermal cycles in positions with different depth in HAZ during the single-pass and multi-layer claddings. Figure 7 was the geometries of the laser cladding samples during the single-pass and multi-layer processes [12, 14, 15].

Figure 8 was the temperature evolution at midpoint of single-pass cladding sample. The temperature of the melting pool reached 1833°C after 3.5 s. Meanwhile, the temperature measurement experiment at the midpoint was conducted by using RaytekMM infrared thermometer. The results showed that the laser beam scan midpoint of the single track after 3.7 s and the temperature of the melting pool was 2175°C, which was consistent with the simulation results.

Figure 9 was the thermal cycles in HAZ during the single-pass cladding. The maximum temperature in regions close to surface of the work-piece reached over 1000°C, which is higher than the complete austenitic temperature Ac3. In addition, the measurement results of the heating rate was around 500°C/s and that of the cooling rate was over 100°C. For regions in different depth of HAZ, the maximum temperatures of the thermal cycles were the main difference, which could directly affected the final microstructure in HAZ. Meanwhile, the simulation results could be provided as evidence for the studies on the mechanism of the microstructure evolution.

Figure 10 was the thermal cycles in HAZ during the multi-layer cladding. In multi-layer laser remanufacturing process, the coating is deposited layer by layer and HAZ was created repeatedly [16]. In addition, the repeatedly thermal effect on microstructure in HAZ was hard to control and predict. Some studies pointed that the microstructure in HAZ experienced approximate tempering during the repeatedly thermal cycles and toughening effect on microstructure was observed. The numerical simulation results showed that microstructure of HAZ experienced thermal cycles with over 600°C maximum temperature during the subsequent deposition process after the first layer cladding. It was considered that the multi-cycle heating and cooling process could affect the final microstructure and mechanical properties in HAZ.

2.3. The continuous cooling transformation (CCT) curve of FV520B

Figure 11 was the schematic of phase transformation for the martensite stainless steel during the laser remanufacturing process. Austenite was assumed to be the initial phase during solidification process. In addition, the martensite transformation occurred during the cooling process due to the high cooling rates. As temperature decreased, the martensite transformation started at Ms and finished at Mf. When a new layer of FV520B was deposited, the previously deposited material experienced a new thermal cycle. Once the rising temperature was higher than Ac1, the transformation of martensite to austenite occurred. The percentage of austenite phase increased linearly as temperature rise.

As the physical properties of material depended on temperature and phases, the accuracy of the physical properties based on the appropriate description on the evolution of phase. In laser cladding process, different regions had different thermal cycles and external load, which could contribute to the change of the kinetics and effect of solid state phase transformation. Therefore it was necessary to carry out systematical research in the influence of cooling condition on the solid state phase transformation during the cooling process [17].

A combined experiment by methods of dilatometry and metallography was conducted and the CCT curve was obtained by the L78 RITA transformation measuring apparatus. According to the thermal cycles of the laser cladding processes, two austenitic procedures were designed to study the effect of the austenitization condition on the microstructure evolution. The process of case 1 was designed as below: the sample heated up to 1000°C at heating rate of 300°C/s and kept the temperature for 5 s. In the cooling period, the sample was cooled to 810°C at the cooling rate of 100°C/s, kept the temperature for 5 mins and then was cooled at cooling rate of 0.02, 0.08, 0.2, 0.5, 1 and 50°C/s respectively. For case 2, the sample heated up to 1300°C at the same heating rate, was kept the temperature for 5 s and cooled to 810°C at the same cooling rate. Then, the sample was cooled at cooling rate of 0.02, 0.08, 0.2, 0.5, 1°C/s respectively. The phase transformation temperatures were obtained using tangent method.

Figure 12 was the continuous cooling transformation of FV520B stainless steel. Figures 13 and 14 were metallograph of microstructures in different austenitic procedures. The results showed that the microstructures were all martensite in the designed cooling parameters due to the hardenability of FV520B stainless steel. For case 1, the microstructure was lath martensite with finely distributed precipitates. In addition, the proportion of martensite with sparse laths increased with the increasing cooling rate in the process. For case 2, the microstructure was lath martensite as well. However, the microstructure of martensite in case 2 was coarser than that of case 1 and dispersedly distributed precipitates were barely observed. As shown in Figures 15 and 16, the phase transformation temperature of case 2 was 20°C lower than that of case 1 and both increased as the cooling rate became higher; the microhardness of case 2 was 60 HV higher than that of case 1 and both decreased as the increasing cooling rate. In general, the maximum temperature of thermal cycle had a dominating effect on the microstructure, microhardness and phase transformation temperature rather than cooling rate.

It was considered that the maximum temperature was 1000°C with short heat preservation in case 1 and the sample experienced complete austenitic process with precipitates partly dissolving in the matrix. In addition, the precipitates further grew coarse and solubility in the matrix decreased, which eventually contributed to the decreasing microhardness. However, the maximum temperature of case 2 reached 1250°C and no ferrite existed at high temperature. In general, the sample of case 2 experienced complete austenitic process with precipitates completely dissolving in the matrix. The increasing solid solubility of alloy elements contributed to the higher microhardness than that of case 1. Meanwhile, phase transformation temperature of case 1 and case 2 were both low and it could caused by the strengthen effect of austenite and growth of precipitates during the cooling period at low cooling rate.

The microhardness of laser remanufacturing specimens was higher than that from CCT process due to the superhigh heating rate in the actual laser cladding process. According to the numerical simulation results, the temperature gradient in high temperature period reached to 1000°C/s during the laser cladding process and the maximum heating rate could be set as 500°C/s due to the restrictions of instrument conditionFigures 15 and 16.

As above, austenitic process and cooling rate had a remarkable impact on the phase transformation temperature and it could be interpreted as the effect of thermal cycle on the valid elements of material composition. It was assumed that the maximum temperature and heating rate during the heating period in the thermal cycle had great influence on the martensitic phase transformation temperature. Moreover, the strengthen effect and reducing of Ms took place due to the low cooling rate during the same austenitic process. In general, the thermal cycles had a significant effect on the metallographic transformation [18, 19].