Investigations on friction stir processing of magnesium based AE42 alloy

Abstract

Mg alloys have vast applications in the automobile industry due to their attractive properties such as outstanding stiffness/weight ratio, excellent castability, good damping capacity and easy recyclability. Mg alloys such as AZ91, AZ31 and AM50 are widely used for automobile components such as cam covers, baffles, oil adaptors, clutch housings and roof frame assemblies. However, these alloys have limited opportunity in the powertrain applications such as automatic transmission cases and engine blocks owing to the degradation of their mechanical properties at high temperature. Due to their inferior mechanical properties such as hardness, the tribological performance of Mg based alloys is also not adequate for such applications. A die casting alloy that was developed for high temperature applications is AE42. This alloy contains 4 % Al and 2 % rare earth (RE) elements. The alloy is characterized by the presence of elongated precipitates of β-Mg17Al12 and Al11RE3 as the major phase along with Mg with an average grain size of nearly 20 μm. These precipitates are formed in-situ during the transformation cooling of the alloy. The presence of elongated precipitates and the coarse grain structure of the AE42 alloy have been reported to be detrimental to its mechanical properties such as yield strength and hardness. Ultra-fine grained (UFG) materials are known for their superior mechanical properties in accordance with the Hall-Petch equation. Therefore, grain size refinement of the material provides one of the possibilities by which material strengthening can be achieved through grain boundary strengthening. The severe plastic deformation (SPD) processes have successfully emerged as one of those processes that can develop UFG structure in the materials in the bulk form. The most popular and successful SPD processes are equal channel angular extrusion (ECAE), high pressure torsion (HPT), accumulative roll bonding (ARB) and friction stir processing (FSP). Amongst all these process, FSP is the most recent SPD process that has evolved as the successful and versatile microstructural refinement process. In the current study, the influence of FSP on the microstructural, mechanical, corrosion and tribological properties of AE42 alloy has been investigated. A specially designed fixture was fabricated to clamp the AE42 samples on a vertical CNC milling machine for performing FSP. The fixture had a hollow rectangular box configuration and was connected to an external chiller unit for undersurface cooling of the friction stir processed (FSPed) specimens. The medium used for cooling was methanol and it was circulated at -20° C. The in-depth studies of the microstructural changes that took place during FSP of the investigated alloy were done using SEM/EDS, XRD, TEM, FIB-SIMS and EBSD analyses. The microstructural observations revealed the occurrence of significant microstructural refinement in the FSPed region. The elongated precipitates were fragmented and some new in-situ precipitates were generated. The precipitate size got further refined with more number of FSP passes. It was found to vary from 200 nm – 1 μm range in single pass FSP to as fine as 50 nm with double pass FSP as compared to about 80 – 90 μm size precipitates in the as-cast alloy. The grain size also got considerably refined during FSP, the extent of which was found to vary with the FSP parameters. TEM, FIB-SIMS as well as EBSD analysis of the FSPed alloy also showed the presence of submicron grains in the FSPed alloy, the percentage of which further increased with increase in number of FSP passes supplemented by undersurface cooling. The microstructural refinement resulted in nearly 60 % enhancement in the hardness of the FSPed alloy. The major factor that contributed to the enhanced hardness of the FSPed alloy was found to be grain size strengthening. EBSD results showed that the grain size varies along the depth of the specimen cross-section. The grain size was found to be more refined with increase in distance from the FSP tool shoulder. The EBSD results also confirmed the occurrence of texture variations across the depth of the FSPed alloy. For determining the most influential FSP parameters effecting the properties of FSPed AE42 alloy, the FSP parameters were varied over a range and a systematic evaluation of these parameters using Taguchi’s DOE technique was done. The most influential parameters were found to be cooling temperature, FSP tool rpm and number of FSP passes. FSP specimens were processed at the optimized set of FSP parameters and mechanical properties were determined using tensile testing, bulkhardness testing and scratch test. The FSPed specimens sowed enhanced scratch resistance and enhanced hardness. The other mechanical properties such as yield strength and ultimate strength got considerably enhanced without comprising the ductility. Instead, the FSPed specimen showed a little improvement in the percentage elongation also. The microstructural studies of the fractured surfaces showed brittle and ductile failure for the as-cast and FSPed AE42 alloy respectively. Further, the FSPed alloy also exhibited greater strain hardening as compared to the as-cast alloy. Attempt has been made in the current study to simulate the thermal history during FSP using a novel technique of developing finite difference equations (FDE’s). The crosssection of the FSPed specimen was divided into discrete number of nodes in the x and y directions. FDE’s were written for each of the nodes and were expressed in the matrix form separating the variables, constants and the temperature terms. The major source of heat during FSP was considered to be the frictional heat between the FSP tool shoulder and workpiece. The variation of axial load at different FSP parameters, required for evaluation of heat input, was determined using a tool dynamometer. A generalized MATLAB code was written for the simultaneous solution of all the FDE’s. The code was generalized in terms of number of nodes and the FSP parameters such as FSP tool rpm, linear speed and axial load. The validation of the numerical simulation of thermal history was done by comparison with the experimental temperature values, determined at different FSP parameters. The actual temperature values were determined using K type thermocouples and data acquisition system. The numerical simulation was found to predict the thermal history at different FSP parameters to a fairly acceptable accuracy with error ranging from nearly 2 to 9 %. Prediction of thermal history using numerical simulation approach assumed that the heat input across the tool shoulder is constant. The effect of viscous heat dissipation was also neglected. However, the heat input varies across the tool shoulder. To consider the influence of variable heat input, effect of viscous heat dissipation and determination of certain other important parameters such as velocity and strain rates, FSP simulation was also done using FLUENT software. The thermal history predicted using numerical simulation was further used for the estimation of the grain growth rates and final recrystallized grain size of the FSPed AE42 alloy under different FSP conditions. The corrosion behaviour of the as-cast as well as FSPed AE42 alloy was investigated using corrosion immersion and electrochemical corrosion tests in 3.5 % NaCl solution. The first phase of the analysis involved determination of set of FSP parameters which can induce maximum corrosion resistance in the AE42 alloy using DOE technique. The second phase of the analysis involved elaborate study of the as-cast and the FSPed specimen that has shown higher corrosion resistance in the first phase. The immersion carrion tests were performed for a period of 48 hours. Weight loss after different time intervals was used for calculating the corrosion rates. The corroded surfaces were analyzed using SEM/EDS. Electrochemical corrosion studies viz potentiodynamic polarization, Rp/Ec and electrochemical impedance spectroscopy (EIS) were conducted on as-cast as well as FSPed AE42 specimens. The tests were performed on a Gamry’s electrochemical workstation using a three electrode cell consisting of saturated calomel electrode (SCE) as a reference electrode, graphite as a counter electrode and as-cast/FSPed AE42 specimen as a working electrode in a 3.5 % NaCl solution. The FSPed alloy showed a higher corrosion resistance in comparison with the ascast alloy. Further, the corrosion rate was found to decrease with decrease in the grain size of the FSPed alloy. The stability of oxide layer in 3.5 % NaCl solution was attributed to the presence of uniformly distributed fine in-situ precipitates. The tribological performance of the as-cast and FSPed AE42 alloys was evaluated using pin-on-disc wear tests. The tests were performed at different sliding velocities and normal loads, which varied from 0.33 m/s to 3 m/s and 5 N to 20 N respectively. The tests were performed for a constant sliding distance of 2.5 Km. Worn surfaces and wear debris were analyzed using SEM and EDS for the determination of different wear mechanisms. The subsurface analysis of the worn surfaces revealed higher work-hardening capability of the FSPed alloy. The FSPed AE42 alloy demonstrated atleast 23 % decrease in the wear rate in comparison with its as-cast counterpart, which may be attributed to the microstructural refinement resulting in enhanced hardness and ductility of the FSPed alloy along with higher work hardening capability. At low loads, wear mechanism transformed from oxidation and abrasive wear at low sliding velocity to delamination at high velocity. At intermediate loads, oxidation and abrasion characterized the worn surface at low velocity, whereas delamination and plastic deformation were found to be major wear mechanisms at high velocities. At high loads, the corresponding mechanisms were abrasion, delamination and plastic deformation at low velocity, whereas severe plastic deformation and delamination at high velocities. Based on the overall results of the present study, a set of FSP parameters may be recommended for the given Mg-based AE42 alloy, which comprises tool rotational speed of 700 rpm, a cooling temperature of -10° C, a linear speed of 60 mm/min and FSP passes as 3. With these parameters, the bulk hardness of the alloy could be enhanced by 60 %, whereas a reduction of atleast 23 % in the wear rate could be achieved. Moreover, the corrosion rate of the alloy in 3.5 % NaCl solution could be reduced by 50 % by its FSP at these parameters.