Microstructure and properties of low-carbon steels processed by high pressure torsion

• Main Author: Husain, Khaled Salman Adwan
• Other Authors: Gao, Nong
• Published: University of Southampton 2015
• Subjects: 620
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.680689

In this study, low-carbon steel (0.1 wt.% C) in ferritic-pearlitic (FP) and ferritic-martensitic (FM) initial states were processed by high pressure torsion (HPT) up to 10 turns under a pressure of 6 GPa at room temperature (RT). The HPT-processed FM samples were tempered at 150oC to 550oC. The microhardness monotonically increased with the number of turns up to 10 turns. After 10 turns, the microhardness had increased to 3 times that of the initial state for both the FP and FM samples. The microhardness of the individual phases, ferrite and martensite, in the FM processed samples increased with strain. The nanoindentation measurement of the ferrite in the FP processed samples revealed that the ferrite nanohardness increased with the number of turns up to 10 turns. The microhardness was maintained when the FM sample processed up to 4 turns and tempered at 450oC while the microhardness slightly increased when tempering at 350oC. Similarly, the ferrite in the FM sample processed up to 4 turns and tempered at 350oC had a nanohardness value higher than that of the sample tempered at 250oC: this can be attributed to the limited formation of cementite particles in the ferrite matrix. Optical and scanning electron microscopy (SEM) revealed that cementite fragmentation in the FP samples started after 4 turns of HPT processing and increased with the number of turns, or strain. SEM investigation revealed the formation of cementite particles when the FM sample processed up to 4 turns and tempered at 450oC: these cementite particles coarsened when tempering at 550oC. The transmission electron microscopy (TEM) investigation of the microstructure in the FM processed samples revealed that the microstructure was characterised by high internal stresses and dislocation density. The dislocation cells evolved during the HPT processing up to 10 turns and the fraction of high angle grain boundaries increased with the number of turns. X-ray diffraction (XRD) analysis revealed that no appreciable lattice parameter (a) expansion occurs during the HPT processing of the ferrite of the FP samples. The lattice parameter of the ferrite of the FM processed samples increased with the number of turns, or strain. The lattice parameter of the ferrite of the FM sample processed up to 4 turns and tempered at 450oC was equal to that of pure α-iron which suggests that the cementite particles formed at the expense of the decomposition of the supersaturated Fe-C solid solution in ferrite. It is well documented in the literature that the strength or hardness significantly increases with the number of turns during HPT processing. However, so far no attempt has been made to predict the strength of low-carbon steels after HPT processing. In this study, the X-ray line profile analysis (XLPA) using the multiple whole profile fit (MWP-fit) was used, as an alternative tool to TEM, to determine the dislocation density in the HPT-processed samples. A model that incorporates both the dislocation and grain boundary strengthening was proposed to predict the strength of the HPT-processed samples. In this model, other strengthening contributions such as solid solution and Orowan strengthening due to different structures such as pearlite, martensite, and carbide particles in the ferrite matrix were taken into account. The strength/hardness model proposed in this study was applied to predict the strength/hardness of six low-carbon steel (0.1 wt.% C) samples in the ferritic-pearlitic or ferritic-martensitic initial states processed by HPT up to 1, 4, and 10 turns. Furthermore, the strength/hardness model was also applied to predict the strength of FM samples processed by HPT and tempered at different temperatures. The strength/hardness model proposed in the current study can be used to predict the strength of all the investigated samples to a good accuracy with an average error of ±8%.