SOME STUDIES ON EFFECTS OF HEAT TREATMENT AND HYDROGEN EMBRITTLEMENT IN P92 STEEL WELDS

Abstract

The Creep strength enhanced ferritic (CSEF) steels are developed to meet the demand by power generation organizations to increase efficiency by operating at higher temperature and pressure. The microstructural evolution and mechanical properties of CSEF steels are affected by various parameters; normalizing temperature is one of them. Grade P92 steel is considered as a candidate material for thermal and nuclear power plants operations at temperatures of up to 650 °C. In the present work, the effect of normalizing temperature on the microstructural evolution and mechanical properties of cast and forged (C&F) P92 steel was carried out. The C&F P92 steel was subjected to various normalizing temperatures (950 °C -1150 °C) and followed by a fixed tempering temperature (760 oC). Considering observation for the optimum combination of strength, ductility, and toughness, the normalizing at 1050 oC and tempering at 760 oC has been suggested for C&F P92 steel. The Laves phase evolution (at 650° C) with varying aging time (upto 3000 hours) in P92 steel have been performed and their effect on mechanical properties have been investigated. The strength and ductility decreased as a result of deprivation of solid solution strengthening and formation of Laves phase. The hardness of P92 steel was also affected by aging time but less pronounced as compared to strength. Charpy toughness was also reduced continuously with increase in aging time as a result of thermal straining of particles and Laves phase formation. In the other work, the formation of Laves phase during aging in P92 steel and an effort to dissolve Laves phase using re-austenitization and tempering treatment was carried out. The as-received material is exposed to 620 oC for 4560 hrs. After aging, the specimens are subjected to austenitization at 1050 oC for 1 hr and followed by tempering at 760 oC for different times (2 hrs and 4 hrs). After re-austenitization and tempering, the dissolution of Laves phase is observed. A model is also suggested to quantify the dissolution of W containing Laves phase. Thermo Calc is used to predict driving forces for precipitation of Laves and M23C6 phases. The predicted driving force map shows that the nucleation and dissolution of Laves phase can be controlled at different temperatures for varying composition of the steel. The double austenitization based normalizing and tempering (DNT) heat treatments have been performed and their effect on tensile behavior and microstructural morphology of P92 steel have been carried out and compared with conventional normalizing and tempering (CNT) heat treatment. The CNT treatment deals with the normalizing at 1040 oC/60 min followed by air cooling and tempering at 760 oC/120 min, followed by the air cooling. The vi DNT treatment deals with the initial austenitizing in the temperature range of 950-1150 oC for 60 min followed by the water quenching and second austenitizing was performed at the 1040 oC for 60 min followed by the air cooling. After the double austenitization based normalizing, tempering was performed at 760 oC for 2 h followed by the air cooling. The fine grain structure was obtained for the DNT 3 treatment. The results of tensile tests were also obtained to be superior for the DNT treatment as compared to CNT treatment. The effect of preheating on the weld bead geometry have been performed in autogenous gas tungsten arc welding (GTAW) and a relation between preheat temperature and thickness of welded plate using Rosenthal 2-D heat equation have been developed for thin plates. The effect of cooling rate on the retention of δ-ferrite was also studied in P92 steel welds. The full penetration was achieved with preheat temperature of 300 oC. A relation between preheat temperature and plate thickness was correctly predicted and a good correlation was found with our experimental data. The volume fraction of δ-ferrite was measured lowest with highest preheat temperature (300 oC) or lowest cooling rate. The dissolution of δ-ferrite in the weld fusion zone using post-weld normalizing and tempering (PWNT) have been investigated and their effect on mechanical properties have also been performed. The autogenous GTAW was used to prepare double-sided welded joints of P92 steel. After welding, post-weld heat treatment (PWHT) at 760 oC for 2 hrs and post-weld normalizing at 1050 oC for 1 hr followed by tempering at 760 oC for 2 hrs were performed. The morphology, composition, structure and hardness of δ-ferrite are verified. In as-welded and PWHT conditions, the retention of δ-ferrite was observed while the dissolution of δ-ferrite was confirmed after PWNT. Dissolution kinetics of δ-ferrite is quantified using Thermo-Calc for Scheil’s solidification calculation and DICTRA for calculating diffusion coefficients of Cr and W at 1050 oC in austenite. In the other work, P92 steel weld joints were prepared using multi-pass shielded metal arc welding (SMAW) process using E911 and P92 filler. A comparative study was performed on microstructural evolution, tensile strength, microhardness and Charpy toughness across the P92 steel weldments in as-welded and PWHT condition. In as-welded condition, transverse tensile specimens were fractured from the fine grain heat affected zone or inter-critical heat affected zone (FGHAZ/ICHAZ) and after PWHT, the fracture location was shifted to over tempered base metal from FGHAZ/ICHAZ. The minimum required Charpy toughness of 47 J (EN1557: 1997) was achieved after the PWHT for both E911 and P92 filler. vii The effect of PWHT and PWNT on mechanical properties of dissimilar SMAW welded joint of P911 and P92 steel using P911 filler have been performed. PWHT of weldments resulted in a negligible increase in ultimate tensile strength and yield strength while a significant change was measured after PWNT. The PWNT resulted in minimum hardness gradient and maximum homogenization of microstructure across the dissimilar weldments. The room temperature Charpy toughness value was also found to be maximum for PWNT treatment. In other work, P91 and P92 dissimilar multi-pass SMAW welded joints have been produced using the different filler rods. The microstructure of the welded joint has been studied for the different filler composition in various heat treatment condition. The room temperature Charpy impact toughness and tensile properties for the different welded joints (different filler) have been also studied and related with the microstructure of the welded joint. From the results, it has been concluded that a higher amount of ferrite stabilizer in P92 filler promotes the formation of the δ ferrite in the weld fusion zone. Granjon implant test and mercury method (for diffusible hydrogen measurement) have been performed on Grade P92 steel welded specimens to study the effect of welding parameters as well as filler conditions on diffusible hydrogen levels and their subsequent effect on hydrogen-assisted cracking (HAC). The three different welding conditions and three different filler conditions are used to measure the diffusible hydrogen level in the deposited metal. Granjon implant test was performed to evaluate HAZ-HAC susceptibility with similar welding conditions which were used in the mercury method. Lower critical stress (LCS) was also evaluated using Granjon implant test. The higher susceptibility of P92 steel welded plate towards HAZ-HAC was observed in case of lower heat input or higher diffusible hydrogen content. However, by considering LCS, fracture mode, embrittlement index, and diffusible hydrogen content, the weld deposited using highest heat input offers great resistance to HAZ HAC. The P92 steel plate welded by the contaminated electrode with a high level of diffusible hydrogen was found to be more susceptible to HAC.