PSI - Issue 23

Hugo Wärner et al. / Procedia Structural Integrity 23 (2019) 354–359 Hugo Wärner / Structural Integrity Procedia 00 (2019) 000 – 000

355

2

1. Introduction

To satisfy the need for more sustainable power generation, the efficiency of biomass-fired power plants could be increased and this can be achieved by increasing the temperature and pressure in the boiler sections (Yin and Wu 2009;Viswanathan et al. 2006). In addition, a more flexible generation of power is fundamental if only renewable power generation is to be achieved and this will increase the number of start and stop cycles (Dietrich et al. 2013). These changes in operating conditions will increase the demands on the materials in the critical components of the power plants. Cyclic operating condition in a long-term high temperature environment is a process that such materials must withstand in order to satisfy the needs for the future power generation. The lifetime of power plants is expected to be 30 years or more and therefore the materials used for critical components need to have good long-term high temperature performance in order to maintain structural integrity and fulfil the safety requirements (Viswanathan et al. 2006;Viklund et al. 2013;Sourmail 2001). Austenitic stainless steels are commonly used for the critical components of power plants (Yin and Wu 2009;Viswanathan et al. 2006). One way to verify the demands on safety and structural integrity for long-term usage is to investigate the cyclic high temperature properties, like the thermomechanical fatigue (TMF) behaviour, of the materials used in new efficient biomass-fired power plants (Hormozi et al. 2015). This study investigates the TMF response of three commercial austenitic alloys and focus on the influence of long term service conditions, i.e. only pre-aged material was studied. A relatively high temperature range of 100 °C to 800 °C was used in order to simulate the effect of increased efficiency efforts of power plants. By the use of finite element analysis (FEA) and low angle grain boundary (LAGB) calculations by electron backscatter diffraction (EBSD), the evolution of inhomogeneous deformation in the mechanical tested specimens was investigated. 2. Experimental work and Materials The investigation involved three commercial austenitic alloys: Sanicro 25 (solution heat- treated at 1220 °C for 10 minutes), Sanicro 31HT (solution heat- treated at 1200 °C for 15 minutes) and Esshete 1250 (solution heat -treated at 1100 °C for 15 m inutes). Sandvik Materials Technology AB have provided and heat-treated the materials. The chemical compositions of the investigated materials in wt.% are given in Table 1. The data used for the material modelling were: E RT = 197 GPa, RT = 700 °C = 0.27, y , RT = 382.3 MPa, E 700 °C = 145 GPa and y , 700 °C = 378.5 MPa. The test procedure employed was strain controlled thermomechanical fatigue (TMF) testing with 5 minutes dwell time at maximum mechanical strain range (Δε mech, max ). The test machine was a servo-hydraulic TMF machine from Instron with induction heating and forced air-cooling. Before the TMF tests, the machine was carefully aligned to prevent buckling and other instability effects, according to "the validated code of practice" by Hähner et al. (2006). This study includes the In-Phase (IP) and the Out-of-Phase (OP) cycle with R ε = 0 and R ε = - ∞ respectively. The temperature range used was 100 °C to 800 °C and the heating and cooling rate was 5 °C/s. The specimens were aged at 800 °C for 2000 hours before testing in order to simulate microstructural degradation from extended service time. The setup of the tests were done according to Hähner et al. (2006), with spot welded thermocouples at a distance of 1 mm between each other and the strain was measured with a high-temperature extensometer with a gauge length of 12.5 mm. The thermal strain was subtracted from the measured total strain, so that the mechanical strain could be controlled as suggested by Hähner et al. (2006). The number of cycles to failure ( N f ) was defined as the point at which the stress range (Δσ) decreases 10 % below the tangent line constructed at t he last point of zero curvature (Hähner et al. 2006). However, the test was not stopped until a load drop of 60 % occurred.