Integrated computational alloy design, single crystal growth, and characterization of nickel base superalloys

Abstract

This dissertation is formed by a part of work conducted in a TUBITAK project (112M783). In the dissertation, first alloy design of a third generation superalloy, and then its production as single crystals, heat treatments, and creep tests are investigated. A combined Neural Network (NN) – PHAse COMPutation (PHACOMP) – CALculation of PHAse Diagrams (CALPHAD) method is applied to alloy development of single crystal Ni base superalloys with low density and high creep resistance. PHACOMP method is used for estimation of the volume fraction of the γ’ and a parameter named Md which is an index showing the propensity of the alloys towards formation of the TCP phases. Neural network is used for modeling the density and rupture time by training and testing a network with a set of the known experimental alloy compositions. Modeling results is combined with data obtained from PHACOMP to render very useful scatter plots for the effect of alloying elements; stress, temperature and volume fraction of the γ’ phase on density, rupture strength and formation of TCP phases in the Ni base single crystal superalloys. A third generation alloy (ERBALLOY) was designed and produced by two methods, vertical Bridgman (VB), and vertical Bridgman with a submerged baffle (VBSB). The effect of low melt height on solidification characteristics of the alloys is studied. Evolution of the phases is simulated by CALPHAD bases Thermo-Calc software. The solution and aging heat treatment of the alloys are modeled with Dictra and TC-Prisma software. Characterization of the microstructure is performed by optical, scanning transmission electron microscope (SEM), and electron probe microanalysis (EPMA). The creep behavior of the ERBALLOY is tested at an intermediate and a high temperature and showed reasonable agreement with NN results. The approach used in this study is in line with the Materials Genome Initiative (GMI) and Integrated Computational Materials Engineering (ICME), and it can be applied for designing low density-creep resistant single crystal superalloys for critical parts of the aircraft/gas turbine engines.