Henrik Larsson

Coupled modelling of solidification and solution heat treatment of advanced single crystal nickel base superalloy

Abstract Two numerical models to simulate microsegregation and phase formation during directional solidification and subsequent solution heat treatment of an advanced experimental ruthenium containing single crystal nickel base superalloy are tested and compared. The first method is based on a one-dimensional front tracking for the primary solidification and a homogenisation method for the final stages of solidification as well as for the solution heat treatment. Calculations for this model are carried out in one-dimension using cylindrical coordinates. The second is based upon the phase field method, applied to solidification and subsequent solution heat treatment, where calculations are carried out in two-dimension. Both models are coupled to thermodynamic and kinetics databases modelled according to the CALPHAD method. A concept of computer based optimisation of solution heat treatments is proposed. The results show that both methods are capable of handling the complexity of contemporary superalloys, and realistic results are obtained from both models.

A random-walk approach to diffusion controlled growth

Abstract The new method does not require formulation of any special conditions at the moving phase interface. Although only binary systems are treated at present an extension to any number of components seems straightforward. A good agreement with conventional front tracking techniques (DICTRA) is found.

Diffusion Process Simulations - An Overview of Different Approaches

Some different approaches to diffusion process simulations are briefly presented. Their varying areas of applicability are discussed. Example simulations using the phase-field method, the DICTRA software, and random-walk based approaches are presented.

Simulation of Coupled Carbonitriding and Internal Oxidation of Steel

Abstract Carbonitriding and internal oxidation have been simulated simultaneously using the DICTRA homogenization model. The predicted carbon and nitrogen profiles as well as the oxide and nitride phase fraction profiles agree favourably with experimental data obtained from the literature, though some discrepancies exist. The present approach should be a very useful tool in process control and optimization.