Grid'5000 experiment

Jump to: navigation, search

study of the ternary phase diagram Fe-Al-C (Networking)

Conducted by

Damien Connétable

Description

It has been largely reported that aluminium-alloyed steels can be hardened by a dispersion of kappa-Fe3AlC carbides. The micro-structure of such steels has been recently investigated in details. However, to control the precipitation micro-structures of steels and also to master the carburising process of Fe-Al alloys, it is necessary to rely on the thermodynamical properties of the iron-rich phases in the Fe-Al-C system, and to know the fundamental properties of these phases. However few and contradictory experimental and theoretical informations are present in the literature about the kappa carbide. As a matter of fact, if the phase equilibrium has been studied in high Mn containing steels, no detailed description of the iron-rich part of the Fe-Al-C phase diagram is available in the literature. First at all, we want to complete the CALPHAD databases (”Calculation of Phase Diagrams”) used to describe the ternary phase diagram Fe-Al-C. CALPHAD method consists in an optimization of the physical parameters on experimental data. Among these parameters, the evaluation of the formation energies of different phases in equilibrium in the phase diagram is needed. Ab initio calculations have been then used to determine them. In a second stage, we want to understand why experimenters don’t identify stoichiometric kappa-Fe3AlC structure but an off stoichiometric system Fe3AlC(1/2). In this way, we have studied the evolution of the formation energy of Fe3AlCx when x varies between 0 to 1.

Status

achieved

Resources

  • Nodes involved: 20
  • Sites involved: 1
  • Minimum walltime: 8h
  • Batch mode: yes

Tools used

No information

Results

We have already finished to generate the CALPHAD database for the Fe-Al-C phase diagram. It was necessary to calculate the formation energies of the different ordered systems in equilibrium in the phase diagram (kappa-Fe3AlC, Fe3Al-L12, Fe3Al-D03, L10-AlC etc.), but we have too simulated desorded binary FexAl(1−x) structures. To simulate such desorded FexAl(1−x) binaries (fcc and bcc structures), we have employed the SQS (‘Special Quasirandom Structures’) framework as proposed by A. Zunger. A Calphad description of the system has been then assessed which is compatible with general databases. We have too finished to study the evolution of the formation energy of the kappa phase with respect to the carbon concentration, and we shown that in accordance with experimental facts, Fe3AlC(1/2).

Shared by: Damien Connétable
Last update: 2008-06-17 11:58:52
Experiment #293

Personal tools
Namespaces

Variants
Views
Actions
Public Portal
Users Portal
Admin portal
Wiki special pages
Toolbox