Conclusions
Due to the periodic boundary conditions applied to the simulation
box, vacancies in the lattice were created, and so the desired effect
of segregation at the twist grain boundary was not observed. Instead,
the Au atoms segregated to the vacancies. To counteract this problem
the way in which the boundary conditions are imposed would have to be
changed. One possibility would be to fix the boundary conditions to a
perfect fcc lattice outside the box. However, this would require
increasing the box size to avoid finite size effects. In turn this
would result in a slower simulation which could be undesirable.
When comparing the results obtained from the EAM model and from L-J
potentials it is apparent that the same segregation behavior was
described by each. However, since this behavior occurred at lattice
vacancies as opposed to a grain boundary, one may not conclude that
the two models would also be in agreement for segregation at a grain
boundary. Nevertheless, it does show that in some cases a simple
approximation (like pair-potentials) to a many-body problem is
sufficient to explain the behavior of physical systems.
Project Home Class Homepage