Page 90 - Read Online

Mooraj et al. J Mater Inf 2023;3:4 https://dx.doi.org/10.20517/jmi.2022.41 Page 15 of 45

suited to study the nucleation and evolution of defects such as vacancies, dislocations, grain boundaries, and
twinning [115-117] .

In the past, MD simulations have also been used to explore phase transformation, mechanical behavior,
nucleation and crystallization processes within HEAs [118-120] . MD simulations can study much larger systems
with faster computation times than ab-initio calculations because they use classical Newtonian mechanics
versus the quantum mechanical interactions on which ab-initio methods are typically based. They can also
accurately simulate non-equilibrium systems due to the rapid time scales over which a simulation is
[113]
conducted . Despite these impressive advantages, a known weakness of MD simulations is that their
accuracies depend heavily on the accuracy of the potential energy functions used to define them. However,
these potential energy functions must first be measured by experimentation or calculated via ab-initio
methods, which can limit the applicability of MD simulations to novel systems that have not been studied
[113]
before . This section presents works that take advantage of the strengths of MD simulations to explore
large composition and application spaces with relatively low computation times.

As previously discussed, many computational methods can be used to investigate and predict material
properties, such as yield strength, hardness, and phase formation. However, MD simulation has the added
benefit of allowing researchers to investigate deformation mechanisms within an alloy via simulation of
[118]
atomic motion under various ambient and loading conditions . This ability is especially important as it is
very difficult and laborious to observe plastic deformation processes under experimental [118,121] . Pan et al.
applied atomic-scale tensile MD simulations to a Fe Mn Co Cr alloy system to investigate
80-x x 10 10
transformation-induced plasticity (TRIP) and twinning-induced plasticity (TWIP) mechanisms in this
[121]
system . In this work, the atomic fraction of Mn, strain rate, and grain size were all adjusted to investigate
[121]
each variable’s effect on the system’s deformation mechanisms . Figure 7A shows a schematic illustration
of the model where green dots represent the FCC phase and white dots denote grain boundaries. The FCC
transforms into BCC and HCP during deformation, and this transformation was found to be most prevalent
when x = 40. The addition of Mn also reduced the stacking fault energy, which facilitated twinning during
deformation, leading to improved strain hardening. Interestingly the transformations and twinning
mechanisms were suppressed for smaller nano-grain sizes, which Pan et al. attributed to the transformation
from the intragranular evolution mechanism at larger grain sizes to the intergranular evolution mechanism
at smaller grain sizes. This study shows the potential of MD simulations to explore compositional space and
to provide a detailed analysis of deformation mechanisms before significant investments in experimental
characterization.

MD simulations can be used to investigate the relationship between the stacking fault energy and
strengthening mechanisms within an alloy system. Understanding this relationship can then provide
guidelines for designing new HEAs with tailored properties and deformation mechanisms suited to specific
[122]
applications . Jarlov et al. performed MD simulations using the Large-scale Atomic/Molecularly
Massively Parallel Simulator (LAMMPS) to investigate the effect of the chemical composition in the Co-Cr-
[122]
Fe-Ni alloy system on the generalized stacking fault energy (GSFE) . The authors used this method to
explore the system’s strengthening and deformation mechanisms during tensile tests. Figure 7B shows the
simulated cell, and the planes marked as I, II, and III indicate the planes displaced during the tensile
simulation. Based on the simulations, it was found that increasing Ni and Co contents led to an increase in
the energy required to introduce stacking faults and deformation, while increasing Cr and Fe contents led to
a decrease in the energy required to introduce these defects. When carrying out tensile simulations of the
various compositions, it was found that the yield strength correlated linearly with the energy required to
introduce intrinsic stacking faults. Thus, the strongest composition was identified as (CoCrNi) Fe ,
90 10