# Are there other kinds of stability of materials? How to demonstrate that with computation approaches?

I have known some kinds of stability of materials, such as:

• Mechanical stability;
• Dynamical stability;
• Thermal stability

In particular, they can be studied with first-principles computational methods:

• Mechanical $$\Rightarrow$$ Elastic constants $$\Rightarrow$$ Born criteria.
• Dynamical $$\Rightarrow$$ Phonon spectrum $$\Rightarrow$$ Is there negative frequency?
• Thermal $$\Rightarrow$$ Ab-initio molecular dynamics $$\Rightarrow$$ Structure distortion.

Then are there other kinds of stability of materials? How to demonstrate that with computation approaches?

• Very interesting question. I can only think of metastable structures but that isn't a type of stability. – Hitanshu Sachania Nov 28 '20 at 15:11
• Would you count radiation stability as a separate item? – Anyon Nov 28 '20 at 19:57
• @Anyon You can list it. – Jack Nov 28 '20 at 23:50
• @Anyon Since this question has gone unanswered for 3 months and I'm trying to shorten the unanswered queue, I wonder if adding an answer about radiation stability and the computation approach used to study it, might be something you'd be able to do? – Nike Dattani Feb 28 at 19:39
• @NikeDattani I gave it a go. The topic is pretty far from my specialty though. – Anyon Feb 28 at 21:58

In some applications or environments (e.g. fission/fusion reactors, space, sterilization of packaging), radiation effects are highly important and can cause significant damage to, or changed properties of, materials. In other cases, radiation-induced changes of properties are in fact desired. For example, polymers are often irradiated to induce crosslinking for superior qualities. Now, both long-term radiation stability of e.g. reactor materials, and such manufacturing methods can be modeled. However, accurately modeling these radiation processes can be very challenging, because the problem can involve everything from initial defect formation to dislocation dynamics to continuum mechanics.

To quote a review article, J. Knaster, A. Moeslang & T. Muroga, Materials research for fusion, Nature Physics, vol. 12, 424–434 (2016),

The effects of irradiation on a material’s microstructure and properties are a classic example of an inherently multiscale phenomenon, as schematically illustrated in Fig. 3a. Length scales of relevant processes range from ∼1 Å to structural-component lengths, spanning more than 12 orders of magnitude. In turn, the relevant timescales cover more than 22 orders of magnitude, with the shortest being in the femtosecond range.

They also write (bracketed comment added by me):

Today, a multiscale approach, based on both computational materials science and high-resolution experimental validation, is used to understand the controlling mechanisms and processes of irradiated structural materials. Figure 3b [shown below] illustrates the hierarchical multiscale modelling methodology, which typically combines ab initio structure calculations on the atomic scale, molecular dynamics simulations, kinetic Monte Carlo simulations, discrete dislocation dynamics, and rate theory with continuum calculations including thermodynamics and kinetics, as well as phase field calculations. Ab initio methods are required to calculate the most stable defect–cluster configurations, their dissociation energies, or the most likely lattice diffusion paths. Results of ab initio studies can be used as input for molecular dynamics, kinetic Monte Carlo, rate field theory and thermodynamics calculations. Additional links between different simulation methods are indicated by the arrows in Fig. 3b.

I think that, in most cases, ab initio methods for these problems means density-functional theory methods. A relevant review article is S.L. Dudarev, Density Functional Theory Models for Radiation Damage, Annual Review of Materials Research, vol. 43, 35-61 (2013).

• Weird that they think finite elements are a mechanical method. It is true that that is where finite elements were first used, but it's just a numerical method similar to finite differences and basis splines that you can use to solve a variety of differential equations. – Susi Lehtola Mar 3 at 15:09
• @SusiLehtola Where was that stated? You're of course right that FEM is general, but I guess that's the main place they get used in these applications where the foremost concern is about radiation damage to structural materials. – Anyon Mar 3 at 16:54
• in the figure you posted – Susi Lehtola Mar 3 at 18:35