...computational materials science is on an exponentially growing curve. Obviously, this won’t continue forever; it can’t. Even so, I foresee a period of around 10 years during which the discipline will undergo rapid growth.
The STFC’s Professor Nicholas Harrison provides his expert perspective of computational materials science: one of the fastest-growing fields of contemporary research...
Professor Nicholas Harrison
Computational materials science is one of the fastest-growing fields of contemporary research, perhaps because it offers practitioners the opportunity to exercise creativity in the truest sense of the word. Computational materials scientists use cutting-edge computer models to simulate material properties. This strategy is not only useful in terms of examining and testing existing materials, but it can also be used to create entirely new materials with desirable features.
One of the core aims of the Science and Technology Facilities Council (STFC) is to drive science and technology forward by harnessing the United Kingdom’s world-leading expertise, facilities and resources.1
In keeping with this goal, the STFC Computational Materials Science Group is concerned with the development of novel modelling techniques that can be applied across fields such as condensed matter physics, surface science and materials science.
In an interview with ScienceOmega.com
, Professor Nicholas Harrison, Head of the STFC Computational Material Science Group and Chair of Computational Materials Science at Imperial College London, outlines some of the most exciting advances from this burgeoning research sector…
I’d like to begin by asking a fairly rudimentary question: for the laypeople amongst us, what is computational materials science?
Computational materials science is an attempt to calculate the properties of materials, and to make predictions about how these properties will affect a material’s operational performance. Allow me to put that into context. Imagine that you are designing an aeroplane at an engineering level. You’ll need to use a computer model to simulate how your vehicle will function. This is a reliable way to calculate how air will flow over your moving vehicle, for example. Essentially, it is possible to design, build and test your aeroplane without leaving the virtual domain. When you come to build your aeroplane in reality, it will function in much the same way as your computer model.
Computational materials science is an attempt to do the same thing, but at an atomistic level. By learning how the atoms of a material behave, we can accurately predict how the material itself
will behave. This means that we are no longer confined by the materials that nature provides; we can design new materials in the same way that you might design a new aeroplane. Select your atoms from the periodic table, arrange them in a particular way, and my goodness – you have something that is both lighter and stronger than steel.
Are computational materials scientists involved solely in the design of materials, or do they help to build them as well?
Although we are at the forefront of the design process, I would say that we only do about a thousandth of the total work that goes into creating a new material. Our job is to discover a material with new properties. Once we have done this, it gets passed onto our colleagues at the laboratory level. They produce a laboratory proof to demonstrate that if you were to make this material, it would have the properties that we predicted. From here, the product is scaled up to the industrial level to see whether or not it can be produced in an economic fashion.
We also work with some large-scale industrial concerns. Typically, this is where a company has an existing material but is unable to figure out why it is behaving in a particular manner. We help to understand and calculate the processes that are taking place within industrial settings. Our techniques are useful not only for materials discovery, but also in terms of rationalising the ways in which materials behave.
So it’s possible to take an existing material, analyse that material and then improve upon it…
It is. An example of this process in action is a contract that we currently have with a high-profile oil company. We are trying to find new ways to prevent steel from rusting by attacking the problem of corrosion at an atomistic level. We want to learn exactly what happens when a material corrodes. In principle, we might be able to answer this question in a way that’s never been done before. We can perform a computational simulation to show precisely how oxygen incorporates into iron to form metal oxide. We can then ask a simple question: why doesn’t aluminium rust but iron does? The supposition that our client has made is that if we can understand exactly how
steel corrodes at this minute level, we can do something to stop it from happening.
How old is computational materials science as a discipline?
It is actually ridiculously new. I’m 47 years old and when I began my scientific career at the age of 20, computational materials science didn’t exist. You see, when you look at things at this atomistic level, you have to account for quantum mechanics. In order to understand how the electrons are behaving, you have to invoke quantum theory. By learning how atoms bind together, you can infer the processes that are taking place right down at the nanoscopic scale.
Of course, it takes a lot of time and effort to develop a computer programme capable of conducting quantum mechanical simulations. The code that we use most frequently, for instance, started being written back in 1974. This code has moved through several generations, and it has taken an average of 20 people working for almost 40 years to get it to its current stage. It has taken a long time, but we are finally in the position to start asking more challenging questions. Consequently, computational materials science is currently undergoing exponential growth.
So this could be the point at which things begin to get really exciting…
I don’t want to oversell it, but if you take any metric – industrial investment, scientific articles published, etc. – computational materials science is on an exponentially growing curve. Obviously, this won’t continue forever; it can’t. Even so, I foresee a period of around 10 years during which the discipline will undergo rapid growth. I would certainly say that it’s the fastest-growing area of science right now.
Are you and your colleagues currently involved in any other interesting projects?
At present, we are collaborating with a global mining and metals company in a bid to optimise mineral-extraction methods: we want to understand how best to extract copper from rocks.
We are doing an awful lot of work on spintronic materials. Traditional electronics is based on the manipulation of electrons. In computing, for example, electrons move through semiconductors in central processing units (CPUs). However, this technology has pretty much reached its limits now. If you scale down these systems any farther, the effects of quantum theory begin to dominate and traditional electronics cease functioning as they should.
This, of course, is a huge disadvantage. Until a few years ago, the average desktop computer was doubling in speed every 18 months or so. As you’ve probably noticed, that has now stopped. We have reached the limits of the speed that can be achieved via conventional methods. So, can we turn this disadvantage to our advantage? Can we use magnetism, as opposed to electronic charge, as our switching mechanism? To this end, we are currently collaborating with academics at the University of Oxford. We are using electron beams to punch defects into graphene. We hope to use these defects to control the spin of an electron.
These are just a few of the projects that we are currently working on. There is so much going on at the moment within our field. It is a very exciting time to be working within the arena of computational materials science.
'About Us', Science and Technology Facilities Council (STFC), http://www.stfc.ac.uk/1775.aspx
For further information about the work of Professor Harrison and his Imperial College London colleagues, check out the Computational Materials Science Group's website...