Tuesday, December 9, 2014

Clay composites modelled in computer

Scientists in London have constructed a computer model that predicts the useful, physical properties of clay composites from their atomic make-up.


Clay-polymer composites can be very strong, stiff and light, making them useful in components for cars and aeroplanes.


But they are usually developed by a costly process of trial and error.


The researchers say their 'virtual lab' system could help take the guess-work out of making such new materials.


Their study is published in the journal Advanced Materials.


Composites that are made by mixing molecules of a polymer, such as nylon, with flaky layers of clay, can have properties that are very different from the separate ingredients.


Predicting those properties can be difficult - even when the molecular structure of the ingredients is well understood.


'Some of the most obvious things about the world around us are harder to extract than you'd think - even though we've had an atomic theory of matter for at least 120 years by now,' said Prof Peter Coveney, whose team at University College London (UCL) conducted the research.


According to Prof Coveney, that difficulty in extrapolating from small properties (forces between atoms and molecules) to large ones (like hardness, density and conductivity) has held up the development of applications for exciting new materials - like superconductors, or graphene.


'Quite often the modern-era Nobel Prizes are awarded to discoveries of interesting molecules, which are supposed to then have lots of interesting applications,' he told BBC News.


'But the actual delivery of those applications can take an awfully long time.'


Prof Coveney and his team have set about tackling the problem using a process called 'multiscale modelling'.


This technique, whose pioneers were awarded last year's Nobel Prize for chemistry, combines rules about tiny interactions - single atoms and electrons, at the quantum level - with an understanding of how larger chunks of matter interact.


It allowed the UCL researchers to test different combinations of polymers and clay molecules, without having to actually make all the composites in the lab.


'Without modelling you'd have to be doing lots of different experiments,' Prof Coveney said.


For example sheets of clay, about five atoms thick, can spread out within the composite, or be stacked like a deck of cards.


The first, spread-out type of structure was discovered by Toyota in the 1980s. Simply by trying out different mixtures of polymers and clay, the car company eventually produced - and patented - the first 'nanocomposites'.


The latter structure, with polymer molecules acting like mortar between tiny, flat bricks of clay, is more tough and durable and is found in natural materials like mother-of-pearl.


Prof Coveney says his research can help reveal what ingredients and conditions produce these different configurations.


His team tried several combinations and then checked that the results of their model matched the known properties of real composites.


'Now we've got a virtual laboratory - a simulation environment that can tell you a lot of detail about why there are so many variations in the products you could form, based on these different chemistries,' Prof Coveney said.


He believes that the work could be extended to model things like the industrial production of graphene.


'There's almost a race on to show that graphene's not only interesting because of its molecular nature, but has actual practical applications.


'That requires the same sorts of issues to be addressed as in the clays - what happens if I have a larger quantity of the material? Will it still have interesting properties?'


Prof Adrian Mulholland from the University of Bristol uses multiscale modelling to study biological interactions, like how drugs affect the proteins in our bodies.


He told the BBC the new study was a useful contribution to a 'very active area'.


The field is important for studying and improving many processes in industry and medicine, Prof Mulholland explained - as highlighted by the Nobel award in 2013.


'It's an idea whose time has come,' he said.


Follow Jonathan on Twitter