A platform for future development and application of the ONETEP software

Computer simulations play a growing role in our society e.g. flight simulators allow pilots to be trained more cheaply and safely than in the air. In science and technology, computer simulation is a powerful tool for understanding or even predicting complex processes in real materials. Simulations are often used alongside conventional experiments, but they can also be used in situations where experiments would be too expensive or even impossible to perform e.g. when studying materials in extreme conditions such as the high temperatures and pressures found in the Earth's core.

The turn of the last century saw the start of a scientific revolution with the discovery of quantum mechanics (QM), a theory that describes the world on the atomic scale with astonishing accuracy, and thus provides the foundation for all of low-energy physics, chemistry and biology. In principle at least, quantum mechanics underlies the microelectronics, chemical and pharamaceutical industries upon which our society relies today. The challenge is that the equations of QM are very complicated. Even on the fastest computers it is only possible to solve them exactly for small molecules, whereas the systems of interest to scientists today contain many thousands. Even the rapid and relentless progress of computer technology cannot overcome this because of the scaling of the problem.

The work needed to complete a task usually increases with its size e.g. the time taken to mow a lawn is proportional to its area: double the size of the garden and it takes twice as long. This is an example of linear scaling, but the effort to do many tasks increases more rapidly. Arranging a hand in a game of cards usually scales as the square of the number of objects involved: triple the number of cards and it takes nine (three squared) times as long. Some are even worse e.g. solving the travelling salesman problem to find the quickest route which visits a given set of locations. Adding one extra location doubles the amount of time to solve the problem. If three locations can be done in one minute, four will take two minutes, and five will take four minutes. Just 22 will take a whole year! Solving QM exactly scales like this when increasing the number of atoms being simulated.

However in the 1960s a leap forward was made with the discovery of density-functional theory (DFT), for which the Nobel Prize was awarded in 1998. The remarkable result of DFT is that the physical properties of the whole system (the answers to the questions we want to ask) can in principle be calculated in a time that scales linearly with the number of atoms. The research proposed here relates to one of the leading pieces of software for performing linear-scaling DFT calculations in the world. This ONETEP code has been demonstrated on systems containing up to 30,000 atoms so far. This method will expand the scope and scale of QM simulations across a wide range of fields. The problems we intend to address using this Platform grant include:
- giving insight into the design of new drugs by simulating the interactions between proteins
- designing new materials for solar energy conversion and storage
- studying the properties of nanoparticles as catalysts for chemical reactions
- understanding the microscopic processes associated with friction that cause machinery to wear out

Our vision is to create a virtual laboratory which uses ONETEP as one of a family of techniques to simulate the results of real experiments e.g. how a material absorbs light. The virtual lab gives total control: you can change the arrangement of atoms in a material or molecule and see the effect. The virtual lab will never replace the real one, but it promises to be a powerful tool alongside it.

In 2004, Accelrys adopted the ONETEP software as the flagship for a new Nanotechnology Consortium aimed at industrial and government partners, that ultimately attracted over 30 members worldwide. Accelrys is a leading scientific research and development software and service company. It is listed on the NASDAQ Global Select market and based in San Diego with European headquarters in Cambridge. It has 1,300 customers including 29 of the top 30 pharma/biotech companies, 4 of the top 5 chemicals companies and the top 5 aerospace companies. In particular, it has an established position in the field of materials modelling and simulation currently based around the Materials Studio application.

Members of the ONETEP Developers' Group (ODG - at that time consisting of the four applicants) contributed to training events across Europe and the US and a new staff position at Accelrys was created to develop the interface of consortium software with Materials Studio. In 2008, after the end of the first phase of the consortium, ONETEP was launched as a new product within Materials Studio. Total revenue from ONETEP to date exceeds £1M from over 200 organisations worldwide.

The licensing of intellectual property (IP) was handled by Cambridge Enterprise. As a result of these IP transfer agreements, new functionality developed outside the ODG has been secured and is being made available to customers of Accelrys through the Materials Studio interface. This pipeline will continue to be used to an increasing extent as the Platform (i) enables the team to expand and (ii) expedites the delivery of new code by maintaining the support of key PDRAs. Thus the knowledge associated with new simulation techniques developed by the ONETEP team for their own academic purposes will be efficiently converted into new tools readily available to ONETEP users across the industrial and government sectors. As the importance of materials simulation for research and development grows within industry, so the development of new methodologies with expanded capabilities such as ONETEP will contribute to the UK economy through the creation of jobs and improved products.

The ODG has direct collaborations with the pharmaceutical company Boehringer Ingelheim and the chemicals company Johnson Matthey, both of which have sponsored CASE studentships at Southampton. At Imperial there are links with the energy company E.ON. We intend to keep working to increase the number of industrial collaborators.

Members of the ODG will continue to work with Accelrys to provide training for customers. The Platform will benefit all users of the code by providing for the maintenance and enhancement by the key PDRAs of online support at the ONETEP Wiki (www.onetep.org). During the Platform we plan to extend a new paradigm for training to industrial and government users of the code through Accelrys, based on the format of the ONETEP Master Class held in Cambridge in July 2011.

Finally, the training of individuals working within the ONETEP team will make a significant impact. PhD students and PDRAs at the three institutions receive a rigorous training in computational science, numerical methods and parallel programming, all within a collaborative environment characterised by an ethos that embraces best practice in software development. Although most of the researchers leaving the team to date have continued in academia, at least in the short term, two recently-graduated PhD students have gone on to use their skills in employment at MathWorks , the developers of the MATLAB software, and another is now working for Arqiva developing software for simulations of wireless audio and video broadcasting. As the Platform fuels the growth of the team, so the number of highly-trained individuals with combined skills in computational science and software development will increase, helping to fill the skills gap identified by the numerous international reports cited in the case for support.