Over 122 PetaWatts of solar power reach the Earth's surface. This figure is vastly greater than the 15 TeraWatts (while still a crazy big number, that's only 0.015 PetaWatts) which is humanity's total power requirement.
So long term, I think it is inevitable that all of our power will come from solar. All the energy in fossil fuels is essentially ancient sunlight, captured by inefficient photosynthesis and geological happenstance.
Solar cells (formally called photovoltaics, as they generate electrical voltage from photons of light) are a beautifully simple engine that silently run on sunlight and thermodynamics to produce electrical power. There are no parts to wear out, no bearings to service, no oil to change. The only problem is that the solar resource is not very concentrated — bright sunlight is about a kilowatt per metre squared. So you have to build big panels to produce a meaningful amount of energy.
Our present best technology for making photovoltaics is based on Silicon. This is an abundant material (sand is mainly Silicon Dioxide), but forming pure Silicon requires a lot of energy, and making Silicon solar panels requires vacuum processing and some pretty nasty chemicals which keeps the cost relatively high.
My research is in trying to understand and design new materials for solar cells. By using computational chemistry and materials science techniques we try to understand how novel solar cells work (in particular, what limits their efficiency or longevity), and how we can make them better. These new solar cells are typically more complex than Silicon cells – relying on a subtle interplay of multiple elements in a particular crystal structure or fiendishly complex molecules dreamt up by an organic chemist.
I work with computational physics techniques. The equations which govern the behaviour of our everyday materials are simple to write down, but impossible to solve exactly for any but the most trivial of examples. Instead we rely on computational algorithms to numerically solve approximate models for the real system. Even very limited and flawed solutions of small systems can require absolutely stunning amounts of computer time. Validating (against reality) these models and understanding the limits of the techniques available, is what I spend the majority of my time doing.
2011: Seconded to Flexink, on the EPSRC Knowledge Transfer Scheme, for rational design of organic photovoltaic materials.
My publications are collated at: Google Scholar - Jarvist Moore Frost
My Academic blog is available at: http://jarvist.github.io/
et al., 2013, Novel BODIPY-based conjugated polymers donors for organic photovoltaic applications, Rsc Advances, Vol:3, ISSN:2046-2069, Pages:10221-10229
et al., 2011, Soluble fullerene derivatives: The effect of electronic structure on transistor performance and air stability, Journal of Applied Physics, Vol:110, ISSN:0021-8979
Frost JM, Faist MA, Nelson J, 2010, Energetic Disorder in Higher Fullerene Adducts: A Quantum Chemical and Voltammetric Study, Advanced Materials, Vol:22, ISSN:0935-9648, Pages:4881-+
MacKenzie RCI, Frost JM, Nelson J, 2010, A numerical study of mobility in thin films of fullerene derivatives, Journal of Chemical Physics, Vol:132, ISSN:0021-9606
et al., 2009, Modeling Charge Transport in Organic Photovoltaic Materials, Accounts of Chemical Research, Vol:42, ISSN:0001-4842, Pages:1768-1778