How do you see the energy field developing in the long-term?
This is a very long timescale. It took 50 years to build up our energy infrastructure to what we have today and, given the size of that infrastructure, it will probably take 50 years to turn it over. Our crystal ball is not that good. The uncertainties are simply too great: we don’t know what technologies may be developed by then or what the impact of renewables will be. Will there be enough land and water available? What will be the impact of large numbers of windmills or solar panels? Solar panels absorb light so they will have a warming effect. I think there are a lot of surprises out there. We have to be very careful when we think about developing new energy sources at the same scale we have today. Also, we don’t know what the geopolitical realities will be. We should not bet on a specific technology but we should provide options for the future. At MIT we’re developing a portfolio of science, technology and policy options that are economically acceptable and that could be used as we move out on that 50 year timescale.

Right now one of our biggest single focus areas is solar energy, but the intention is to be pretty broad. You can see that in the research fields we have with companies – transportation with Ford, deep ocean with Chevron, energy conversion with BP, solar with Eni ...

Which energy technologies do you believe show the most potential?
Long-term, solar is the long pole in the tent. It’s the renewable resource with the biggest potential – there’s plenty of capacity there. The challenge with solar is to do it affordably so that we can deploy it at large scale, to understand in advance any environmental surprises that might come, and to get the business systems to enable it to happen. Using solar, unlike oil and gas, is a huge cultural change. We would move from what you might call a hunter-gatherer society model, where we send out our expert geologists to go find a resource such as oil, gas or coal and they bring it home. In a renewables world it would be more like a farming society, where you have this resource basically everywhere – granted some places better than others, better in the deserts in Arizona and California than in Seattle or in the UK, but it is still very widespread.

The question is then one of efficiency and harvesting renewably so that you don’t use all your resource. There are MIT faculty members who are starting to worry about unexpected issues as you scale up. Right now an expectation might be that you put large solar farms in desert regions where there is a lot of uninterrupted solar radiation. But those desert regions generally reflect back much of the incident radiation into space. Solar cells are designed to absorb energy. In organic systems we can convert maybe less than 10% of the incident radiation; with the best solar systems we convert around 20%. The rest of that energy is there as heat. So what does it do to global temperatures if you start to absorb energy instead of reflecting it back? If you were to supply current US energy needs with solar you’d need around two per cent of US land mass converted to solar cells. If you’re absorbing energy over that amount of land it’s going to have some effect.

We have to think about it before we go a long way down that route. There was a time when researchers thought there couldn’t possibly be any problem with burning fossil fuels, because if you do it very efficiently all you produce is carbon dioxide and water. They are already both present in the atmosphere so what harm could it do? The answer is, well, no harm at all if you burn just a little bit of fossil fuel. But when you consume the magnitude of fossil fuel that we do, as we have found out the hard way, you start to have a substantial impact on the amount of carbon dioxide in the atmosphere, which in turn affects the global mean temperature. And here we are facing a potentially disastrous consequence.

Here at MIT, Ron Prinn from the Department of Earth, Atmospheric and Planetary Sciences and Jake Jacoby from the Business School are leading a joint programme on the science and policy of climate change. One of Ron’s main concerns is that we have not thought about renewables at scale; all of these systems do have an effect on their environment. He would like us to avoid surprises like we’re experiencing today, if possible.

Ron has started a list of potential problems – one of them is wind power. It may be that with wind you will change convective patterns of heat transfer from the Earth’s surface and reduce the cooling effect. A windmill takes energy from the wind, so you trade off some of the windspeed for power produced by the windmill. The question is what effect does that removal of energy from the wind have on climate. I think that if you lose some wind, there’s not likely to be any impact at all. But I think it’s worth asking before you go too far down that road.

We are so far away from being at fossil fuel level. About five years ago renewables provided around one per cent of global energy needs, so we just have no idea what these could do on the kind of scales we’re talking about.

What is your take on biofuels?
In the US we’re starting to farm more and more corn (maize) for biofuels and we’re discovering that impacts the price of corn, food prices for people and feed prices for cattle and other domestic animals. There’s a whole food chain that’s affected by that, so for many reasons that’s probably not a good idea.

I don’t see any reason why you can’t get energy from biofuels but it seems like you need to do it much more thoughtfully than we’ve been doing it. To begin with, burning your food is not a good idea. You want biosources that don’t compete with people’s food. That’s the first thing that we haven’t thought out. We’ve gone down that road apparently because there’s a very effective corn lobby in the US.

It’s pretty clear that corn, at least the corn grain, is not the right feedstock. And, certainly at MIT, it’s not at all clear that ethanol is the right fuel to get from that or any other biomass. There are other possibilities – you could go to longer chain alcohols such as butanol or pentanol. These have the advantage that because there are more carbons in the backbone they have a higher energy density. Ethanol has 70% of the energy density of gasoline. If you fill your tank with ethanol your car goes 70% as far down the road. I don’t think many people realise that.

Or you might not go to an alcohol at all. By metabolic engineering of the organisms that ferment sugars into alcohols, you can design them to make other kinds of fuels such as alkanes. At the beginning of an era of biofuels, we need to be thinking “forget what the options are, what are the impacts of the different options?” and making a more reasoned decision about them. It might not be the same globally, I would expect other regions of the world to have different biomass sources that are preferable. Brazil has got this wonderful climate that enables it to grow lots of sugar cane. They avoid the need to hydrolyse cellulose into sugars, and can go straight from sugars to alcohol. But lots of the world can’t grow sugarcane. The question is what could you grow and what could you grow sustainably?

Besides energy, our MIT president’s other major thrust is the convergence of life science and engineering. That fits very nicely to the energy bioscience area as well. There’s a large number of faculty members looking at different biological issues around energy, ranging from what you might do with algae, which are very efficient converters, to basic metabolic engineering – how you would turn any of these materials into the right fuels, and from downstream processes to business models and how you would put together an energy-based infrastructure that relies on biofuels. That would be very different to oil. We literally ship oil all around the world. We produce it in the Middle East, in Norway, in the North Sea, ship it to anywhere in the world and then refine it. Biomass is way too bulky to ship that far. You have to grow it locally and then start to process it locally.

Whether biofuels are more efficient or cost-effective than oil is not clear. With oil, coal and natural gas you start with one huge advantage: nature and time have done a lot of the work for you. With biomass you’ve got to do pretty much all the processing yourself. With a crop, the sun and the plant do some of the work but you have to do a lot of the actual conversion.

How are you funding the initiative?
We have constructed a set of industrial sponsors that interact with us on particular aspects. We have enlisted nine in the first year covering a wide range of energy technologies, including BP, Italian oil company Eni, Schlumberger, Chevron, Enel, Bosch, ABB, b_Tech (a Spanish renewable energy business based in Barcelona) and Ford.

Each one has a sponsored research programme – a one-on-one research relationship with MIT. As an example, BP has chosen a theme of energy conversion – converting low value, high carbon feedstock into high value liquid fuels, i.e. coal to clean fuels. There are a variety of programmes within that – catalysis, gasification and so on. Eni has highlighted a major emphasis on solar – I would characterize it as "beyond silicon" – looking at advanced materials, and a long range view of new technologies to make solar affordable and possible at the kind of energy densities that will be needed to make a major impact on global energy supply.

BP and Eni are the two founding members. Founding members must commit to a minimum of $5 m a year for five years and sustaining members must commit to $1 m a year for five years.

In addition we have a number of elements of the programme that are shared. Companies contribute to an energy research seed fund, in which we put out calls for proposals to the MIT faculty twice a year to identify novel ideas. It’s a good way to get new faculty members involved.

The first funding round was completed in December. We awarded about $1.7 m seed funding, covering a broad range of technologies, including new materials for solar – Earth abundant materials. That last project’s a partnership between a materials scientist who’s an expert in computational materials design and one of the mechanical engineers with a lot of expertise in solar photovoltaics. The idea is between the two of them to do a rapid screening of candidate materials so that they can identify and test the most likely candidates for solar materials. There’s a lot of research going on in solar technology but often it involves materials that are too rare or too expensive ever to be realistic to scale up. To be hard-nosed, for solar to ever have a major role in energy supply, you have to have a system that is scalable up to the size of global reach.

What do you think the energy field will look like by 2030?
The biggest thing I expect to see on the research side is that we will be devoting a lot more resources to energy research than we do today. This is my vision, I hope it’s not just a wish. Certainly industry has come around and hopefully governments around the world will step up to the plate and start funding in a big way. We won’t have solved anything by 2030 but hopefully we’ll have some good starts in a variety of directions and perhaps a better sense of where to focus our efforts. If you think about how subtly energy infrastructure turns over, the world in 2030 is going to look a lot like the world in 2005.

One of the key messages is, if you want to meet the projected doubling of energy demand this century, the biggest single contributor is likely to be energy efficiency. That’s the one that doesn’t cost anything in terms of carbon dioxide – it’s the energy you don’t use. I expect by 2030 we will see many projects in energy efficiency, because that’s the low-hanging fruit. There’s a lot of possibility, for example at MIT there’s much work on energy-efficient buildings, more efficient transportation systems, better engine designs, lighter weight cars ...

Energy efficiency is the biggest potential contributor but there are others. I expect to see more biofuels contributing by 2030. One of the things we are advocating as a clean way to meet projected energy needs is clean coal – coal combustion and carbon sequestration. The projection though is that that’s not likely to have much impact by 2030 – one factor is the lifetime of coal plants, which is 40 or 50 years, the slowness with which we are developing a carbon strategy and the unwillingness, I think justifiably, of utilities to build a very expensive power plant without knowing what the policy and regulation are going to be.

For our domestic production in the US, policy and regulation has to be done at the federal level. To be effective we need global agreement as well, whether that’s carbon prices or taxes or cap and trade. Right now the price is external to the marketplace, nobody suffers economically for carbon emissions now – we will all probably suffer down the road and our kids will suffer.