MIT Institute Professor and former CIA Director John Deutch yesterday warned that the US needs to revamp its “innovation system” if we want to make timely progress on global warming and energy security. The former Deputy Secretary of Defense and Undersecretary of Energy offered his views on the challenges to remaking our energy system in a plenary talk at the Optics and Photonics for Advanced Energy Technology meeting at MIT.

Deutch’s main point is that the researchers and entrepreneurs who are rushing to tackle the energy problem are following the traditional model of technology innovation: identify a problem, come up with an idea to solve it, engineer the solution for specific applications, and bring the solution to market.

This linear, technology-push approach is running up against the hard economic and regulatory realities of the energy market as currently embodied by Congress and exemplified by the Waxman-Markey Bill.

Innovation needs to be more of a team sport, Deutch said. You have to start the science, engineering and business aspects at the same time. “So this traditional distinction we have… between discovery and application is blurred because discovery here depends upon the character of the application,” he said.

On top of the systemic challenges, the Waxman-Markey Bill poses a particular challenge for energy innovation, Deutch said. On one hand you have Renewable Portfolio Standards (RPS) that dictate specific amounts of wind, solar and other renewable energy sources. And on the other you have cap-and-trade, which attempts to place a price on carbon emissions.

The two methods clash, Deutch said. “You have an inconsistent set of measures that are supposed to guide our energy future. The problem is RPS hides the cost of the generation technologies that are going to replace CO₂, and the CO₂ cap-and-trade system recognizes the costs,” he said. “It makes an uneconomic basis for technology choices.”

I hate the label “clean coal”. The simple alliteration and the pairing of opposites (coal = dirty, clean = clean) makes the term irresistible, especially for headline writers. Talk about a PR manager’s dream.

Even the cigarette makers didn’t have the chutzpah to label their low-tar products “healthy”. Granted they didn’t have such catchy wording available.

It’s a shame that research and development resources, particularly taxpayer money, are going toward new ways of burning coal. Even if we could get to zero emissions through some kind of gasification process, coal mining will remain an environmental nightmare. (Like virtually all Americans, I’m guilty of passively allowing someone else’s mountaintop backyard to be blown to smithereens so I can keep my lights on.)

I know the mantra: we have to keep all options on the table for dealing with global warming and energy security. But I worry about skewed priorities, especially given the significant political clout of the coal interests. I’m not against developing lower-emission coal plants, but I have deep reservations. I’d hate to see a single coal plant — however much better than existing plants — built if a renewable energy power plant could be built instead.

There’s also the issue of the world’s huge existing fleets of coal-fired power plants. MIT has published a report summarizing a March symposium on retrofitting coal power plants to reduce CO₂ emissions. The report argues that we need to tackle the problem if we are going to make much headway on global warming.

Developing carbon capture technologies for existing coal plants is akin to bandaging a gaping wound. No reasonable person would leave a wound untended, but also no reasonable person would call the bandaged wound “good as new” and mean it literally.

We should stanch the flow of greenhouse gases into the atmosphere. But our focus must remain on getting off of fossil fuels entirely as quickly as possible.

Call it the Jeffersonian energy model. MIT chemistry professor Daniel Nocera’s “personalized energy” plan has every home generating its own electricity and fuel from sunlight, and has power plants, transmission lines and gas stations fading into history.

Nocera has outlined the technology needed to realize this vision. And he says we’re in sight of making the technology economically viable.

The scheme, detailed in an essay in the journal ChemSusChem, boils down to using photovoltaics to generate electricity and solar water splitting to generate hydrogen. The hydrogen would fuel vehicles and power fuel cells for nighttime electricity.

Nocera’s lab developed a key enabling technology last year: a stable, inexpensive water-splitting catalyst made from cobalt and phosphate. Other research teams are working to integrate the catalyst with semiconductor materials in order to power the water-splitting reaction entirely by sunlight.

Nocera assumes that the average American home uses 20 kilowatt hours of electricity a day (the US Energy Information Administration puts the figure at about 30). A 3- by 2.5-meter solar water splitting panel can generate the hydrogen equivalent of 20 kilowatt hours from 5.5 liters of water in three hours, he said.

The challenge to making personalized energy affordable is lowering the cost of the scheme’s four components: water-splitting electrolyzer, photovoltaics, hydrogen storage and fuel cell. The electrolyzer is the most expensive component but the cobalt phosphate catalyst “puts a dent in that,” Nocera said.

In general, three things must happen to make personalized energy affordable, said Nocera.

First, researchers need to make fuel cells more efficient and less expensive. The key is finding alternative cathode materials. Today’s fuel-cell cathodes use platinum, which accounts for 38 percent of a fuel-cell’s cost. Researchers are looking for materials that are abundant, inexpensive and minimally damaging to extract and refine.

Second, researchers need to develop good hydrogen storage systems. Many research teams are working on materials that can store hydrogen at reasonable temperatures and pressures, but this is long-term research. We might see something practical in a decade or two. Another option is storing compressed hydrogen gas in tanks, said Nocera.

Third, researchers need to find ways to use sunlight to power water splitting. Many researchers are working on the problem, and many of them are looking at titanium dioxide nanostructures. The cobalt phosphate catalyst developed by Nocera’s group has opened new possibilities here.

University of Washington researchers are working to combine the catalyst with iron oxide, commonly known as rust. The work so far allows sunlight to generate some of electricity needed to drive the water splitting reaction, said Daniel Gamelin, an associate professor of chemistry at the University of Washington.

In the long run, all of the power should come from sunlight, said Gamelin. “As a community, we’re not so incredibly far away from this objective.”

The electricity generating aspect of personalized energy has been around for years in the form of residential photovoltaic systems, said Martin Green, a photovoltaics pioneer at the University of New South Wales. “The challenge is clearly the cheap storage systems,” he said.

Personalized energy is the most direct path to solving the energy challenge and should be a major goal of national and global energy policies, said Nocera.

I like his vision of everyone on the planet owning the means of their own energy production. I’d also like to see an economic and environmental analysis comparing regional, local and personal energy generation and storage.

When do economies of scale make energy produced in large centralized plants less expensive than personalized energy after factoring in transmission/transportation and environmental impact? Is it better economically for a small village to have a shared energy system than personalized energy? How about city centers, where sunlit surface area and storage space are smaller on a per person basis than in neighborhoods, suburbs and rural areas?

If we take strong action to curb greenhouse gas emissions we’ve got a good chance of minimizing the impact of global warming by the end of the century. But if we don’t, the consequences may be worse than previously thought. If we continue as usual, the planet will warm twice as much as previous models had indicated, according to a study based on an MIT computer model.

The MIT model, updated from a 2003 version that had projected 2.4 degrees Celsius warming by the end of the century, now projects 4.1 or 5.2 degrees Celsius warming, depending on whether recent ocean temperature data is included.

The model includes economic activity as well as atmospheric, ocean and biological factors. Changes from the 2003 version include improved economic modeling and data and account for cooling in the second half of the 20th century due to volcanic eruptions.

Spent fuel from nuclear reactors is a thorn in the side of those pulling for a nuclear energy renaissance. Radioactive waste just keeps piling up at nuclear power plants across the country, and there’s no consensus on what to do about it. The Obama administration’s decision to pull the plug on the controversial Yucca Mountain nuclear waste repository means there is no long-term solution (good or bad) on the horizon.

At Senator Tom Carper’s forum on nuclear waste at MIT today, the panelists — Matthew Bunn from Harvard, and Charles Forsberg, Ernest Moniz and Andrew Kadak from MIT — agreed that today’s interim storage is good for another half-century or so and we have time to work on a permanent solution.

The source of their confidence is the dry cask system of entombing spent fuel in big metal and concrete cylinders. The Union of Concerned Scientists supports the use of dry cask storage for 50-year timeframes but has expressed concern about safeguarding the casks.

Moniz called for a semantic change. We should call on-site storage “managed” storage rather than “interim” storage, he said. “We have to stop thinking of this as kicking the can down the road.”

Forsberg and Carper expressed interest in reprocessing spent fuel but Bunn, Moniz and Kadak soundly rejected the idea of reprocessing using today’s technology, which produces large quantities plutonium and increases the risk of nuclear proliferation. The three voiced support for continuing the US government’s moratorium on reprocessing. “Today we have about 270 tons of separated plutonium essentially in storage in multiple countries,” Moniz said. “Not a pretty picture.”

There was consensus on the need for a geological repository for spent nuclear fuel. They also all disagreed with the administration’s decision on Yucca Mountain.

Bunn, who is an expert in nuclear nonproliferation, called for the government to take ownership of nuclear waste. I asked Senator Carper after the forum about the idea of shifting the burden of nuclear waste management from rate payers to taxpayers. “We’re running a deficit of 1.6 or 1.7 trillion dollars,” he said. “That’s not a subject that’s going to get a lot of traction, to be honest with you.”

MIT report

Meanwhile, MIT has updated its 2003 report on nuclear energy. The original report called for a terawatt of nuclear power worldwide by midcentury in order for nuclear power to have a measurable impact on climate change. The update’s authors lament the slow start toward that goal over the last six years.

“We don’t appear to be shovel-ready,” Moniz said at the forum. “There is no new plant under construction in the United States, and even if one looks at the 44 plants under construction around the world, the fact is we are not on a trajectory for a terawatt by midcentury.”

I’m not sorry we’re not shovel-ready. Nuclear power is expensive and there are real concerns about plant safety and fuel security. And having half a century to work on the spent fuel storage problem doesn’t mean we’ll solve it.

Electric cars are poised to get two-speed transmissions. Compared to today’s one-speed transmissions two speeds cuts energy use by 5 to 10 percent, according to the transmission maker. Here’s the Green Car Congress story.

I had a good conversation today with Chris Mines, head of Forrester Research’s green IT market research practice. We talked about how businesses are coming to grips with energy and carbon issues.

A key point Chris raised is the institutional gap between IT and facilities departments. The IT department is responsible for the flow and management of data, and the facilities department for the flow and management of energy.

Does IT bear the cost and facilities gain the benefit of bringing the two together? Chris said it’s a hurdle organizations have to overcome as they work out how to lower their energy costs, secure their energy supplies and manage their carbon footprints.

This brought to mind an article I cowrote for Network World over a decade ago about tying embedded systems into organizational networks. One of our examples was the University of North Dakota, where they used the campus network to integrate environmental sensors into the school’s HVAC system.

The university saved hundreds of thousands of dollars by using the campus network to tie in the sensors. But what jumps out at me now is that the university saved $750,000 a year — 25% of its energy costs — by having automated real-time control of its heating system.

The project was run by the university’s physical plant department. Facilities departments today can buy these technologies from environmental control vendors like Honeywell, Johnson Controls and Siemens.

IT departments are addressing the energy issue by getting their own houses in order and becoming more efficient energy consumers. Green IT is largely about more efficiently powering data centers and reducing computational loads through virtualization.

So what does it mean to bring information and energy together?

I think it boils down to bringing efficiency and conservation to new heights. People are motivated to protect the environment and organizations are motivated to cut costs, but we know that it’s difficult for people to change their behavior when they can’t see the consequences of their actions. Telling people to remember to turn off the lights only goes so far.

If we can come up with effective, unobtrusive ways of monitoring our energy and carbon budgets in real time, most of us will probably spend more wisely.

When Energy Secretary Steven Chu delivered the Compton Lecture at MIT this afternoon he let his Bell Labs roots show.

During the talk, titled The Energy Problem and the Interplay Between Basic and Applied Research, Chu switched up the usual formula of breakthroughs in basic research advancing applied research. He said fundamental breakthroughs also come from highly focused applied research. His prime example — Shannon’s information theory — is a shining moment in Bell Labs’ history.

Chu also talked about modeling energy research laboratories on Bell Labs. He described a management structure where labs are run by the top practicing scientists whose intimate knowledge allows them to quickly deploy resources and help researchers connect with colleagues. He also described a place so rich in ideas that people are not obsessed with secrets.

He cited the Joint BioEnergy Institute as an example of a “Bell Lab-let”. The Institute is a partnership of three national laboratories and three universities focused on developing biofuels. The lab is headed by Jay Keasling, a UC Berkeley professor, Berkeley National Lab scientist and pioneer in the field of synthetic biology. “Great science is going to come out of this,” said Chu.

The brief question-and-answer session was dominated by the topic of money. Chu said he was going to be sending letters to university presidents and deans and heads of scientific associations asking for volunteers to help review grant proposals. The several billion dollars the Department of Energy is awarding for research is a large load on the system, he said.

The next question prompted him to say that the money we’re devoting to energy research and development as a nation is a small fraction of what’s needed. Most of our energy system is still based on burning oil and coal. We need to make it more high-tech, Chu said. The US spends over $1 trillion a year on its primary energy market. Following the model of high-tech companies we should be investing at least 10 percent of that in R&D, which would amount to about $100 billion per year, he said.

“Science and technology will be a cornerstone if not the cornerstone for how Americans are going to prosper in this century. And so what we’re investing now is nothing,” said Chu.

The Obama administration’s decision to cut spending on fuel cell vehicles came as a bit of a surprise, especially given the money the Department of Energy is putting into alternative vehicles in general.

The Wall Street Journal’s Keith Johnson asks if the decision amounts to the government picking a winner. I’d say they’re picking off a loser.

No alternative technology is likely to knock the gasoline-powered internal combustion engine off its throne by 2020, according to a Boston Consulting Group study. But batteries and biofuels are a better bet for making an impact in the next decade than fuel cells.

Electric vehicles benefit from the success of hybrids and the anticipated success of the coming wave of plug-in hybrids, particularly with oil prices expected to be in the midst of a significant rebound when plug-ins hit the market. Carmakers have stepped up their investment in battery R&D.

Although biofuels are a harder case to make than batteries, they benefit from a low cost of deployment. Assuming we can develop an affordable domestic supply of cellulosic biofuels, converting petroleum-based vehicles and fueling infrastructure is a minor cost.

In contrast, today’s fuel cells are expensive and have short life spans.

The clearest evidence against the picking-a-winner thesis is the carmakers. If they’d had a lot invested in fuel-cell vehicles you’d have heard a big outcry. How many images of sexy new prototype fuel cell vehicles have you seen in the years since the Bush administration touted the coming hydrogen economy? And how does that compare to images of electric vehicles, or even E85 vehicles?

At an MIT conference in March, Ford’s John Viera said the company is devoting more of its R&D to electric vehicles than hydrogen vehicles. You could make the case that instead of picking a winner, the government is following the market. More likely everybody is seeing the same thing, and all that’s changed is that the DOE’s Steven Chu is the first to abandon the talking points about agnosticism and say, and publicly act on, what everybody is thinking.

Viera made another point at the conference: alternative vehicle technologies are expensive and it’s going to take innovative business models to make them affordable. His comments were in the context of praising Better Place, the electric vehicle infrastructure company whose CEO is skilled at charming heads of state. If there are people working on innovative business models for fuel-cell vehicles, they’d better hurry up and make a case.

It’s also important to note that even though the DOE has thrown fuel-cell vehicles under the electric bus, they’re still investing in long-term hydrogen-related research. The work at two of the 46 Energy Frontier Research Centers is applicable to generating hydrogen, the work at two others will advanced fuel cells, and the work at another will help efforts to improve hydrogen storage. None of this long-term research is likely to make an impact in the next decade, but in 20 or 30 years some of it could lead to breakthroughs that will have us talking about hydrogen again.

If you learned that someone had invented a car that runs on water, you’d probably be thrilled. But if you found out that the car consumes 50 gallons of water for every mile driven, you might wonder if it’s worth it.

Of course any vehicle that requires 50 gallons of any liquid fuel is a nonstarter given the volume and weight of the fuel, but for the purposes of this thought exercise the issue is using up all that water.

Something very like this scenario is rapidly becoming a reality, and is even mandated by law. It turns out that producing ethanol from corn uses an awful lot of water, and the Energy Independence and Security Act of 2007 requires the US to produce 15 billion gallons of corn ethanol annually by 2015.

A study by researchers from Rice University, Clarkson University and Missouri University of Science and Technology found that it takes 500 to 4,000 liters of water to grow feedstock to produce one liter of ethanol, depending on the crop and where it’s grown.

Given an 800-to-1 water-fuel ratio and a car that gets 16 miles per gallon of ethanol (ethanol has a lower energy density than gasoline, which means lower mileage), you’d use 50 gallons of water per mile. This is the case for Nebraska-grown corn. You’d use 23 gallons per mile for Iowa corn and 115 gallons per mile for Texas sorghum.

The 15 billion gallons of corn ethanol mandated by 2015 is only about 10 percent of the transportation fuel the US is likely to use that year, but producing it will require the equivalent of 44 percent of the corn grown in the US in 2007. Agriculture today accounts for 80 percent of the water consumed in the US, and our freshwater supply is already under a lot of pressure.

The water-use scenario is very different for cellulosic feedstocks, particularly drought-resistant plants like miscanthus that require far less water. In theory, many types of grasses can be grown without any irrigation.

This makes efforts to come up with economical and scalable cellulosic biofuel production all the more urgent. Sorting out the land-use issues around biofuels is challenging enough without worrying about water.