Review: 2002 Hoffert et al paper

Fusion, or Space Solar Power?

In November 2002, Marty Hoffert and 17 co-authors published a landmark paper on future energy options capable of handling realistic future demands for the next century and more. The following review was first published in kuro5hin, an online discussion forum, shortly after the paper came out.

National energy policy! California blackout! Fuel-cell cars! Wars, terrorism and oil money! Enron! Giant wind farms! Kyoto!

So many headlines of the past few years have a common background theme: the dependence of modern economies on a steady, dependable supply of energy, and the consequences of our current fossil fuel dependency for global stability and climate. Clearly this cannot continue forever. Worse still, most of the people of the world do not even live under modern economic conditions as yet, and as China, India, and other similar nations continue to progress, world energy needs will almost inevitably double or triple from their current levels. So where is all that energy going to come from?

In the November 1, 2002 issue of Science, Marty Hoffert of NYU and 17 co-authors have published an analysis of the energy options that will be available to meet world demand a few decades from now, under the constraint of constant or reduced carbon dioxide emissions. While there are many short-term measures that could make a difference, the only long-term viable alternatives seem to be fusion and space-based solar power.

This article received some main-stream press coverage, including in The New York Times, space.com and MSNBC, as well as a large number of more local outlets, with different aspects of the report emphasized to varying degrees. It's not clear the primary message has gotten through though: we're not ready to replace fossil fuels yet, and we should be spending an awful lot more money (Manhattan or Apollo-project sized) on research into the long-term alternatives, so we can be ready in time.

As the article states, Kyoto is paradoxically both "too weak and too strong: Too strong because the initial cuts are perceived as too much of an economic burden by some (the US ...); too weak because much greater emission reductions will be needed, and we lack the technology to make them."

What are the technology alternatives available now? Improved efficiency is the maxim of the environmental advocates - a lot can be done there, both with improved technology and altered lifestyle choices (if that's acceptable in a democracy). But as the authors point out, the limited potential effects of improved efficiency (at best a factor of 2 or so) in the Western countries would be overwhelmed by increased energy requirements in China and India. Efficiency improvements are worth doing, but will only delay the inevitable, unless other measures are taken.

On the technocrat side (exemplified by the position of such US agencies as the Department of Energy) the long-favored solution has been power from nuclear fission. Surprisingly, this is as much a limited resource as oil, which the article estimates at 6 to 30 years-worth of reserves of land-based uranium, if used to sustain 100% of world energy usage. More uranium is available from seawater, but the authors point out extracting sufficient uranium to power the world would require processing seawater at a rate at least five times the total outflow of the world's rivers to the ocean. Fission power can be stretched through production of plutonium in breeder reactors, but most of the world has decided that is an unsafe course to follow.

Use of hydrogen as a fuel does not directly help since hydrogen is not available in raw form in sufficient quantities to serve as a major fuel: original energy of some other form is needed to produce it. Hydrogen may help in relocation of the CO2 production from end-users to central power plants, however. A recent set of large-scale engineering proposals involves "sequestration" of carbon dioxide; capture and removal to the deep ocean, or underground in old oil fields, etc. This will likely be a valuable short-term measure, but if nothing else is done the sequestration rates required to stablize global CO2 levels will be enormous, and we will run out of places to store it all safely.

That pretty much leaves the renewables, and fusion. With renewable energy sources, by definition, we will not run out (at least while the Sun still shines), so they definitely provide a long-term CO2-free solution. All renewables, however, suffer from low power density: you need to occupy a lot of land area to capture the 3-10 TW (electric) estimated needed. For example, with solar photovoltaic panels the land area needed would be 200-600,000 square kilometres (100-250,000 sq. miles) or up to 7% of the land area of the United States. Using biomass (green plants and incinerator/generators doing the work of the solar panels) even 3 TW would require as much land area as is currently used by human agriculture.

The most intriguing renewable option is space-based solar power. Due to the direct exposure to the sun and ability to produce energy 24 hours a day, a space-based solar option would require only 1/4 the area of one on Earth's surface (after accounting for microwave transmission losses). The technology was researched in the 1970's with a plan for large satellites ("the size of Manhattan", i.e. each about 20-30 square miles) in geosynchronous orbit; on the order of 1000 of these would meet the space-based area requirements for 3 TW electric worldwide power supply. The Earth-based land area requirements for receiving this energy via microwave beams would be substantial, but less than for land-based photovoltaic, and would allow dual-use (for example as range land) where land paved with photovoltaic cells would be difficult to use for anything else. Alternatives to geosynchronous orbit may be economically more attractive, particularly use of the Moon as suggested by David Criswell in a number of articles recently, notably in the April issue of The Industrial Physicist, which received considerable follow-up.

Space solar power has the advantage over fusion that no real fundamental further research is needed to figure out how to do it - the primary challenges are those of large-scale space engineering. Criswell has estimated the start-up costs for his lunar solar power system would be on the order of $150 billion before the system would start to pay for its continued expansion - a substantial (Apollo-project-level) investment, but given the scale of the problem facing the world and the expected cost of most other solutions, not an impossible figure.

One other space-based option for climate change mitigation is to directly block the sun with a mirror, placed at the L1 semi-stable point between Earth and Sun. A mirror the size of the United States would block about 2% of solar energy, roughly compensating for a doubling of atmospheric CO2. Of course the sun would appear to have a permanent spot right in the middle!

Fusion research has progressed over the years, and several major new test reactors are in development, most prominent among them being the International Thermonuclear Experimental Reactor (ITER), with construction costs estimated at about $10 billion. The next major milestone, aside from understanding the engineering challenges with confining large, hot, plasmas, is to demonstrate net electric power production from a self-sustaining fusion reaction. The authors of this study find it unlikely that fusion could be relied on to help with the CO2 problem before 2050 or so, but continued research on this long-term option is very important.

The world has a serious problem here. We're clearly not ready to deal with it. Whether on the space engineering or fusion physics/engineering side of things, a massive investment in research and development funding will be needed to get these long-term energy projects moving. Tragically, in the recently declared US national energy policy, only one of the more than one hundred recommendations even mentions fusion energy, and that recommendation also suggests promoting hydrogen and fuel cell use (confusing given that hydrogen cannot be a primary energy source). Solar energy and other renewables (along with conservation and other short-term measures) are mentioned in quite a few recommendations; however the possibility of large scale space-engineered systems is completely overlooked. Further evidence of a lack of thought for the future can be found in the 2002 budget for the US Department of Energy; $2 billion over 10 years for "clean coal", $1.4 billion over 10 years for "weatherization assistance", solar and renewable technology money contingent on ANWR approval, and overall a $700 million cut from the previous year's budget. The request for the 2003 fiscal year (still not yet passed by Congress) doesn't look any better, with actual dollar cuts in the (already small) levels requested for renewable and nuclear energy research.

Is there any real hope for such massive R&D projects? Are we, the democratic western societies, still capable of such things? The somewhat aimless International Space Station has been beset by funding troubles, but the cost involved is a few times less than total costs for new energy solution R&D would be. But we did succeed with the Apollo project, roughly $100 billion in cost, at a time when US GDP was less than 1/4 its present size (in constant current dollars). And the cost scale is not so far out of the range of current major power systems: a typical multi-GW modern power plant costs several billion dollars, and hydropower plants already involve engineering on an enormous scale here on Earth - for example China's Three Gorges Dam is estimated will cost over $20 billion to build, with a reservoir 370 miles long, and the concrete dam itself 1.2 miles across and about 600 feet high. And even that can only provide a small fraction of China's future energy needs.

Personally I'm an optimist; I think most people just don't yet realize the scale of the problem or of the research effort needed to overcome it. When this is widely understood the world will surely rise to meet the challenge. Won't it?

Created: 2005-01-18 01:53:49 by Arthur Smith
Modified: 2005-01-18 02:52:02 by Administrator