Fusion Power Is Here Now!

Mark Kleiman writes:

The Reality-Based Community: Making progress backwards: If you're my age — that is, if your childhood reading included books such as Our Friend the Atom — one of the things you've always known is that all energy problems are transient, since eventually we will learn how to do controlled fusion and the world's oceans will be its fuel supply. In the late 1950s, fusion power was commonly said to be thirty years away. But for decades the horizon seemed to recede as fast as we moved forward; by 1975, fusion was expected to be ready sometime early this century, and by 1990 the schedule had slipped back to something like 2020.

A nuclear engineer of my acquaintance tells me that this is no longer the case. Instead, we're moving backwards. Now no one seriously expects commercial-scale fusion power in the next thirty years. The more we learn about the problem, the harder it gets, and each passing year adds to, rather than subtracting from, our distance from the blessed day when we can thumb our noses at ExxonMobil and the House of Saud.

It seems to be possible that fusion will arrive about the Twelfth of Never.

Nonsense! Mark Kleiman merely needs to look up! (At this time of year, between the hours of 7:30 AM and 6:30 PM.) Gravitational containment of fusion power is very practical, and currently delivers about 175 petawatts of power to the earth.

Turning the energy generated by this gravitational-containment fusion reactor into more useful forms, or starting up another, smaller gravitational-containment reactor at a more convenient place do, however, pose considerable engineering difficulties.

Comments

[shah8]

I did always figure that antigravity is a prerequisite for fusion power. It's also a prerequisite for extensive manned space travel as well.

Such a smooth, finely grained power curved is necessary for so much.

And that's why we're gravity's...er, well you know what...

Unless we can woo it, or something...

[vt codger]

I don't know where Kleiman gets his information, but I have to say that I was around in the 1950s. Over the years I've heard (and rejected) lots of silly assertions about energy including fission power "too cheap to meter", and claims a decade ago that deregulated electricity would be dramatically cheaper. I never heard anyone who sounded like they had a clue promising fusion power before 2040 give or take a little. Not in the 1950s, 1960s, 1970s, 1980s, 1990s and not today. Currently, commercial fusion power is claimed to be about 35 years away. Which puts in pretty much on schedule for 2040 give or take a little. Personally, I'd guess that it might be a decade late (If it can be done at all which is still not 100% clear).

As for solar power, it has its place. Domestic hot water in warm climates is a no brainer. Some domestic heating. Some electricity someday, but its really hard to recover the start up costs in later savings today.

I live in a town where the days around Christmas are 10 hours long and the maximum sun angle then is 22 degrees. Nobody has ever explained to me how I can have solar power that works year round. At least not without rebuilding the house into a structure that no sane person would live in and cutting down a bunch of trees that I am moderately fond of.

[Ronad Brak]

I've read science books for kids written about 25 years ago that promised we'd have fusion reactors by now. Also, we were supposed to have magnetic levitation trains ten years ago. But we did get the portable miniture phones, however we don't wear them on our wrists.

I'm optimistic about the costs of solar power dropping. One low cost thermal solar power design that has been tested here in Australia stores heat to provide baseload power and the company responsible insists it will be able to compete with fossil fuels on price in sunny areas. Even if it only manages to come close to being competitive with fossil fuels I will be impressed.

[Kaleberg]

That's the difference between active and passive solar power. In active solar power, you have to create your own miniature sun. When I was in grad school at MIT in the 70s one of my housemates was a fusion physicist. He told me flat out, "Fusion is the power source of the future. Always has been. Always will be."

A lot of new technologies get the thirty year estimate thing. Powered flight put up with it for at least 50 years, and we've been getting it for genetic medicine since the 60s. Technology is impossible to predict. The question is whether it is whether we want to spend the money on it as opposed to other pies in the sky. Not all of the pies ever land, but enough have landed that we like the taste.

We considered solar power in the far northwest, near Seattle. The problem was six cent kilowatt hours and a 30 year recovery. If we were living in NYC, we could have paid for it in about 12 years.

[jhe]

Other than fusion, it's pretty much all solar power - it's just a question of whether you're unpacking it from coal, oil, wind or taking it direct from the tap.

[bigTom]

I worked in "fusion research" back in the early eighties. That was about the time when the twenty five years until we have it bagan to increase. The current best guess for most scientists is never, although it still still worth spending some resources on it as a long shot.

Generally most people think of electricity when we talk of solar power. Space heating from solar is thousands of years old. Even just opening your blinds on south facing windows during the winter as a way to get at least some heat is doable -but underexploited in the US. Now the electric variety, whether photovoltaics or solar thermal is still a ways off (I will consider it here when a decent fraction -say 1/4 of new generating capacity is solar). If you take a growth rate of 40% a year, and compound that over 15 years, thats about how long it will take to "arrive". The commonly used polysilicon panels have been production constrained the last couple of years, buts lots of new capacity is scheduled to come online the next couple of years -especially in China. Perhaps we will see a growth spurt -beyond that 40%/year exponential rate, but even so, I think the age of solar has not yet dawned.

[bigTom]

Brad, A related issue that I'd lot economists to start thinking about. Lets assume for the sake of discussion that in say 25years solar power becomes too-cheap to meter, i.e. the incremental cost of photovoltaic roofing over traditional roofing materials becomes too low to bother with the later. Then as vt codger alludes to, we have a seasonally varying power base. Seasonal storage is likely to be too expensive, and covering the full wintertime deficit with other forms of generation is likley to also be pretty expensive, as these plants would not be used often enough. So we have an productive system which can only do certain energy-intensive things during part of the year. How do we evolve an economic system which can handle that sort of time variation?

[NervousRex]

Go read about ITER. It should come closer to breakeven and be used to test materials for long-run reactors.

Being cynical about progress is really cool and stuff but thank goodness someone is willing to roll up their sleeves and build something that might not make money right away.

http://www.iter.org/

And if it's not? We wasted what exactly, a half-year of potato chip sales over a period of 10 years?

I'm completely pessimistic about our response to Peak Oil production but quite certain fusion is worth the effort.

Meanwhile, wind, solar and conservation, and expect difficulties.

NR

[Steven]

Seasonal storage is likely to be too expensive, and covering the full wintertime deficit with other forms of generation is likley to also be pretty expensive, as these plants would not be used often enough.

Maybe I'm missing something obvious, but...don't we already have a situation like this? For example, I live in New England, and my natural gas usage is (obviously) much higher in the winter than in the summer. This doesn't seem to cause too much trouble. Financially, I can choose to have the company spread out my gas bills over the entire year, which is a good choice for me (the bill is more predictable) and a good choice for the company (their revenue is more predictable, and essentially they get the money earlier than they otherwise would). Now, presumably we're really talking about coal and nuclear as well as natural gas, but it seems like the same principles would apply (not sure about hydroelectric).

[Steven]

Sorry, that first paragraph should've been in quotation marks; I forgot that italic tags don't work on this blog.

[NE1]

The best carbon/power summary out there: http://eands.caltech.edu/articles/LXX2/powering.pdf

Solar's the way it's got to be, and we've got to start *hard*. We don't have time to put all our eggs in the fusion basket.

[NE1]

err... too bad it's 10meg

[Winston]

The technology for cost-effectively harvesting solar energy is finally here and from a bit of an unexpected place. While photovoltaics are still about a factor of 3 too expensive to be really practical, a couple of solar thermal systems are in the $0.10/kWh range, which is economical in California (where energy of any kind is expensive). The best of these uses a modern Stirling engine (an early competitor of the steam engine which is very efficient) combined with a dish that tracks the sun. They can get convert 40% of the sunlight that hits the dish to electricity, which makes also take up less space per Watt than photovoltaic systems. Unfortunately, it is really only suitable for rural areas (37 foot diameter, reflective, parabolic dishes don't fit well in cities).

The first large scale (500 MW) plant of this kind will be online in about 2 years.

[jerry]

Do we include wind farms as solar energy?

And a cheap botec shows that lining the sides and median strips of our major western interstate highways with average efficiency photovoltaic cells provides a major source of electricity, and does so in a way that is distributed, environmentally sensitive, and easy to maintain.

[bigTom]

nervous:
I know about ITER. Achieving energy breakeven -or even sustained ignition for a couple of runs is lightyears away from an economically competitive energy source.

winston:
If your stirling engine dishes are the ones I heard about, they were not projected to become cheap enough. The linear parabolic systems from AUSRA, which PG&E and FPL are seriously looking at for peaking applications are pretty interesting however.

steven:
Your heating system, and your AC are only used part year. I think the way our economy runs productive (i.e. business) investment being able to run a manufacturing plant for only part of the year would present problems. And yet large scale adoption of alternative energy is likley to lead towards that sort of constraint. Thats what I want the economists to start thinking about.

[Winston]

bigTom

San Diego Gas and Electric went with Sterling Energy Systems dishes for a 500 MW installation and PG&E just bought the same size solar trough system, so it is likely that the two approaches are cost competitive. There are advantages to both. The parabolic trough approach is less expensive to build, but requires a heck of a lot more land and doesn't provide power over as long a period of time as does the tracking dish approach.

I'm under the impression that the dish system is marginally cheaper, but manufacturers of both claim that they're less expensive.

BTW, here are the web addresses of the two companies that are installing big solar thermal systems in Southern California

www.stirlingenergy.com/
http://www.solel.com/

[Neil']

Should we really write off cold fusion? It hasn't shown clearly favorable results, and the theory is flaky, but lots of apparently good researchers say they've seen something interesting.

[Rich Gibson]

"i.e. the incremental cost of photovoltaic roofing over traditional roofing materials becomes too low to bother with the later. Then as vt codger alludes to, we have a seasonally varying power base."

bigTom is worried about a seasonally varying power base. I'm going to go out on a limb here, but I _think_ we have several available or darn near available, technologies to convert surplus electricity into a storeable form.

The whole 'hydrogen economy' becomes more reasonable if it is simply a way to convert surplus summer electricity into winter energy.

Also, we have a seasonally varying power demand-summer electricity demand is much higher than winter.

[bigTom]

winston: nice reference to the SES site. It is good to have competing technologies available. The amount of storage, and the ratio of collector to turbine power can be varied for the trough systems. I.E. enough storage could be added to adapt it to baseline power needs if desired. That still wouldn't cover the seasonality of the power. For a horizontal surface solar insolation varies by roughly a factor of three at typical US or European latitudes. A tracking dish or trough wouldn't see quite as much variation, but it is still significant. Near term the seasonality isn't a major issue, but as we try to eventually scale solar up to be a major component of power it could become one. Of course if solar hydrogen becomes economically feasable, then storage in the form of chemical energy would work well. Or else we could use natural gas for the off season. Either way expensive capital equipment goes un/underused for part of the year. That will adversely effect the economics.

My favorite in this space is SolFocus, whereby miniature reflectors focus sunlight onto 1mm square multi-junction PV cells. This enables expensive 40% efficient cells to be used. But all these utility grade solutions require the additional expense of solar trackers. Especially large trackable dishes such as needed for the stirling approach strike me as pretty ungainly and hence expensive.

[M. Simon]

Here is an alternative that may provide a quick way to fusion:

http://powerandcontrol.blogspot.com/2007/03/mr-fusion.html
Bussard Fusion Reactor

http://powerandcontrol.blogspot.com/2006/11/easy-low-cost-no-radiation-fusion.html
Easy Low Cost No Radiation Fusion

It has been funded:

http://powerandcontrol.blogspot.com/2007/08/bussard-reactor-funded.html
Bussard Reactor Funded

The above reactor can burn Deuterium which is very abundant and produces lots of neutrons or it can burn a mixture of Hydrogen and Boron 11 which does not.

The implication of it is that we will know in 6 to 9 months if the small reactors of that design are feasible.

If they are we could have fusion plants generating electricity in 10 years or less depending on how much we want to spend to compress the time frame. A much better investment that CO2 sequestration.

BTW Bussard is not the only thing going on in IEC. There are a few government programs at Los Alamos National Laboratory, MIT, the University of Wisconsin and at the University of Illinois at Champaign-Urbana among others.

The Japanese and Australians also have programs.

[Randolph Fritz]

Actually, a fair amount of engineering research has already been completed, and more is accumulating (albeit at a slower rate) all the time. There are many good solar technologies now available. They cannot, however, compete with the embodied energy of petroleum or even coal.

People interested in pursuing this are encouraged to visit:

http://www.eere.energy.gov/
http://www.worldchanging.com/

For solutions to problems in building design, I suggest: Brown & Dekay, *Sun, Wind, and Light*.

[Porlock Junior]

vt codger:
"I never heard anyone who sounded like they had a clue promising fusion power before 2040 give or take a little."

Well, "sounded like they had a clue" allows a certain amount of room for differences in judgment; but how about the president of the Geneva conference on Peacdful Uses of Atomic Energy in 1955? Dr. Homi Bhabha had at least that much cluefulness, along with whatever comes from being on the Indian Atomic Energy Commission. In his opening address he "predicted that within 20 years it will be possible to derive energy from tbe controlled fusion of heav hydrogen nuclei." (Scientific American, October 1955, pp. 29-30)

"Derive energy" is not the same as practical fusion power production, but he was not talking about possiblites 85 years in the future. (If Niels Bohr and I. I. Rabi were doing a nudge-nudge-wink-wink at this speech, it's a pity they didn't let on to the Sci Am reporter, who reported it straight, along with skeptical reactions.

More fun in the Sci Am: "depending on ... assumptions on the level of effort and the difficulties ahead, the time it would take to produce a large prototype reactor could range from as much as 50 years to as little as 10 years." (From the end of a long and detailed article by two AEC experts, February 1971, p. 64)

Both of these are a lot cagier about committing to timetables than the stuff one saw in the popualar press; but neither is talking about 2040 or anything close to it at the high end.

I note, though, that it took more effort to find these than I had expected.

I conclude that the perennial 30-year predictions (quite familiar to this ca codger from the 1950s and on) are more like stuff that the public is told, quite uncontradicted, than the material that was presented commonly to people moderately well informed.

[iris]

Big Tom is exactly right. Just look at solar cost curves and solar will be very (very) big by 2023.

It will be useful in some areas before then, but Brad can go a few doors down and tell Berkley's Environmental Policy physics dude who is always on NPR that no, it is not "here" yet.

Im serious about this. Why can't Brad give this energy "expert" a few econ 101 lessons. In exchange, he can give Brad science lessons.

[Winston]

bigTom,

We'll get to see how both technologies perform on a large scale in the Mojave desert in the next couple of years. You're right that the tracking dish approach is more expensive, but you get a lot more energy from a given amount of collector. Both systems require maintenance and have good and bad points. It will be interesting to compare PG&E's and San Diego's systems (since they're essentially in the same location and being built at the same time). I think that California's requirement that utilities provide 20% of their power from renewables may be the push solar needs to get it into wide use (as California's wind farms were in many ways for wind power). I just hope that California does a better job taking advantage of the technology than it did with wind.

As for seasonality, I think that right now this is an advantage because solar power plants' peak output is at about the same time as California's peak demand.

[vt codger]

Hopefully, we will know about fusion by 2020. ITER should tell us. But even if things go well it'll likely still need two or three decades of engineering to actually come on line. This is field where real, but slow, progress seems is being made. (It could still stall out). But it's not like flying cars or space colonization. Those look to be basically sitting around betting on some deity bringing forth a miracle. 35 years if everything goes well is a long time to just muddle through. The free market will resolve all that of course. Great idea if you are a masochist or were planning to emigrate anyway. Solar and Wind. Good ideas. I don't think they can solve the US energy problems, but they can moderate them just as hydro does. Hydroelectric is, unfortunately, pretty much fully developed in the US. But maybe we can find some additional hydro energy from stationary turbines. Maybe we can squeeze a quadrillion BTU per year from the tides as well (about 1.5% of our needs). And a perhaps a bit more energy from various technologies that work in special circumstances -- a bit of geothermal. Some biofuel. A few more wood powered power plants. We can still conserve quite a bit more. And we could slow down to 50mph where drag is close to negligible. And we can try to pull folks back from the distant countryside at least into villages where basic services like groceries and video rentals can be obtained without a ten mile drive. None of those are solutions. But they'd help. A lot of new fission plants would help. If we can find reasonable sites, and if the environmentalists will get a grip on reality (some have) and refrain from suing them into unprofitability. And if the self centered jerks in Nevada will agree to a reasonable bribe to accept and store nuclear waste in a place where there are not now and are not likely to be in any realistic future, any people. Drilling the ANWR (which I'd normally oppose) might help. It doesn't make sense to run the Alaska Pipeline at half capacity just to protect a few reindeer that may not even be bothered by the drilling. And we and Canada really need to build a gas pipeline to tap the North Slope and MacKenzie Delta fields. And there is always coal. We have a lot of it and it's cheap. Of course, if we insist on using vast amounts of it in a Carbon emission sensitive world, we will very likely find that the US is the target of a world wide trade embargo. Don't laugh, it can happen and it won't be a joke. No magic solutions in any of those. Maybe deposits of deep sea methane clathrates will give us a decades breathing room at some point. Could happen. But it'll just delay the inevitable. It's basically just another form of natural gas with some unpleasant potential for blowing out spectacularly. As Jimmy Carter tried to tell us 30 years ago, this country needs an energy policy other than Don't Worry-Be Happy. Fortunately enough of Carter's policies survived that we have a lot of cars with decent mileage, houses with decent insulation, and industry that is moderately energy efficient. But remember that many other industrial countries have much lower per capita energy usage without much lower standards of living. This is going to become painfully clear if petroleum prices stay high -- which I'm guessing, they will. If, because of energy costs, it costs Americans more to make a widget than it does Frenchmen and Japanese to make the same widget that's not exactly a competitive advantage even for things that the Chinese can't make even cheaper.