UW fusion reactor concept could be cheaper than coal | Hacker News Reader

UW fusion reactor concept could be cheaper than coal

washington.edu

72 pointsmrfusion53 comments

After so many years of hype around fusion it's hard to take press releases seriously. This bit at least offers some hope:

"The researchers have successfully tested the prototype’s ability to sustain a plasma efficiently, and as they further develop and expand the size of the device they can ramp up to higher-temperature plasma and get significant fusion power output."

Unfortunately they don't mention how long the plasma was sustained.

Hype is a double edged sword. Without it, nobody pays any attention and you can't raise money. With it, everyone ends up disappointed and you can't raise money. :)

Yeah, that seems to imply they've already achieved a sustained fusion reaction. Seems like that would be big news on it's own.

The dynomak paper abstract (http://www.sciencedirect.com/science/article/pii/S0920379614...) makes it all sound conceptual/simulated.

If I had a penny for every time I see how some so-called breakthrough technology "could be X", I'd be very rich.

And the bigger the impact after the "could be", the lesser the chance that it'll ever happen.

"University of Washington engineers hope to change that. They have designed a concept for a fusion reactor that, when scaled up to the size of a large electrical power plant, would rival costs for a new coal-fired plant with similar electrical output.

. . . .

"Right now, the UW’s concept is about one-tenth the size and power output of a final product, which is still years away. The researchers have successfully tested the prototype’s ability to sustain a plasma efficiently, and as they further develop and expand the size of the device they can ramp up to higher-temperature plasma and get significant fusion power output."

Well, good luck to them. But this is a press release about a "concept," and it has not been tested out as a source of electrical power (as contrasted with being a test-bed for plasma containment) at all yet, at any scale. Solving the plasma containment problem for a fusion reactor is something I have been waiting for more than forty years. If it was easy, it would have already happened. Maybe the new concept will help solve that problem, but maybe not. At the very least, we still have NO IDEA what the economics of running a power plant will be at scale when using this concept.

Many, many submissions to HN are based at bottom on press releases, and press releases are well known for spinning preliminary research findings beyond all recognition. This has been commented on in the PhD comic "The Science News Cycle,"[1] which only exaggerates the process a very little. More serious commentary in the edited group blog post "Related by coincidence only? University and medical journal press releases versus journal articles"[2] points to the same danger of taking press releases (and news aggregator website articles based solely on press releases) too seriously. Press releases are usually misleading. Not all press releases spin their statements as badly as one very bad example a skeptical blogger commented on,[3] but all of them should be compared to independent sources for a second opinion. Whether in medical research or in reports about new breakthroughs in energy production, press releases often run ahead of the facts.

The most sure and certain finding of any preliminary study will be that more research is needed. All too often, preliminary findings don't lead to further useful discoveries in science, because the preliminary findings are flawed. The obligatory link for any discussion of a report on a research result like the one kindly submitted here is the article "Warning Signs in Experimental Design and Interpretation"[4] by Peter Norvig, director of research at Google, on how to interpret scientific research. Check each news story you read for how many of the important issues in interpreting research are NOT discussed in the story.

[1] http://www.phdcomics.com/comics/archive.php?comicid=1174

[2] http://www.sciencebasedmedicine.org/related-by-coincidence-o...

[3] ?

[4] http://norvig.com/experiment-design.html

FTFY (well, almost) ;-)

Thanks; yes you did do a good job of fixing. (This is why I didn't like following the pattern that users requested me to follow, which I have done at their request, of showing external links as footnotes. Sometimes I misplace or misnumber the footnotes.) I appreciate your help.

Here is footnote 3:

[3] "Anatomy of a Press Release" http://www.skepticblog.org/2012/06/21/anatomy-of-a-press-rel...

You know, the previous time he copy-pasted this same comment on a different article, the footnotes were in there. If it happens again, we'll know where to look.

> If it was easy, it would have already happened.

I'm sometimes a tad skeptical of this framing of things. Everything always seems "easy" in retrospect, so our definition of easy based on what we've already done is highly suspect and distorted. Everything was once unbelievably hard until someone(s) figured it out, at which point it became obvious and easy.

I think what I'm saying is that generating gigawatts of power at all was once just as hard as fusion power is today. Past performance is not indicative of future returns on the downside, either.

If you don't like reading just a press release, the scientific paper describing this concept is at: http://www.sciencedirect.com/science/article/pii/S0920379614...

This article concludes with "The research was funded by the U.S. Department of Energy.", does this impacts the availability of the research to other researchers? Even with the University of Washington in my backyard, I hope the advancements in such an important field are shared as widely as possible.

On the fusion topic, energy.gov reports 63 active US Fusion Program Participants, and over 100 total (including gov't, private companies, international with US pressence)[1]. But last year the Boston Globe reported that MIT was shutting down, "The shutdown will leave only two fusion experiments in the United States, one at Princeton University and the other at General Atomics, a company in San Diego", due to a cutoff of federal funding.[2]

Anyone have any insight on the actual current state of fusion energy R&D in the U.S.?

[1] http://science.energy.gov/fes/research/fusion-institutions

[2] http://www.bostonglobe.com/metro/2013/05/19/fusion-energy-re...

The state of funding isn't great but off the top of my head I can name two locations doing lots experiments: University of Rochester's Laboratory for Laser Energetics and The National Ignition Facility at Livermore National Labs. That Boston Globe article is likely referring to the type of fusion being researched at MIT.

The DOE actually provides research funding for most 'hard' sciences. Particle physics and supercomputing are primarily funded by DOE, for example. Unless it's classified, the research is open.

How this works depends on the department that's funding the research. With NSF/NIH grants, it's at the desecration of researchers on whether they want to apply IP rights. Other departments (mostly DoD) do contracts for research and own the IP afterward, but I'd bet DoE is in the former camp.

The specific contract with the DoE may have defined ownership and distribution of the results, but in general it should be possible for the creators to get IP ownership from the DoE if for example they want to create a company around the technology.

Fission was supposed to be "too cheap to meter."

This seems to be getting way ahead of itself. Seems like they should look to hit important milestones like producing more energy from the reaction than is required to sustain it, before they start worrying about producing electricity for pennies per kilowatt-hour. Walk before you run, and all that.

If we had stayed with fission instead of bailing after relative non-incidents (Chernobyl and Three-Mile Island), wouldn't it have drastically increased the power supply to almost that point?

The casualties, even the most indirectly computed casualties, from nuclear accidents are far, far less than the casualties from coal. It's going to get even worse once the sea levels really rise.

A single fission plant used to supply about a third of my energy; if they had proliferated to the extent people thought they would have, it energy would have, in fact, been a lot cheaper.

Democratic ignorance and regulation killed that dream, though. At least we still have "clean coal."

China which cares little about what people think and has huge air quality problems. Yet it is still building a lot of coal plants instead of nuclear. Which might suggest there not the ideal solution you think they might be. France was actually the most reliant on Fission of any country and even they are stalling far back from there peak despite having few actual problems of any kind.

There are 72 nuclear reactors being built around the world right now. 29 of them are in China. They're building more reactors than the next four builders combined. That is, more than Russia, India, Korea and the United States put together. [1]

The Chinese government has stated that they want to decrease their reliance on coal by building more natural gas, nuclear and solar power capacity. [2]

Aside from a few regions, I don't think they've stopped building coal plants entirely, but the Chinese government pretty clearly likes nuclear power.

[1] http://www.iaea.org/PRIS/WorldStatistics/UnderConstructionRe...

[2] http://www.forbes.com/sites/christopherhelman/2013/09/12/the...

Of course they want to divirsify 77% of the total electrical capacity is coal. More importantly they had to shut down 2.5% of the nation's coal plants (58 units or 14,020 MW of capacity) in 2008 due to coal shortages.

However, in the end these charts says more than I can:

http://en.wikipedia.org/wiki/Electricity_sector_in_China#med...

http://en.wikipedia.org/wiki/Coal_in_China#mediaviewer/File:...

PS: Of total annual production: Hydroelectric = 20.7% in 2006, Coal = 68.7% in 2006. = 89.4% Compared to France which get's 75+% of it's power from nuclear and has little in the way of coal reserves. http://our-future.nl/wp-content/uploads/2012/02/coal_reserve...

Edit: To be clear, even after that construction is over Nuclear is still going to be a small fraction of overall capacity.

"wouldn't it have drastically increased the power supply to almost that point?" No. Building and running fission power plants costs a lot of money, people would stop doing so as soon as they couldn't make a profit out of it.

It's hard to say, but since nuclear power paranoia didn't really kick off until around the 70s, even the most optimistic scenario there would put "too cheap to meter" status at 3+ decades after the first self-sustaining controlled fission reaction in 1942. We're still some unknown time before the first self-sustaining controlled fusion reaction, so discussion of extreme cheapness seems premature.

A lot of fusion researchers think we really ought to consider economics when we choose what reactor designs to pursue. It'd be a shame to spend a lot of time and money on an approach that can't compete in the market.

It would be a shame to spend a lot of time and money on an approach that can compete in the market if it works, but doesn't work.

Maybe it's my programming bias at work here, but it seems like getting anything to work should come before trying to optimize it to be cheaper than anything else out there. Until you have a working system to experiment with, how can you even be sure where the true costs will lie?

It's not that hard to project the costs of a reactor. Tokamaks for example have to be very large for net power, and the chamber has to be surrounded by neutron-shielded superconductors. That's going to be expensive.

Whether the plasma will behave the way we want is a tougher question, but roughly how much a device will cost to build is relatively easy to answer.

There's another benefit to cheaper approaches: you can afford more experiments, and you can probably do them faster. Taking a couple decades and $50 billion to build one experimental reactor is not necessarily the fastest approach to getting something to work. ITER is the ultimate waterfall project.

A lot of the cost will be ongoing operating costs, and a lot of that probably can't be easily predicted without operating experience. Of the expenses involved in running a fission plant (even in a hypothetical generous regulatory regime), how many of them could have been accurately predicted in, say, 1930?

Sure, costs could be somewhat higher than expected. All the more reason not to go with designs that have very high expected costs to begin with.

Neat! I wonder what the next steps for commercial use would be, since the patent is filed through the UW's Center for Commercialization and funded by the US Department of Energy. Could BP or whoever come in and say "we're interested" and throw money at them to build a full scale reactor (in a few years)?

UW student and patentee here. That's what the UW C4C is for—to license UW-developed technology to interested parties. They front the money for filing the patent and then recover the cost when the patent is licensed to an outside party (or by the inventors, if they want to commercialize their work on their own).

What's the rationale for patenting work funded by taxpayers?

The research was funded by the U.S. Department of Energy.

This is definitely something I've wondered about before. Links to some decent references are below. Basically, the idea seems to be that giving patent rights to grant recipients gives them an incentive to develop and commercialize ideas that will benefit the public. Taking a concept from the lab to the marketplace is often a non-trivial affair, and patent rights are a way to drive that process.

[1] http://en.wikipedia.org/wiki/Bayh%E2%80%93Dole_Act

[2] http://www.mofo.com/resources/news/2002/10/the-governments-p...

but doesn't this act also stymie concurrent tech developments from competitors in their respective fields?

Is a monetary incentive for the grant recipient really a greater moving force for the market within a specific industry than capitalistic competition amongst the existing forces in that industry?

Some academics do work on specific problems with a known commercial application, and sometimes this research is funded by commercial entities. The primary goal of most academics isn't to produce commercial products, but in the course of research useful things can pop up. Giving patent rights to grant awardees gives them the ability and incentive to commercialize offshoots of their research. Here's an example (I'm not a biomedical guy, so please excuse the errors):

Dr. Brown researches some disease. He's not particularly trying to come up with a treatment for it, just trying to understand it better. He's in academia, so he needs to publish papers to get tenure and receive grant funding. The NIH funds his research because it's somewhat promising and because he will do a good job educating students. Suppose his group happens to discover a chemical compound that seems to be effective in treating this disease. What happens next depends on whether the university can patent this discovery.

Before the treatment can be prescribed by doctors it needs to receive FDA approval, which requires expensive clinical trials. Without patent protection, no one (neither the university or outside companies) is willing to fund the trials because they won't have exclusive rights to sell the treatment once it's approved. Dr. Brown could work on the trials himself, but that would mean neglecting his other responsibilities. No one has a sufficient financial incentive to make the next step, so the treatment never makes it to the market.

> What's the rationale for patenting work funded by taxpayers?

The rationale for patenting it is the same as the rationale for patenting anything.

The rationale for the funding agency allowing it to be patented and then licensed out by the funded researchers is probably the real question, but there's probably at least an argument that it serves the public good by decreasing the direct public cost of research by sharing the upside reward with those actually doing the research work (which is only proprietary for a limited time, and that's an actually limited time, rather than the limited-in-name-only time that applies to copyrights.)

Patenting the work, then licensing it allows companies to acquire the technology in a very straightforward and well-defined way. You want to build an X, you work out a deal with UW's patent office and they give you the information.

As a side benefit, UW gets some money and (depending how they do things) shares some with the individuals who did the work. Some of my co-workers recently went out for a lunch paid for by one of their patents--it's not a ton of money, but it's a neat thing when your past work buys you lunch.

> You want to build an X, you work out a deal with UW's patent office and they give you the information.

AFAIK, the US patent law says that you can receive a time-limited monopoly for your invention in exchange for describing it in enough detail that others can reproduce your invention. if the patent matter is only reproducible using further information excluded from the patent description then the patent is invalid, IIUC.

Good point.

So rather than saying "we have a totes sweet fusion reactor design, if you pay us we'll show you the schematics", UW publishes it to the world and if somebody thinks it looks good, they pay UW some money to use the design outlined in the patent, which is theoretically sufficient to reproduce the reactor.

I think "a few years" is optimistic. I would assume a lot of the community is holding its breath to see how ITER fares once it's switched on in 8-12 years. Nobody knows for certain if it's possible to generate a plasma on earth that can heat itself, and then produce excess energy. And by that time the anti-fusion lobby will have poisoned the mind of the average power user to believe that fusion = fission.

I doubt it. There isn't much ITER can do that will surprise anybody. Even if it works perfectly, it still won't produce economical fusion. People already know that if you want practical fusion you've got a huge opportunity to bypass ITER.

Wait, there's an antifusion lobby? Who are they?

But yeah, people will still think negative when hearing about "nuclear". Maybe we should change that to "Atomic" to make the distinction and give it a less negative connotation :-).

Exxon, BP, Shell, Chevron...

Those are mostly oil financiers providing capital to contractors who do the work, and project management. They won't care. Too much oil is used for transport and petrochemical (aka plastic) industry.

Some of the peculiar niche uses for petrochems are much less price sensitive than bunker A boiler oil, so they would make more money overall in the long run if people used oil more wisely, which sounds weird but is an economic truth. You care a lot about the price of gas in your car, but you probably don't really care about the cost of plastic in your legos, so if it went up by 10x it would still be a rounding error at the checkout line although in the long run they'd make more money.

The coal companies, on the other hand, are likely concerned.

I can't find any examples of any of them lobbying against fusion (which is unsurprising, since lobbying against something that is currently impossible is not a good use of funds).

If anything, they'd be the first to invest in working fusion technology. They already invest billions in alternative energy research - to secure their own domination in the future, but still...

From what I understand its mostly a materials problem because the plasma just erodes everything.

There's an anti-fusion lobby?

Greenpeace and others. http://www.greenpeace.org/international/en/press/releases/IT...

It was editorial license by johlindenbaum to call it an "anti-fusion" lobby, but we can all agree that there's a pro-fossil-fuel lobby.

Greenpeace is very outspoken against fusion, as they believe it carries the same dangers as fission. http://www.greenpeace.org/international/en/press/releases/IT...

Just the people who've been hearing "fusion is right around the corner!" for the last 40 years.

Meanwhile, Thorium could be done now...

What's stopping it, though? There are new nuclear plants being constructed in Georgia and there's ~$1 trillion budgeted for renewal of the US nuclear arsenal over the next 30 years (see http://www.nytimes.com/2014/09/22/us/us-ramping-up-major-ren...).

This is about as favorable an environment for Thorium reactors as can be imagined, certainly compared to the last few decades. While I think the technology is promising, if it's such a done deal where's the private capital seeking a building permit?

Reply on news.ycombinator.com