What is the trick with nuclear fusion?

[toadspittle]

@Alex_Dubinsky

Isn’t the last word on cold fusion still, “there’s something really interesting going on, we just wish we could make it happen regularly enough to make sense of?”

It looks that way. This Wired story covers the whole thing in great detail. It seems that while something is happening, the results are so irregular and so hard to reproduce that it’s not worth any individual lab’s time and money to keep at it.

[Researchers Edmund and Carol Storms] succeeded partly because they were inhumanly persistent. "We tried every conceivable permutation of every variable we could think of. We ran 250 experiments, taking one whole year, and I think 13 made excess tritium.

…

He shows me a box containing 90 little tags of palladium. “I’ve learned,” he says, “how to determine in advance whether a sample will work. I can predict it with about 50-50 accuracy, where it was a 1-in-20 chance before.”

He analyzes various properties of the metal, such as its tendency to crack, which limits its absorption of deuterium. “That’s what makes cold fusion so nonreproducible,” says Storms. “You have to load the palladium with very high concentrations, and many samples simply won’t tolerate it.”

…

At Los Alamos, Tom Claytor likewise is thwarted by lack of money. He would like to see a massive trial-and-error program to test every possible palladium alloy, since tiny impurities seem to catalyze dramatic performance gains. “This is how ceramic superconductors were developed,” he points out, “by testing 5,000 different compounds.” But no laboratory wants to mount such an effort for cold fusion.

Consequently the field is languishing, while its key scientists grow older, and few newcomers venture in.

[Stranger_On_A_Train]

@MarcusF

Stranger - I think you are being too pessimistic. IANAPhysicist but my understanding from the physicists is that the problems are understood and routes to resolving them can be seen. Thislink (Note: pdf file) is to a paper setting out a fast track route to commercial fusion power production. It dates from early 2005 - most of the work was done in 2004 - but it gives a pretty good summary of the issues to be addressed and the facilities needed to work on them.

In the three years since then no show stoppers have turned up and a start has been made on building ITER and designing IFMIF. As always there are technical - and organisational - problems in constructing something as complex as ITER, particularly as an international collaboration, but it will happen - too much political capital is tied up in it to let it fail now.

Certainly the basic physics with regard to the conditions necessary to create fusion reactions is known; there are, however, many unsolved problems in the control of a high temperature, high density charged plasma, along with the practical engineering issues of developing materials and processes that can cope with the kind of environments imposed by such a system. These are all extensive research programs, not niggling details. Even the UKAEA “Accelerated development of fusion power” study you site places operating fusion power plants at 40+ years, and the ITER doesn’t even come on line until 2015; this, of course, assuming that all technical hurdles are cleared on schedule; fourteen of the fifteen major issues they list have “solution is a requirement” for a functioning commercial power plant.

I don’t know how much faith you have in a large project meeting the deadlines of high level conceptual schedules, but my experience is that anything of significance never comes in on time and in budget. This is particularly true when trying to make projections involving unsolved or unforeseen problems.

Stranger

[MarcusF]

*Originally posted by * Stranger On A Train:
These are all extensive research programs, not niggling details.

Sorry - I probably wasn’t as clear as I should have been as I posted in a hurry. I didn’t want to suggest developing commercial fusion power was trivial with only niggling details to be dealt with, I just wanted to slightly counter-balance your post which I felt made it seem like an impossible dream and not worth spending time and money on.

The rewards are so great - with 6, 7, 8 billion people all wanting a standard of living approaching that we have in the West today - that it is worth doing the hard research and difficult engineering to deliver a base-load power source that has virtually unlimited fuel, does not emit green-house gases, and does not leave a legacy of extremely long term radioactive waste. Fusion is not a solution for our immediate problems but there is a pretty good chance that it might be a solution for our children (or more likely our grand-children!). If there were known show-stoppers you would want a different approach - more basic research oriented - but my point was that we are now at the stage where what is needed is hard engineering and industry getting involved.

No, I don’t have much faith in large projects being delivered to time and budget - I would be amazed if ITER is operating before 2020 - but all the more reason for getting going now. What that “Fast Track” road map in the paper I cited indicates to me is that, if we want to have commercial fusion around 2050, there are a number of things to learn and techniques to develop - and we won’t do this if we don’t start soon.

[Stranger_On_A_Train]

@MarcusF

The rewards are so great - with 6, 7, 8 billion people all wanting a standard of living approaching that we have in the West today - that it is worth doing the hard research and difficult engineering to deliver a base-load power source that has virtually unlimited fuel, does not emit green-house gases, and does not leave a legacy of extremely long term radioactive waste. Fusion is not a solution for our immediate problems but there is a pretty good chance that it might be a solution for our children (or more likely our grand-children!). If there were known show-stoppers you would want a different approach - more basic research oriented - but my point was that we are now at the stage where what is needed is hard engineering and industry getting involved.

No, I don’t have much faith in large projects being delivered to time and budget - I would be amazed if ITER is operating before 2020 - but all the more reason for getting going now. What that “Fast Track” road map in the paper I cited indicates to me is that, if we want to have commercial fusion around 2050, there are a number of things to learn and techniques to develop - and we won’t do this if we don’t start soon.

I don’t disagree with any of this; ultimately, nuclear fusion holds the promise to provide clean (or at least much cleaner) and nearly unlimited energy for the foreseeable needs of humanity. It will also, provided it can be made sufficiently compact, provide the means for human exploration and habitation of planetary space, and perhaps even interstellar journies. But any credible timeline must be caveated by the reality that we haven’t solved all problems with regard to commercial or even self-sustaining fusion power production, nor do we have in hand a certain way of doing so. What we do have is a defined set of gateposts (in current systems engineering parlance) that let us know how close we are to achieving practical self-sustaining fusion, but not precisely when or indeed if we can solve them with extant or speculative near-term technology. Developing practical fusion power would be worth almost any price for the bounty of energy it would provide, and the freedom from hydrocarbon or fissile fuels that are located in limited and often politically inconvenient areas of the globe. But it’s not a straight, unmetered road to get there.

Stranger

[MarcusF]

We seem to have reached complete agreement

[Alex_Dubinsky]

Thanks, toadspittle! That Wired article is an amazing read.