Practical Nuclear Fusion - how close?

[brenan]

Can anyone say how close we are to obtaining a workable nuclear fusion generating
plant?

I understand the largest trial is being built in France (albeit with many delays apparently)
Does anyone have any understanding as to what the best guess is to this technology
being available to the point where it works (I don't care about financially viable - just working)

I know its hard to guess scientific breakthroughs but could we be talking
10, 15,50 or perhaps more years until we can do this? Is there any "well informed"
guess on this?

 

[mfb]

The most optimistic scenario has commercial power plants operating in ~50 years:
ITER tests from 2020-2035, DEMO construction ~2025-2035 and operation ~2035-2045 (putting power into the grid, but the main purpose will be the experiment), afterwards construction of the first commercial power plants. It is hard to accelerate this - ITER test results are needed to finish DEMO, and commercial power plants will wait for the results of DEMO. And construction is always on the timescale of 10 years.

The most pessimistic scenario has no commercial power plants at all.

The reality? Probably something in between.

 

[brenan]

Thanks for the input mfb - this is very helpful.
Thanks for the link too - I don't know how I missed that one.

 

[the_wolfman]

Can anyone say how close we are to obtaining a workable nuclear fusion generating
plant?

I understand the largest trial is being built in France (albeit with many delays apparently)
Does anyone have any understanding as to what the best guess is to this technology
being available to the point where it works (I don't care about financially viable - just working)

I know its hard to guess scientific breakthroughs but could we be talking
10, 15,50 or perhaps more years until we can do this? Is there any "well informed"
guess on this?

To expand on what mfb said.

The problem is that following break-even we really need a demonstration power plant (DEMO). DEMO is going to be an experiment first, and a power plant second. For this reason, DEMO like ITER is going to most likely be government funded and its going to be expensive. Its going to take some time to assemble the political and economic resources to build DEMO. How long is really impossible to judge.

We can look towards ITER to get an estimate. The USA and Russia first stated talking about ITER in 1987, and the current timetable for ITER has its first plasma in 2020, but the first D-T plasmas won't be until 2027. That's 40 years from the first talks to break-even (at best).

Assuming that DEMO follows a similar time line starting at the point of break-even. Then we can expect some power to the grid around 2067 for a magnetic confinement reactor. Now breakthroughs in magnetic confinement fusion or highly motivated countries can both accelerate that time line.

Another route to fusion is inertial confinement fusion. Along this lines the results at NIF have been disappointing. But they are already doing experiments at conditions relevant for break-even. If they make a significant breakthrough, then serious talks on a demonstration inertial power plant might be begin in earnest, and we might be looking at an experimental power-plant around 2053.

Again this is all speculative, and I stress that the time-line depends as much on politics as it does on the science and technology.

 

[bobsteinhaus]

I really believe that fusion will come in forms smaller, cheaper, and sooner than most energy analysts now anticipate.

Some people like the idea of fusion and an abundant energy generator that produces non-radioactive helium as its nuclear waste. What is less exciting is the fact that only one type of nuclear fusion has ever produced net energy (more energy out of the fusion experiment than it takes to run the fusion experiment). That form of fusion is "impure fission-fusion" that uses the power of fission to create the conditions needed for fusion.
There is real reason to be optimistic about fusion today (not just hype and rah-rah). Several small fusion experiments are getting genuinely close to achieving "break-even" and fusion ignition.
Here is a link to an article at The Next Big Future Blog that does a good job reviewing the current status.
http://nextbigfuture.com/2013/05/nuclear-fusion-summary.html

Fusion has a real advantage over fission today as regards the current level of regulatory obstruction from NRC.
While this may not seem significant, it could make a real difference in how quickly fusion will emerge as a commercial technology in forms people will want to build and own to produce power.


Some of the small, low cost fusion approaches are getting genuinely close to fusion ignition and practical power generation [1].

One of the keys to practical production of fusion power is to meet the basic conditions required for fusion (the famous Lawson criteria of temperature, plasma pressure, and confinement time). At least one small fusion experiment at Lawrenceville Plasma Physics headed up by Dr. Eric Lerner has at this point technically already met the minimal conditions for D-T fusion.

D + T -> 2He-4 (3.5 MeV) + neutron (14.1 MeV)

The Lawson criterion for fusion ignition and break-even with D-T fusion is about 4 x 10^15 keV s/cm^3.

Eric Lerner of Lawrenceville Plasma Physics reports in peer reviewed literature that his experiment has achieved 5 x 10^15 keV s/cm^3

This report makes things sound like LPP has achieved the long sought fusion goal of break-even energy already, but (as usual) things are more complex than a single number like Lawson criteria.

First off, the temperatures LPP has achieved are actually too hot for ideal D-T fusion. At 150 keV they would need longer confinement times and/or higher densities than they have
achieved (or so far announced) to reach break-even. So if they were actually looking to produce fusion ignition via D-T fusion they would aim to make their plasma a little cooler and a little denser.

In theory, LPP should not be that far away, but they are not putting effort into D-T or D-D fusion at this time (although in the past they have made many runs using both D-T and D-D fusion). D-T fusion is the fusion reaction that is easiest to achieve - but LPP is not currently interested in going that direction.

That's because LPP isn't really interested in D-T type fusion. The "problem" with D-T fusion is that a lot of the energy in D-T comes in the form of 14.1 MeV high-energy neutrons, which tend to make reactor materials highly radioactive via neutron activation. In theory you can capture that energy the same way you do with a conventional nuclear reactor (that is, by using the radiation to get something like liquid Lithium hot and using it to boil water or heat molten salt to run a turbine). Part of the problem with D-T fusion is the tritium itself, which is radioactive unlike deuterium fusion fuel, and LPP doesn't want to get into all those radioactive-materials handling issues. Tritium is also very expensive to purchase (about $30,000 per gram as estimated by Los Alamos National Lab).

So LPP has chosen to set their sights just a bit higher. They plan to bypass easier to achieve D-T and the cheap and very sustainable D-D fusion reaction in favor of aneutronic fusion (for me, the is a questionable and perhaps unfortunate choice - neutronic fusion producing a huge number of high value neutrons is valuable and could be used for applications like nuclear waste burning, medical isotope production, and manufacture of fuel for fission reactors, as well as producing electricity from fusion).

The fusion reaction LPP prefers is called p-B-11, which uses conventional hydrogen (which becomes a bare proton, p, when ionized) and boron-11, which is over 100X more difficult to ignite in a tokamak (and is about 20X as hard to ignite in a properly designed Inertial Confinement Fusion device).

[1] - LPP at Google's Solve for X Conference - still leading the field - http://www.lawrencevilleplasmaphysi...olve-for-x-conference-still-leading-the-field

 

[mfb]

Here is a link to an article at The Next Big Future Blog that does a good job reviewing the current status.

That looks more like a wish list than the current status. If you ask the developers, their devices are just a few million dollars and some years away from the breakthrough: infinite, cheap energy, probably giving world peace and a cure for all diseases as well ().

Despite FF-1’s low cost of less than $1 million, the results LPP published showed FF-1 has achieved two out of three conditions—temperature and confinement time—needed for net fusion energy.

Two out of three conditions might sound nice, but a simple bottle of hydrogen achieves two out of three as well (density and time).

Reaching breakeven is nice, but not the real goal - it is just a necessary step.

 

[adlhancock]

Engineer working on both DEMO and ITER here. If you want a detailed explanation of where Europe sees the future of magnetically confined fusion going, there's a link here to the EFDA roadmap. The bottom line is electricity from DEMO to the grid in the 2040s, with commercial plants being built shortly afterwards.

 

[phinds]

Engineer working on both DEMO and ITER here. If you want a detailed explanation of where Europe sees the future of magnetically confined fusion going, there's a link here to the EFDA roadmap. The bottom line is electricity from DEMO to the grid in the 2040s, with commercial plants being built shortly afterwards.

Your link gives a 404 error

 

[LabratSR]

Your link gives a 404 error

Try this

http://www.efda.org/downloads/fusio...-realisation-of-fusion-energy/jg12-356-web-2/

 

[phinds]

Try this

http://www.efda.org/downloads/fusio...-realisation-of-fusion-energy/jg12-356-web-2/

Gives me nothing but a title ... no article and no link to one.

 

[LabratSR]

Gives me nothing but a title ... no article and no link to one.

The title is clickable, at least for me.

 

[phinds]

The title is clickable, at least for me.

Very weird. The link now gives me a different look/feel and the title IS clickable.

EDIT: BUT ... I click on the title and I'm just taken to a page that has the titlle and no further info and no link.

 

[LabratSR]

Very weird. The link now gives me a different look/feel and the title IS clickable.

EDIT: BUT ... I click on the title and I'm just taken to a page that has the titlle and no further info and no link.

Weird. Should be 5 links on that page.

http://www.efda.org/2012/05/36-of-the-energy-market-for-fusion/

http://www.efda.org/2012/04/european-energy-conference/

http://www.efda.org/2012/06/energy-goes-digital/

http://www.efda.org/2012/10/former-jet-scientist-appointed-fusion-for-energy-director/

http://www.efda.org/2007/07/didier-gambier-appointed-as-director-of-fusion-for-energy/

 

[mfb]

It looks like the forum is adding weird stuff to the URL. Click the first link, remove everything behind "pdf" and it works.

Alternatively, try this:
http://www.efda.org/wpcms/wp-content/uploads/2013/01/JG12.356-web.pdf

 

[adlhancock]

Apologies for the dead link.
Let's try again: http://www.efda.org/wpcms/wp-content/uploads/2013/01/JG12.356-web.pdf
or just google "EFDA roadmap", as I did.
EDIT: It's a fairly technical document, but the executive summary's been written for MEP's so should be
reasonably accessible.

 

[Greg Bernhardt]

This recent phys.org article explains a bit

Nuclear Fusion Reactors 'Possible' Report Scientists - But Not Yet
http://phys.org/news/2013-09-fusion-weve.html

 

[bluecore]

Well, it was 20 years 40 years ago! :-D But this time, this very time - it really looks to be journey for just another 20 years... probably.

 

[adlhancock]

This recent phys.org article explains a bit

Nuclear Fusion Reactors 'Possible' Report Scientists - But Not Yet
http://phys.org/news/2013-09-fusion-weve.html

Hmm. Yes, though it's worth differentiating between inertial (e.g. NIF) and magnetic (e.g. ITER/DEMO) confinement fusion. I'd argue from a clearly biased, though somewhat informed viewpoint that the challenges for NIF are the same as for ITER, with a few notable additions. NIF, has uses for modelling nuclear weapons detonations, which is partly why they're well funded in the US compared to magnetic fusion. I think the "we will be able to do it, trust us" is interestingly timed in the light of recent failures, and I read this article as actually saying that, "it's much harder than we expected" rather than, "we've made real progress".

 

[brenan]

I've been following the thread with interest (I already have what I needed) and just thought
I'd mention my researches suggest there is as much "contract protecting" waffle involved in fusion as there is in
many other areas of "science."

This page http://www.efda.org/fusion/ in particular is very dissapointing. It reads like a schoolboys
homework project to promote a favourite toy. The figures (both of them) mentioned are very much at odds
with the infromation I've gleaned by asking specific questions in places I trust.

There is also the matter of fusion being sold as a "clean and safe" alternative to fission. It seems that
is being oversold too.

I'm willing to bet it will be at least another 100 years before any significant contribution is made
by any nuclear fusion technology.

And that presents us with serious problems - not least of which is world war three as the rush
for arctic fossil fuels gets out of hand.

Whilst it all looks a lot better than building fission reactors I am no longer convinced it will be the
solution we need. Unfortunately I can't see a viable alternative. And that seems to be its only selling point.

 

[adlhancock]

brenan, I'm sorry to see you end up with such a disappointed view of fusion, so will attempt to very briefly address your individual conclusions:

1. "contract protecting" - indeed. Sadly the nature of things when applying for limited funding in today's economic climate. It's worth digging through the dross and surface "polish" however. There are roughly
1000 people here at Culham who depend on the often fickle short-term views of European and global
governments for the security of their work and careers!

2. The EFDA website - (Being careful, since they indirectly pay much of my salary!) Think carefully about the intended audience of this particular page, in the light of where it's coming from. Enthusiasm and simplicity are necessary for many non-specialist readers. Remember too that those writing these documents and proposals are often scientists first and communicators second (or third!). In the case of organisations like EFDA, add the fact that there aren't all that many native English speakers working there.

3. "clean and safe" oversold? - There's a lot more depth behind the scenes. Spend some time looking particularly at the EFDA roadmap, and if you can get hold of it, the "PPCS" (power-plant conceptual study) work done a couple of years ago, which covered lots of ground here. It included a range of suggested conceptual designs, as well as thinking about the details of actual costs, risks, material availability, timescales, economics and the like. The inherent safety does seem "too good to be true" at first glance, but as engineers we're taking this seriously.

4. "100 years" - I'll take your bet! ITER's frustrations are primarily organisational rather than fundamentally technical, and DEMO's challenges are "can we make it economically viable" rather than "can we generate electricity from fusion at all". If you say 100 years before a significant percentage of the world's power is generated via fusion, you may be right. I'd say half that before we put the first generation of commercial plants on the grid in half a dozen different countries.

Lastly "the solution we need" - I agree, somewhat, since I'm sure that fusion will only ever be a part of a wider landscape of energy sources, mixing all the benefits of renewables, advanced fission, and probably even some heavily modified fossil fuel tech in the medium term. I'd argue that fusion has significant strengths that make it worth investing sensibly in bottoming out whether it's worth doing long term.

 

[nikkkom]

I don't know, the monster machines like ITER don't look economical to me, even if they would have zero radiation-related problems (e.g. disposal of the activated materials).

All this while solar and wind becoming steadily cheaper per watt, and no signs of Sun dimming or Sahara disappearing. I have a feeling I know who will win this one.

 

[adlhancock]

@nikkkom, it's not really a question of "winning", as far as the strategists in the fusion community see it (apart from a few sensationalist outliers). The progress in solar and wind tech (two former wind engineers sit in my office), is obvious and should ensure that they take a central role in the future energy landscape, but there's definite advantages to having additional larger-scale, base-load, always-on sources too, which is where fusion and genIV fission would be looking to sit. One potential game changer in this area is energy storage; if there are massive improvements there the need for steady-state supplies reduces somewhat, but never goes away completely.

I'm a big fan of David MacKay's work at http://www.withouthotair.com/. The site's not "pretty" and I wouldn't support everything he says unreservedly, but the numbers are well thought out.

EDIT: With economics, that's one of the challenges we're looking at. Cheaper fusion plants will produce more expensive electricity and v.v., so it's an optimisation exercise. Best estimates put fusion power as "competitive" based on forecasts of future energy costs. I don't have the numbers immediately to hand, and should probably get back to actually working!

 

[nikkkom]

I'm a big fan of David MacKay's work at http://www.withouthotair.com/.

Great work!

However: "Photovoltaic panels with 20% efficiency are already close to the theoretical limit".

Looks like "theoretical limit" has been raised since then ;)

 

[mfb]

"how close" inertial fusion is? NIF can deliver something like 3 laser pulses per day, and needs no-idea-how-many GJ to produce that 1-2 MJ per laser pulse. I don't know how many orders of magnitude are missing to cover the real energy needed to generate the laser pulses, but the repetition rate has to go up from some shots per day to some shots per second - 5 orders of magnitude.

All this while solar and wind becoming steadily cheaper per watt, and no signs of Sun dimming or Sahara disappearing. I have a feeling I know who will win this one.

Even if photovoltaic cells would be free, the sun won't shine at night.

Looks like "theoretical limit" has been raised since then ;)

Cells with three or even four layers are a completely new type, they have higher limits.