Sunday, November 04, 2007

Holding Back Fusion

The Government Accountability Office (GAO) has just released a report on the state of nuclear fusion in America. It is not good. Here is an excerpt from the executive summary.

GAO has identified several challenges DOE faces in managing alternative fusion research activities. First, NNSA and the Office of Fusion Energy Sciences (OFES), which manage the inertial fusion program within DOE, have not effectively coordinated their research activities to develop inertial fusion as an energy source. For example, they do not have a coordinated research plan that identifies key scientific and technological issues that must be addressed to advance inertial fusion energy and how their research activities would meet those goals.

Second, DOE may find it difficult to manage competing funding priorities to advance both ITER-related research and alternative magnetic fusion approaches. DOE officials told GAO they are focusing limited resources on ITER-related research activities. As a result, as funding for ITER-related research has increased, the share of funding for the most innovative alternative magnetic fusion research activities decreased from 19 percent of the fusion research budget in fiscal year 2002 to 13 percent in fiscal year 2007. According to DOE officials, this level of funding is sufficient to meet research objectives. However, university scientists involved in fusion research told us that this decrease in funding has led to a decline in research opportunities for innovative concepts, which could lead to a simpler, less costly, or faster path to fusion energy, and reduced opportunities to attract students to the fusion sciences and train them to fulfill future workforce needs. Finally, while the demand for scientists and engineers to run experiments at ITER and inertial fusion facilities is growing, OFES does not have a human capital strategy to address expected future workforce shortages. These shortages are likely to grow as a large part of the fusion workforce retires over the next 10 years.
Inertial fusion is all about using laser pulses to create enough pressure to cause a pellet of fuel frozen to near absolute zero to implode with enough pressure to fuse the frozen elements. So far there is no plan to turn this into a power producer. Brilliant management. Just brilliant.

In addition they have no plan to meet their manpower requirements by training scientists and engineers. They should try reading The Mythical Man Monthby Brooks. They are setting themselves up for a regenerative failure.

Another inertial approach is the beam or IEC approach. Standing for Inertial Electrostatic Confinement. This uses electrostatic fields to focus and accelerate the beams with various methods used to reduce beam collisions with the accelerator electrodes. The Bussard Fusion Reactor is one example of such a device which uses magnetic fields to reduce losses. There are others.

Then we have the problem of ITER sucking up funds like a runaway Hoover. Choking off other promising approaches. Like alternative magnetic fusion approaches such as the Spheromak. Between all the magnetic approaches such as ITER, other tokamaks, other magnetic confinement approaches, and laser implosion, the budget for various IEC approaches is tiny indeed.

Here is an excerpt from the full report.
The ITER Organization faces several management challenges that may limit its ability to build ITER on time and on budget and may affect U.S. costs. Many of these challenges stem from the difficulty of coordinating the efforts of six countries and the European Union that are designing and building components for ITER and, as members of the ITER Organization, must reach consensus before making critical management decisions. The key management challenges include (1) developing quality assurance standards to test the reliability and integrity of the components made in different countries; (2) assembling, with a high level of precision, components and parts built in different countries; (3) finding a new vendor if a country fails to build a component on time or does not meet quality assurance standards; (4) developing a contingency fund that adequately addresses cost overruns and schedule delays; and (5) developing procedures that describe which countries will be responsible for paying for cost overruns.
I smell a boondoggle. The Euros had this problem with the Airbus A380 Fiasco. So you can't say they don't have enough experience to screw things up. They have had practice.

Here is more about the laser inertial confinement program.
DOE has three separately funded inertial fusion research programs: NNSA’s inertial fusion research activities related to the nuclear weapons program, a High Average Power Laser Program (HAPL) to develop technology needed for energy for which funding is directed by a congressional conference committee, and OFES’s inertial fusion research activities aimed at exploring the basic science for energy applications. Experiments in each of these programs help advance inertial fusion energy, but these experiments are not coordinated and each program has a separate mission and different scientific and technological objectives.
Evidently the European management model is popular in the USA too. Who knew?

I'm not sure exactly what program is being referred to here. It looks like IEC which is distributed among a number of labs and university locations.
As another alternative to both the laser systems and the Z-machine, OFES is funding experiments using heavy ion beams to produce fusion energy at the Lawrence Berkeley National Laboratory. Heavy ion beams are made by a particle accelerator—a device that uses electrical fields to propel electrically charged particles at high speeds. The heavy ions, which are heavier than carbon atoms, collide with the targets and cause the compression and heat needed to release fusion energy.

However, in fiscal year 2006, OFES spent about $21 million to fund 25 small-scale experiments at 11 universities, 4 national laboratories, and 2 private companies to test 7 types of magnetic fusion devices with different shapes and magnetic currents. This level of funding represents a decline over the past 6 fiscal years—from $26 million in fiscal year 2002 to $20 million in fiscal year 2007. University scientists involved in innovative fusion research told us that this decrease in funding was not consistent with a 1999 DOE fusion energy science advisory committee study that recommended OFES increase funding for innovative magnetic research activities. OFES relies on this advisory committee to establish priorities for the fusion program and to provide a basis for the allocation of funding.

However, since that report, the share of funding for innovative research activities has decreased even as funding for fusion research has increased. The share of funding has dropped from 19 percent of the fusion research budget in fiscal year 2002 to 13 percent in fiscal year 2007. In addition, while OFES’s 5-year budget plan shows an increase in funding for fusion research activities in fiscal years 2008 through 2011, most of this funding will be used for ITER- and tokamak-related research activities at the major facilities. DOE officials also told us there are planned increases in funding for innovative devices, but only to maintain the same level of research. According to university scientists, a number of innovative approaches are ready to advance to the next stage of development that would test the feasibility of producing fusion energy or conduct more sophisticated experiments, but DOE has no plans to advance any of these approaches because it may require an increase in funding to conduct more sophisticated experiments. DOE’s fusion energy advisory committee has not assessed the appropriate level of funding between ITER- and tokamak-related activities and innovative concepts since 1999, before the U.S. joined ITER and it became a priority.
So they are choking small money fusion research to pay for ITER. This is nuts when any one of the small approaches migh deliver a breakthrough that could reduce the time and money to develop actual fusion power.

So you get the idea. Typical big governmentitis. The ideas with the most political clout win. Ideas with small experiments, few researchers and low cost results get squeezed out because they lack a constituency.

Pretty much what Dr. Bussard said in the audio found here and the video found here.

If you think it is time for a change, contact your government.

House of Representatives
The Senate
The President

Give them an earful. The future will soon be upon us and we need to be ready.


Anonymous said...

Count your blessings. More research money does not necessarily mean quicker results. It just means more research. Tokamak is an excellent example of this effect.

The ONR has probably provided enough money to find out if the Bussard device has promise, and the results-oriented military is a much better taskmaster than Big Physics.

If the IEC approach has some success, there'll be plenty of people jumping on it, especially since the scale of the equipment required to get started is nearly garage-ish and there's a "cult" following built up by way of YouTube, this site, etc.

M. Simon said...


You are in part correct. I have seen as many projects killed from too much money as from not enough.

I think ONR might not have come up with the money if this project didn't have as many fans as it does.

Anonymous said...


Time to gain a bit of perspective here: The US ITER contribution of $1.9B over the next nine years absolutely pales in comparison to the $465B that has been spent to date on the Iraq war. Most rational people would realize that such a tremendous waste of money for a war we cannot win plays a significant role in undermining progress in many areas of science, technology, and engineering. For proof of this, take a look at the Administration's FY budgets for basic and applied research since '03. They have been gutted.

I also find it amusing that you pick and choose GAO reports to suit your crusade against ITER. Since you are a big fan of the War, missile defense, V-22 Osprey, Aegis destroyers, etc, and in general a consummate ex-military flag waver, you might also consider perusing the scathing GAO reports generated for those items. I suppose you'll just call those reports mere examples of technology "teething" problems and brush them aside.

M. Simon said...

Sure we can't win the war.

Pictures like this are proof positive.


We could easily double the spending on fusion. If it all went to ITER it would be just as big a mistake as if it was never spent. It is not just the dollar amounts. It is the priorities.

M. Simon said...

I don't know those other projects in depth I do ITER.

ITER will get us to break even in 30 year and power production in another 20 - if we are lucky.

The minimum size is 17 GWe (the utilities have no use for plants over 1 GWe and they mostly buy 100 MWe and less plants these days). They will cost too much: well above $6,700 capital cost for 1 KWe - current plants are mostly in the $500 to $1,000 capital cost per KWe output.

You can learn what Vincent Page of GE has to say at Fusion Symposia

Small fusion has way more to offer.

M. Simon said...

Plasma Physicist Dr. Nicholas Krall said, "We spent $15 billion dollars studying tokamaks and what we learned about them is that they are no damn good."

Unknown said...

I think we would all do well to remember that the tokamak started out as small fusion in the USSR. The stellarator was the big fusion concept at the time having itself displaced the magnetic mirror, the previous big fusion concept. The tokamak became the present big fusion concept because it was demonstrably better even at small scale than other contemporary devices. The current tokamak path including ITER can be derailed any time a new concept can be demonstrated to be better. Till then, it is tokamaks and ITER. Stellarators, ONR, levitated dipoles, IEC, RFPs and others need to prove themselves before they can become the new big fusion concept. I predict that even if that should happen someone would complain about how Bussard's concept is getting all the money and choking the development of new innovated concepts that are more promising.

As for ICF, it has no fear of failure and no need to produce a viable plan for energy production. The vast amount of its funding comes from DoD for stockpile stewardship. As long as it meets their goals, it doesn't have to do more than talk about energy.

As for capital cost, that is only part of the story. Fuel costs and cleanup and pollution need to be considered as well. Tokamak power plant studies look more attractive on the cost per kW basis that is within a factor of 2 of current coal and fission plants. Given the increases in Uranium costs and CO2 sequestration and that seems pretty reasonable. Projections will be more reliable once DEMO (~2GW not 17GW- I don't know where 17 GWe comes from?) is built. And considering timelines, new power infrastructures have always a large portion of 100 yrs to establish themselves. Look at any other candidate for providing TW of power to the grid and you'll see similar time scales. It's not easy. And power density and scale matter.

M. Simon said...


Good points.

The 17 GW figure comes from the estimated size of an economic tokamak.

As to electrical production costs Vincent Page of GE has some things to say.

Fuel costs for any of these devices will be less than 10% of operating costs. i.e. economically negligible. (ITER may be an exception due to the requirements for a huge Li6 blanket to produce fuel.

Unknown said...

I did understand from that the 17GW size comes from an estimate of an economic reactor. Its just that that number is about an order of magnitude higher than any other estimate I'd seen. I was wondering what the methodology is. I had read Page's presentation and he doesn't directly discuss that figure though I suppose it could be inferred in someway. The word "economical" hides many assumptions and is not very quantitative. Are we talking about a target CoE or payback period. What are the assumptions on cost of other energy suppliers' costs?

Most reactor studies come up with numbers like 2GW at 0.10 $/kW-hr. I would think a 17 GW plant would exceed the material limits of plasma facing components or at least divertors.

As for ITER, as far as I know, no Li blanket is planned. Perhaps some test modules, but that is all. ITER is an experiment, not a prototype power plant - that would be the follow on DEMO project. ITER will be fueled with externally obtained tritium. Principally from heavy water fission plants.

Happy Holidays. Interesting discussions.

-john (DT)

M. Simon said...

Coal, nuclear, and the best wind sites produce energy in the 2¢ to 4¢ a KWh range. ITER even at 10¢ a KWh is no competition until coal and nuclear resources become really scarce.

In any case who can be sure how much it will cost? Results (power to the grid) are 60 to 80 years out.

I have yet to see any suggestions of 2 GW for power plant size using tokamaks. Do you have a link?

Unknown said...

See the well ARIES studies at
in particular the ARIES-AT study overview. The abstract gives the plant size as 1GW and p19 (the 17th page) gives 4.7 cent/kw-hr not 10 as I said, but I haven't read this in a while and anyway as you point out, these numbers can't be taken as completely accurate. Here is the direct link to the article:
There are also links to studies of non tokamak power plant.

Wind is a great technology, I hope to see more of it. To grow beyond a few percent of load it will need to deal with intermittentcy and storage which will raise the cost significantly. Coal and nuclear are limited by fuel. Nuclear may run out a lot sooner than we expect if the much talked about nuclear renaissance develops and breeders take a long time to significantly multiply fuel. Fuel costs have already increased an order of magnitude over the last few decades for Uranium. Coal, if all costs are internalized and sequestration is required doubles in cost, if it can be made to work.

I agree that for all unproven technologies (for which I include all fusion concepts) cost estimates are at best guidelines for constraining parameters and should not be taken too seriously.

Also I would repeat that ITER is a physics experiment and will not produce power at any price. DEMO is the technology proving project and even it will not be at 10 cents/kw-hr.

I expect that since fuel plays almost no role in the cost of fusion what ever the concept and is in principal not limited, 10 cents may look cheap in a few decades.

Unknown said...

OK, I meant "well know study" and it looks like the link got cut, so here it is in proper htmlese:
The ARIES-AT overview is here and the ARIES website with all references and reports links is here

That's what I get for not hitting 'preview' first.

M. Simon said...

ITER which will lead to no power to the grid for at least 40 rears is just a jobs program for physicists:

Plasma Physicist Dr. Nicholas Krall said, "We spent $15 billion dollars studying tokamaks and what we learned about them is that they are no damn good."


I have a plan (if the experiments green light further work) for delivering the first power to the grid from a Bussard reactor in 3 to 5 years.

With production of such reactors scaled up to 100 GW a year in the 5 years following that. That would be 1,000 100 MW reactors a year. At that rate all the world's power plants could be replaced in 10 years.

For all these devices it is going to depend on capital costs. I can tell you that a production Bussard plant will initially come in at $20 to $50 million. i.e. 20¢ to 50¢ a watt. Currently wind comes in at $1.00 a watt with production capacity at 1/3 name plate rating. Wind at those prices competes with coal and nuclear.

BTW the intermittancy of wind becomes a problem at above 20% of total grid supply. We are at .5% in the USA.

Distributed wind can supply 20% of nameplate rating as base load.

M. Simon said...

Let me add that the Bussard recirculating power will be on the order of 5% to 10% of total plant power.

In addition with the smaller size and no steam plant the time from a plant order to plant delivery will be on the order of 6 to 12 months.

About 80% of the cost of a fission plant is the steam/turbine system. Get rid of that and the costs and delivery schedule are reduced accordingly.

M. Simon said...

A neutron wall load of 1.5MW/m sq?

So far the high end thermal load considered practical is 1 MW/m sq. With the preferred load being 300 KW/m sq.

Also. These plants will be competing with batteries for their Li supply. There is not enough Li in the world to do everything that is planned for that material.

M. Simon said...


The initial Bussard plant will be built with materials currently available. Including super conductors. MRI type magnets (with modifications) should suffice.

If the advances in superconductors posited by the Aries design are possible improvements could be made.

i.e. the Aries design is based on unobtanium. It could take decades alone for the superconductor advances.

And then you have the need to replace the superconductors often due to neutron flux damage.

By burning pB11 neutron fluxes are reduced by at least a factor of 1,000. No way in hell a tokamak can burn pB11. It can barely burn D-T.