Tens of billions of
dollars have been invested over the past forty years in trying
to make controlled
nuclear fusion a practical
reality. The ITER experiment was initially expected to cost
about $3B, but the latest estimate is $14B. After 15 years
of design effort, construction may begin on ITER within a year.
Preliminary testing of the vacuum vessel and magnets is expected
in 2018, and ignition with real fuel is projected for late
2026. Projections in early 2008 were for the next and much
larger experimental reactor, DEMO, to begin preliminary testing
in 2035; and testing of the first pre-commercial reactor was
expected in 2045. The projected test date for the first pre-commercial
reactor seems to get about 2 years closer every decade. At
that rate, perhaps it will really happen in about 300 years.
The ITER demonstration is projected to be able to generate up to 500 MW
of thermal power for extended periods of time, which in principle could
be used
to produce ~140 MW of electrical power. (The cooling water leaving the
blanket would probably not be able to exceed 420 K, even briefly, without
major changes, so conversion efficiency would be very low.) The plasma
heating power going directly into the reactor will be only 50 MW, but operating
the neutral beam injectors, plasma rf heaters, cryo-refrigeration
system, etc. will probably consume
about 110
MW of electrical power. So its net output (assuming additional power generating
equipment were included) could be about 30 MW. With continued progress,
10
years into operation, it could be operating nearly half time for a few
months before it would need to be shut down.
The intense neutron irradiation of the structural materials
in a
fusion reactor will quickly turn the reactor into a huge, dangerous mess
of extremely radioactive material. To minimize this problem in ITER, it
will actually
operate with fusion fuel (tritium and deuterium) for only a very short
time after being in service for at least 10 years with "play fuel" (protium,
1H) before firing it up and making the reactor
dangerously radioactive. Still, after decommissioning and 50-years of
cool-down, it will leave behind thousands of
tons of radioactive material for long-term radioactive
waste disposal. Our analysis indicates the waste disposal problems with
tokamaks will be greater than with fission reactors, though some others
disagree.
A large fraction of ITER's cost will be the 1500 tons of Nb3Sn
superconducting magnet windings and other mature, expensive technologies
(cryo-refrigerators,
rf transmitters, magnet pulsed-power supplies, particle beam accelerators,
turbo-molecular pumps, etc.), where it is unrealistic to expect that major
cost reductions can ever be achieved. However, it does seem realistic to
believe
that enough
progress will be made to enable the much larger DEMO to be built 25-30
years from now for about 4 times as much and generate 500 MWE about
half-time for several years.
Hence, a rough estimate is that the capital investment that will be needed
for the first 500 MWE fusion plant in 2035 would be $50B, and
it would produce about 1 GW-year of electrical energy. The operating costs
alone of ITER are expected to be approximately $1M/day, and it would have
a maximum
electrical output (for a few months) similar to that from a 20 MW natural
gas power plant. The operating costs for a 20 MW gas power plant 10 years
from now are expected to be less by a factor of about 50.
The other general approach to fusion that has received enormous DOE and
DOD support over the past four decades is Inertial Confinement Fusion (ICF).
The specific method that has received (by far) the most support is laser
fusion, where, for example, 192 ultra-high-power lasers zap a very small
bead of fuel in an attempt to fuse solid deuterium and tritium into helium.
The National Ignition Facility at Livermore National Laboratory has received
a lot of press recently, as recent experiments suggest the prospects for
achieving “ignition” in 2010 are fairly good.
And exactly what would “ignition” mean from a energy perspective?
It would mean that they might be able to claim that this $3.5B experiment has
allowed them to generate 20 MJ of fusion heat output (similar to about a nickel’s
worth of coal) from an input laser beam pulse energy of 8 MJ – which probably
required 10 MW of xenon flash lamps that used a thousand times more energy than
was produced by the “ignition”. Spending $3.5B to generate the energy
in a small lump of coal per day can’t begin to make sense, so this work
is “justified” from a defense rational.
And what about the other ideas that have been hyped, such as bubble fusion, current-pulse
compression of fine wires, cold fusion, and mechanical piston compression. All
have been soundly refuted as having no potential for net energy production.
A $3B investment in WindFuels could produce 40 GW-years of liquid-fuel
energy over the same time frame. Thus, a reasonable estimate is that even
if magnetic confinement fusion works as projected, it will be over 300
times
more expensive
than WindFuels.
Fusion has no chance of making a significant contribution to our energy
needs for the next six decades. Hence, it seems extremely irresponsible
to invest $70B into fusion experiments and research over the next 30 years.
It would be better for DOE to apply the money to a new budget for unsolicited
proposals. Such a program could attract
innovative solutions to our energy and climate challenges (such as WindFuels).
References:
Controlled thermonuclear fusion is theoretically
impossible:
F. Cellier, “The Future of Nuclear Energy: Facts and Fiction – Part
IV: Energy from Breeder Reactors and from Fusion?”, 2009,
http://europe.theoildrum.com/node/5929
http://www.nature.com/news/2009/090527/full/459488a.html
http://en.wikipedia.org/wiki/Thermonuclear_fusion
http://en.wikipedia.org/wiki/Fusion_power
http://en.wikipedia.org/wiki/Nuclear_fusion
http://www.iter.org/
WE Parkins, “Fusion Power: Will it Ever Come?”,
Science, 311, 1380, 10 Mar, 2006.
D Clery, “ITER’s $12B Gamble”, Science, 314,
235-242, 13 Oct., 2006.
The following reference contains a lot of interesting history,
but it is of little value from a scientific perspective:
Charles Seife, “Sun in a Bottle – the Strange History of Fusion
and the Science of Wishful Thinking”, Viking, 2008.
