IEEE Spectrum October, 2016 - 24

NUmBerS DON'T Lie_By VACLAV SmiL

NUCLEAR ELECTRICITY:
A SUCCESSFUL FAILURE
The age of commercial nuclear elecTriciTy generation
began on 17 October 1956, when Queen Elizabeth II switched on Calder
Hall, on the Cumberland coast of England. Sixty years is long enough to
judge the technology, and I still cannot improve on my evaluation from
about 10 years ago: a "successful failure."
The success part is well documented.
After a slow start, reactor construction began to accelerate during the late 1960s, and
by 1977 more than 10 percent of U.S. electricity came from fission, rising to 20 percent
by 1991. That was a faster penetration of the market than photovoltaics and wind turbines have managed since the 1990s. Today the world has 448 reactors, many with
capacity factors of better than 90 percent. That's the share of the reactors' potential
output that they averaged year-round, producing more than twice as much electricity
as PV cells and wind turbines combined. Nuclear power provided the highest share
of electricity in France (77 percent), but Swiss reactors contributed 38 percent and
South Korea got 30 percent, as did Japan before Fukushima. The U.S. share remains at
almost 20 percent. The failure part has to do with unmet expectations. The claim
that nuclear electricity would be "too cheap to meter" is not apocryphal: That's what
Lewis L. Strauss, chairman of the U.S. Atomic Energy Commission in 1954, told the
National Association of Science Writers in New York in September of that year. And
equally audacious claims were still to come. In 1971, Glenn Seaborg, a Nobelist and
chairman of the Atomic Energy Commission then, predicted that nuclear reactors
would generate nearly all the world's electricity by 2000. Seaborg envisioned giant
coastal "nuplexes" desalinating sea water, geostationary satellites powered by compact nuclear reactors for broadcasting TV programs, nuclear-powered tankers, and
nuclear explosives that would alter the flow of rivers and excavate underground cit-

OCT 2016

NOrTh AmeriCAN

SPeCTrUm.ieee.OrG

OPiNiON

ies. Meanwhile, nuclear propulsion would
carry men to Mars.
The project to generate electricity
from fission stalled during the 1980s, as
demand for electricity in affluent economies fell and problems with nuclear
power plants multiplied. And three failures were worrisome: Accidents at Three
Mile Island in Pennsylvania, in 1979; at
Chernobyl in Ukraine, in 1986; and at
Fukushima in Japan, in 2011, provided
further evidence for those opposed to
fission under any circumstances.
Meanwhile, there have been cost
overruns in the construction of nuclear
plants and a frustrating inability to come
up with an acceptable way to store spent
nuclear fuel. Nor has there been much
success in switching to reactors that
might be safer and less expensive than
the dominant design of pressurized
water reactors, which are essentially
beached versions of U.S. Navy submarine designs of the 1950s.
As a result, the Western public remains
unconvinced, utilities are wary, Germany and Sweden are on the course of
shutting down their entire industries,
and even France plans to cut back. In
August 2016, 61 reactors were under construction worldwide, too few to make up
for the capacity that will be lost as aging
reactors are shut down in coming years.
The only leading economies with major
expansion plans are in Asia, led by China,
South Korea, and India, but even they
can't do much to reverse the decline in
the share of nuclear power in worldwide
electricity generation. That share peaked
at nearly 18 percent in 1996, fell to 11 percent in 2015, and is expected to bump up
to just 12 percent by 2040, according to
the International Energy Agency.
There are many things we could do-
above all, use better reactor designs and
act resolutely on waste storage-to generate a significant share of electricity from
nuclear fission and so limit carbon emissions. But that would require an unbiased examination of the facts and a truly
long-range approach to global energy
policy. I see no real signs of that. ■
illustration by

Chad Hagen

http://SPeCTrUm.ieee.OrG

Table of Contents for the Digital Edition of IEEE Spectrum October, 2016