The subject of Thorium reactors keeps coming up and there are always serious errors and misconceptions in almost every treatment. These problems are often not caused by special characteristics of Thorium reactors but a lack of background in the science behind the entire subject. That is too large a topic to treat in a single post so I'm not going to even try. Much as it pains me to do so when such issues arise, here I will just state conclusions without backing them up.
What follows, as indicated in the title, is a general overview of thorium
reactors with special attention to some of the more common errors and
misconceptions.
For some background I'll start with a very high level description of
thorium reactors. They are devices that use nuclear fission to release large
amounts of energy that is converted to electricity by producing steam to drive
turbines. Nuclear fission is the process where a large nucleus breaks apart,
usually into 2 unequal pieces and a few neutrons. Usually this is caused by the
absorption of a neutron. The neutrons then go on to hit other nuclei causing
them to fission as well. In a few cases the number of neutrons is large enough, and the likelihood that absorbing a neutron will produce fission is high enough, that a chain reaction will occur. Such materials are called
"fissile".
Another important classification is "fertile" substances. When
these absorb a neutron they are, usually after a series of decays, converted to
a fissile substance. The fissile material in a thorium reactor is 233U,
an isotope of uranium. This does not exist in significant quantities in nature so it is produced in the reactor itself from 232Th which is fertile. Reactors
like this, where neutrons are used both to produce fission and to breed new
fuel, are called breeder reactors.
There are a few general categories that cover the most egregious
problems in treatments of thorium reactors:
Reactor Safety v Fuel Choice
Reactor Design v Fuel Choice
The effect of fuel choice on waste
Misinformation
I'll take them in reverse order.
One of the errors that motivated me to write this was a misstatement of the
half-life (the time it takes for ½ of the material to decay) of thorium.
Thorium on Earth is 99.98% 232Th. This isotope has a half-life of
around the age of the Universe, 14 billion years. I've seen assertions that thorium's half-life was short, on the order of dozens of years. I've also seen
assertions that its half-life is MUCH longer than it is. I'm a bit troubled by
these errors because the real number is so easy to find. I've also seen the
assertion that since thorium doesn't need to be enriched it is less useful as a
source of nuclear weapons material. This is wrong in multiple ways.
I also commonly see that the only reason thorium reactors weren't developed
is that they don't produce material for nuclear weapons. It is true that thorium
reactors don't easily produce weapons grade material but neither do any
reactors that are designed for power production. Weapons material is produced
in reactors optimized for that purpose, in the case of plutonium. In the case
of uranium this done with centrifuges. Using the waste, or partly spent fuel,
from a power reactor would be a difficult way to make weapons grade material. A
significant design restraint of early non weapon producing reactor systems was
to power submarines. Uranium is a far better choice than thorium for this. The
relative likelihoods of alternate histories are difficult to quantify but,
that's one of the major reasons that things worked out as they have. Another
factor is a detail in the breeding of 233U that caused Enrico Fermi,
one of the major figures in the development of nuclear technology, to disfavor thorium.
A solution to this problem was found but not until attention had focused on other
paths.
Much of the discussion around nuclear reactors in general is really about
the waste produced. Much of that discussion is based on several
misunderstandings about radiation and the health dangers associated with it. As
I said above I'm not going to even try to explain my conclusions but here's the
bottom line. The idea that radioactive materials must be sequestered until
their emissions can't be detected is silly. We live on a radioactive planet.
Even more radiation is added from space. From a health point of view it is
clear that the amount of background radiation we get from the average location
on Earth is not a significant health risk. Many evaluations of the danger of
nuclear waste are based on ignoring this fact. This is true of all nuclear reactors, no matter what fuel is used.
It is often stated that the waste problem from thorium reactors is less dangerous. One specific claim is the assertion
that thorium reactors produce waste that is less radioactive, with the
implication that this makes it less dangerous. The first problem with this is
that it isn’t true. The waste produced by Thorium reactors is more radioactive than from existing
reactors. But this turns out to be a good thing.
The major issue surrounding nuclear waste is that it needs to be kept
secure until the radiation gets to an acceptable level. As mentioned above,
there is disagreement about what that acceptable level is. But there is no
disagreement that the important characteristic is the half-life of the
material. Let’s say you have a particular number of atoms of a radioactive
material. If they decay in a short time there will be lots of radiation
produced in that short time. So a short half-life means a highly radioactive
substance. The longer the half-life the longer it needs to be sequestered to
allow it to decay away. So, from a waste sequestration point of view, highly
radioactive material is better than lower levels of radioactivity. The facts
here are good for thorium reactors.
The core of an operating nuclear power plant has three classes of
materials:
1) The fuel itself.
2) Substances created when the fuel fissions, called fission byproducts. These
have half-lives ranging from tiny fractions of a second to around 10 million
years.
3) Substances formed as the result of unwanted nuclear reactions. The most
important of these are mostly elements in the actinide series of the Periodic
Table these are called actinides. These have half-lives ranging from around 6
to several billion years. Thorium reactors create less actinide waste. It
should be noted however that thorium reactors still produce waste products
with very long half-lives, up to about 17 million years, but only in quite
small amounts. This is contrary to the commonly seen assertion that no such
waste is produced in thorium reactors.
However, if we take a sensible approach to how long radioactive materials
need to be isolated thorium reactors are a big improvement. The reduction in
the creation of actinides is the primary reason for this. Despite the existence
of some long lived substances, the waste from a thorium reactor is about as
radioactive as naturally occurring ore in only a few hundred years as opposed
to many thousands for current designs.
Another issue is the overall design of the reactor. For this discussion I'm
referring to a very basic question: Is the fuel solid or liquid? For technical
reasons beyond the scope of this treatment thorium reactors are best implemented
as liquid fueled or Molten Salt Reactors (MSR). In particular reactors that use
molten fluoride salts known as Liquid Fluoride Thorium Reactors (LFTRs) are the
ones getting the most attention. It is important to realize that MSR reactors
can be operated with any of the nuclear fuel cycles that have been studied. MSR
reactors are more complicated than current LWR designs but they have several
advantages. Most of the advantages cited for thorium reactors such as more
complete utilization of fuel are actually benefits of being a MSR and unrelated
to the use of thorium as a fuel.
One of the most common misrepresentations in this category is that a LFTR
was built in the early days of nuclear reactor development. This is NOT true.
An MSR was built and for a time it was fueled with the same isotope of uranium that
fissions in a LFTR. The second step, the breeding of the fuel was also
demonstrated, but in a different reactor. This is valuable experimental
verification that a LTFR might be a workable design but no LFTR was actually
built. This misinformation is used to present LFTRs as proven design that is
ready to commercial use.
The danger of a meltdown is probably the best known and most common concern
with nuclear reactors. A meltdown occurs when the energy produced in the
reactor isn't removed fast enough and it heats up to the point that the
physical structure is damaged or destroyed allowing the highly radioactive
fission byproducts to escape. The basic design of a LFTR makes it very easy to
ensure that this cannot happen. This is often mentioned as an advantage of
thorium reactors. This is wrong for at least two reasons. First, all MSRs share
this characteristic. This is true no matter what nuclear fuel was used in the
reactor. Second, it is also possible to design a solid fueled uranium reactor,
very much like current reactors, that is inherently extremely resistant to
meltdown.
In conclusion, much of the information about thorium reactors that you’re
likely to run across is wrong. But it is true that thorium reactors, LFTRs in
particular, are an interesting technology with many potential benefits that show
lots of promise and they address the reasonable concerns with current commercial reactors. However many of
those benefits are not because they use thorium and can be applied to reactors
that of multiple designs and fuel choices. Nuclear reactors are likely to be central
in any successful attempt to combat climate change. Thorium reactors, LFTRs in
particular, are a particularly promising approach.
