THORIUM and the Third Fuel
by Joseph M. Dukert

The Understanding the Atom Series

Nuclear energy is playing a vital role in the life of every man, woman, and child in the United States today. In the years ahead it will affect increasingly all the peoples of the earth. It is essential that all Americans gain an understanding of this vital force if they are to discharge thoughtfully their responsibilities as citizens and if they are to realize fully the myriad benefits that nuclear energy offers them.

The United States Atomic Energy Commission provides this booklet to help you achieve such understanding.

Edward J. Brunenkant,
Director Division of Technical Information
UNITED STATES ATOMIC ENERGY COMMISSION
Dr. Glenn T. Seaborg,
Chairman James T. Ramey
Wilfrid E. Johnson
Dr. Theos J. Thompson
Dr. Clarence E. Larson

Jubilation and ceremony marked the startup in October 1968 of the Oak Ridge National Laboratory Molten-Salt Reactor, which uses uranium-233. This was the first time this fuel had been used in any reactor. ORNL Director Dr. Alvin Weinberg points out a detail of the control panel to Dr. Glenn T. Seaborg, co-discoverer of uranium-233.

“…someday the world will have commercial power reactors of both the uranium-plutonium and the thorium-uranium types…” –Dr. Glenn T. Seaborg, Chairman U. S. Atomic Energy Commission, March 25, 1969

INTRODUCTION

More than half the electricity that will be generated by power plants now being built in the United States will come from nuclear reactors. Other booklets in this series explain how heat, released by fissioning (or splitting) of uranium 235 atoms inside reactors, can turn water into steam and drive turbine generators.

The demand for energy is increasing so rapidly, however, that we can’t use uranium-235 indefinitely as our only fuel in nuclear power plants. Uranium-235 (235U) is about 50 million times as efficient as coal, but it is eventually consumed in a reactor and is relatively scarce to begin with.

Only one uranium atom out of every 140 found in nature is 235U. It isn’t easy to separate these readily fissionable atoms from other atoms or to enrich natural fuel by increasing the percentage of 235U that it contains.

Yet the world’s growing and increasingly industrialised population will probably need about three times as much energy during the next few decades as it has used in the millions of years since man discovered fire. Where will we obtain nuclear fuel to supplement our dwindling supplies of coal, gas, and oil? Is there a practical substitute for 235U that could help stretch our atomic resources?

Luckily, there are at least two such supplements—both of them man-made. One is plutonium, which is produced in a two-step process when a subatomic particle, called a neutron, is absorbed by the most common variety of uranium—238U. The second is uranium-233, an isotope that appears as a result of radioactive decay after neutrons have been absorbed in still another material—thorium-232.

Design of the fuel elements for the High-Temperature Gas-Cooled Reactor at Peach Bottom, Pennsylvania, is shown clearly during non-nuclear test. The plant converts fertile thorium into uranium-233 (the third fuel) while generating electricity.

Uranium-238 and thorium-232 are both called fertile materials because they provide a base from which nuclear engineers can “grow” new nuclear fuel. Theoretically, the neutrons released by fissioning atoms can be soaked up by fertile material so that fresh fuel is produced faster than the original fuel is consumed. Such a reactor is called a breeder reactor.

So far, no commercial power reactor has used fuel elements manufactured from 233U. They may begin to appear before long, however, and they can he expected to play an important role in conserving our natural resources of fissionable material.

The AEC does not expect a thorium mining boom to follow the pattern of uranium prospecting that occurred in the 1940s and 1950s. Nevertheless, power plants are already operating in various parts of the world to convert thorium into 233U; and in some ways this “third fuel” is the best of all.

The Element of Thunder and Gas Lights

Baron Jons Jakob Berzelius could not have chosen a more appropriate name for the element he isolated (from minerals on a little island off the coast of Norway) in 1828. Thor was an enormously powerful Scandinavian god, who often gave humans a helping hand. Although he was never quite as important as the father god, Odin, Thor controlled thunder and lightning. He was a source of power to he reckoned with and so is thorium.

Thorium is neither rare nor commonplace. It is the 35th most abundant chemical element in the earth’s crust, ranking just behind lead and molybdenum. Thorium is several times more abundant than uranium, but statistics are inexact because even today there is no thorium mining industry. Most thorium is produced as a by-product from the processing of the rare earths or uranium ore.

The first practical use of thorium developed more than 75 years ago, when an Austrian baron named Carl Auer von Welsbach noticed that burning gas gives off a bright, steady light when surrounded by a fabric cylinder covered with thorium oxide. The brilliant glow of incandescent gas mantles was thorium’s major contribution to the energy industry until the discovery of nuclear power. As late as 1952 more than 65% of this country’s thorium production went into this original application. By 1964, this figure was down to about 30%.

The coming of electricity didn’t completely eliminate the value of thorium in illumination, because it is also useful in light bulb filaments. The addition of about 1% thorium to tungsten makes the material easier to draw into filaments, and it also keeps the hair-like wires from becoming brittle and cracking. For similar reasons, thorium improves the efficiency of tungsten electrodes in high-temperature welding.

Inside an electron tube, thorium electrodes help keep the inert gas pure by reacting quickly with any oxygen or nitrogen that may he present. Thorium oxide is very stable, and its melting point of almost 6000° Fahrenheit is so high that the material has even been used in making crucibles for specialized melting operations.

The principal use of thorium today, however, is in metal alloys. This accounts for about 150 of the 250 tons of thorium oxide produced annually in the US. Some of the magnesium-thorium alloys can be used at temperatures up to 650°F (more than 360°C), and these are being used increasingly in machinery “hot spots”, such as compressor housings and in the casings of large rockets.

Some physical and chemical properties that make thorium valuable for non-nuclear purposes also affect its possible use in atomic reactors. As a metal it can be extruded, rolled, forged, or swaged. It forms ceramic compounds readily, and these stand up well under nuclear radiation and high temperature.

PHYSICS AND ECONOMICS

Thorium’s potential role in nuclear energy can’t be summarized adequately in just a few words. To understand thorium’s value, we must know a bit about nuclear physics. And to realize the reason for its relatively slow development, we have to consider production problems and economics.

This may sound complicated, but it need not be if we examine the parts of the thorium story one at a time. Furthermore, there is an added advantage in understanding the place of the third fuel in the nuclear industry of the future: It will aid our understanding of new developments related to the two atomic fuels we already use (uranium-235 and plutonium-239).

Inside the Atom

Thorium itself isn’t a nuclear fuel. The nucleus of a thorium atom generally doesn’t break up or fission when it is struck by a tiny neutron, which can set off an energy releasing chain reaction in 235U or plutonium. Instead, the neutron is absorbed; and the most common thorium nucleus, which consists of 90 protons and 142 neutrons, suddenly includes an extra unit, making it much less stable than it was before.

Thorium-233 is so unstable, in fact, that even when produced in quantity it soon decays radioactively. Within 23 minutes, half the 233Th atoms in any given mass will decay by emitting one negative charge each. The loss of each negative charge (called a beta particle) in this fashion transforms one of the neutrons inside the affected nucleus into an additional proton. But as soon as the number of protons in a nucleus changes, the atom itself is transformed into a completely different element. An atom with 91 protons in its nucleus isn’t thorium at all; it’s a substance called protactinium.

Protactinium-233 is only slightly more stable, having a half-life of 27 days. It goes through a second decay process, losing another electron; and the result this time is an atom containing 92 protons and only 141 neutrons. This is 233U, which can fission when struck by a neutron transforming part of its mass directly into energy. Uranium-233 can support a chain reaction in a nuclear reactor. It is the third fuel of the Atomic Age.

Slow Balls versus Fast Balls

A fissionable atom is one that may split and release nuclear energy when hit by a neutron. But nature doesn’t guarantee that this will always be the result of the impact—any more than you can be sure that a baseball star will connect solidly with every pitch when he is at bat.

First of all (as in baseball), speed is a critical factor in nuclear reactions. Just as some batters prefer slow halls and others like fast balls, neutrons of various energies are more or less likely to react with different types of fissionable atoms.

Scientists have applied different names to neutrons at various energy levels. If one of these tiny particles is a relative slowpoke, moving along only slightly faster than the speediest airplane ever built, it is called a thermal neutron. On the other hand, if it has enough energy to buzz along at several hundred times the speed of our fastest moon rockets, scientists call it a fast neutron. Neutrons whose velocities lie somewhere in between these two extremes are called intermediate or epithermal neutrons.

Slow or thermal neutrons are much more favored by 235U nuclei than they arc by those of 238U; and this simple fact has dictated the design of almost every nuclear power reactor built so far. Most use a fuel mixture of relatively few 235U atoms and many 238U atoms, even though the latter arc more likely to absorb a neutron than to fission on impact. These reactors can sustain a chain reaction because they use a moderator (often ordinary water) to slow down the flying neutrons without absorbing them. The thermal neutrons that result tend to get past the more plentiful 238U atoms and concentrate on the 235U nuclei, which are more likely to fission than 238U.

The probability that a free neutron of a given speed will intercept one sort of atomic nucleus as opposed to another has been calculated more carefully than any major leaguer’s batting average. The likelihood of an atom’s being hit by a neutron is called its neutron cross section. It is measured in units called barns.

In the water-moderated reactors used by most US nuclear power plants, most of the neutrons that come into contact with 238U are absorbed—changing the 238U by a two-step decay process (like the one described for 232Th) into a new fissionable clement, plutonium. The amount of 239Pu produced is small compared with the 235U present, so it doesn’t have as many fission-producing collisions with the “slow balls” as does 235U. Thus, plutonium contributes little to the energy production of that particular reactor. Eventually, of course (after the reactor core has been removed from operation), plutonium can be separated chemically from the other materials in the fuel elements. Then it can be reprocessed and used as fresh fuel in another reactor later on. But reprocessing and refabrication into new fuel elements are usually long and expensive procedures, and their costs must be factored into business decisions by utilities.

What does all this have to do with thorium and the third fuel? Simply this: 233U, which is produced when thorium is used instead of, or in addition to, 238U as a fertile material in nuclear reactors, has a relatively high cross section for thermal neutrons. Furthermore, when 233U “accepts” a slow neutron, it is even more likely than 235U to fission and release energy immediately, thus adding fuel to the nuclear fire.

Or, to pursue the baseball analogy, you might say that adding thorium to the lineup helps keep our team at bat for a longer time during each inning.

The Extra Base Slugger

A nuclear chain reaction is possible because a fissioning atom releases one or more neutrons from its nucleus at the same time that it gives off energy. These neutrons are free to cause other nuclear fissions, releasing still more “atomic fast balls”. Some of the fresh neutrons leak away from the fuel completely; others are absorbed by the reactor structure or by fertile material. Still others, as we have seen, may come into contact with a fissionable atom but for some reason don’t produce fission. A chain reaction will actually begin and continue only so long as at least one neutron released in each fission succeeds on the average in producing another fission.

Usually a fissioning nucleus releases two or three new neutrons. And, obviously, it’s easier to keep a chain reaction going in a material where the individual reactions are more productive to begin with. Here’s where 233U scores again: In the shorthand of nuclear engineering, the neutron productivity of a fuel material is designated by the Greek letter eta, ?. This value equals the average number of fissions produced for each neutron absorbed by the fuel and multiplied by the average number of neutrons released in each fission.

Obviously, ? must be greater than 1.0 for a chain reaction to take place. If it isn’t, the number of available neutrons will drop every time one is captured. And the higher ?? is, the more energy you are likely to get from a nuclear fuel.
In a thermal reactor just starting up, ? is 2.02 for 239Pu — more than ample to produce a chain reaction.
• For 235U, it’s even a little higher-2.09.
• But ? for 233U under the same conditions is 2.23.
This is another reason why efficiency-minded reactor engineers are attracted to this third fuel.

A demonstration fuel assembly containing uranium-233 is checked for radiation.

Although its advantage over the other man-made nuclear fuel (plutonium) shows up best in thermal reactors, 233U can also be used in intermediate (epithermal) systems.

There Are Some Problems

A few power reactors use thorium as a fertile material — either interspersed with the fuel in the core itself, or in a “blanket” around the outer edges of the core. Fuel elements made of 233U have been tested to experimental reactors. Yet, for all its potential advantages, thorium is still regarded as a minor factor in the nuclear industry to date. Why?

The main reason is that the thorium –233U cycle simply doesn’t provide new fuel as rapidly as the 235U-plutonium cycle. But there are some other factors, too.

First of all. thorium doesn’t exactly compete with uranium; it can never replace it completely. Thorium may supplement 238U as a fertile material, but you can’t fuel a reactor with thorium alone. It must he used in conjunction with one of the fissionable materials -235U, plutonium, or 233U. But the process of enriching natural uranium (i.e., increasing the relative percentage of 235U it contains) produces a substance that is already an intimate mixture of fissionable and fertile material. The tendency is to use what is already on hand, even though something else might offer theoretically greater benefits.
Technological costs are also somewhat of a chicken-and-egg matter. There has been far less experience with thorium and 233U than with 235U and 238U — or even with plutonium – as reactor fuel. Until more experience is acquired, costs will stay high.

Finally, there is a real technical bugaboo in the thorium cycle that we haven’t mentioned so far. It is the side production of 232U.

Real life is seldom as simple as physical formulae. The usual course of events that occurs when a neutron is absorbed by 232Th was shown in an earlier series of diagrams. But occasionally the unstable 233Th starts its decay process by giving off a couple of neutrons instead of a beta particle. A nuclear engineer would indicate the reaction like this:

232Th —n.2n— 231Th —?— 231Pa

The symbols above the arrows mean that first one neutron (n) was absorbed by the thorium and then two (2n) were emitted. Without any further bombardment, the thorium next lost a beta particle (?) and turned into an isotope of protactinium with an atomic weight of 231. This represents the beginning of several possible nuclear reactions (each a combination of particle absorptions and emissions), which may
produce 232U instead of 233U.

A demonstration fuel assembly containing uranium-233 is checked for radiation.

Uranium-232 is “bad medicine”. It causes no particular trouble immediately, but its radioactive decay eventually produces daughter* isotopes, which give off dangerous penetrating radiation called gamma rays. The presence of only a few hundred parts of 232U per million parts of U is enough to require thick, heavy shielding between the uranium and the workers who form it into reactor fuel elements.

Nuclear fuel elements normally have to be manufactured to precise specifications. Fissionable material of any kind is far more valuable than gold, and fabricating it into elements with the close tolerances needed is an exacting (and expensive) job at best. But when the job has to be done by remote control, behind lead and concrete shields, it’s easy to understand why developments and experience with the fuel material in question have come slowly.

Actually a great deal of progress has been made, and some of this will he discussed later on. Furthermore, some of the fabrication problems may he reduced by using the newer reactor types described in the final section of this booklet. Despite all potential difficulties, research and development is moving ahead slowly but steadily because of our growing energy needs, our limited resources, and the conflicting factors of economics.

Where Are Those Golden Eggs?

Compared to our supply of coal, oil, and natural gas, the earth’s resources of fissionable material seem enormous. And they are!† Yet not all uranium in the earth’s crust is easy to obtain. As we gradually remove the richest and most readily available ores, the raw material cost of reactor fuel is bound to rise. Still, nothing is likely to stop the incredibly rapid rise in the public’s demand for electricity. The use of air conditioning in the summer, electric heat in winter, and other electric conveniences all year round has created a situation in the United States in which the total use of electricity now doubles every 10 years.

In some fast-growing areas the doubling interval is every 6 years. Such a voracious appetite for electricity would really threaten to exhaust both our fossil fuel and our supply of reasonably priced uranium if it weren’t for the promise of breeder reactors, which can produce new fuel at a faster rate than the original fuel is used up.

*A daughter is a nuclide formed by the radioactive decay of another nuclide, which in this context is called the parent.
†Sec Sources of Nuclear Fuel, a companion booklet in this series.

A breeder reactor produces more fuel than it consumes, but early versions are likely to take 20 years to double their content of fissionable material. Demand for electricity, which now doubles every 10 years, and the even more rapid multiplication of power plants using nuclear instead of fossil fuel to generate electricity. may require the power industry to use even more expensive uranium ores or to introduce a “third fuel” via the thorium cycle.

If we kept using only 235U even if it were practical to pay as much as $100 a pound for unenriched uranium oxide (that’s about 16 times what it costs now) — we would reach the bottom of our natural fuel reservoir early in the next century. Luckily, however, breeders can postpone that day almost indefinitely by transforming 238U into fissionable 239Pu.

As wonderful as this is, there are certain dangers in becoming too complacent about the plutonium breeders that are expected to come into their own in the 1980s. For instance, we must not forget that the breeders have a doubling time of their own. This is the minimum time a reactor takes to produce enough new fissionable material (allowing for the amount that “burns in place”) to fuel another reactor of the same size and power output. For the early plutonium breeders this is likely to be about 20 years, although by the end of this century it might he reduced to as little as 6 years. If the demand for electricity keeps rising, however, we probably won’t be able to stop mining uranium even then. Breeders just don’t breed fast enough.

A second point to remember is that breeders themselves will have to be refueled periodically – even though the core that’s removed will contain more fissionable material than it started with. In time the fuel elements wear out physically, subjected as they are to high temperatures, radiation, and corrosive materials. Besides, nuclear operation creates fission products, which eventually begin to absorb so many neutrons that they, may interfere seriously with the chain reaction.

Poisoning by fission products is balanced partly by the increasing amount of fissionable material in at breeder reactor. But this, in turn, presents other problems. The control system must be adjusted constantly to take care of the additional fissionable material, and – even more serious front an economic standpoint – the valuable new fuel is almost invariably in a physical arrangement that offers less than maximum efficiency. The new fuel, which costs money to produce, may be contributing so little to the reactor’s output that it is not even “earning its keep” at this point.

It takes a complex computation in specialized engineering economics to decide the extent to which newly bred fuel should be used in place instead of being removed for reprocessing. A single fuel load for a reactor may represent more than 15% of the entire capital cost of a nuclear power plant, so it would seem wise to make each core last as long as possible. This is especially true in a fast-breeder reactor, where the initial fuel cost is higher than for a thermal system of the same power output. Nevertheless, there are other factors to consider. After several years of operation a reactor may still have in its core a good many of the fissionable atoms it started with (plus newly bred fuel), but it play be generating energy less efficiently. Considering the amount
of money invested in its original core and the potential profit from reselling the fresh fuel now present, it may make economic sense to replace the core more frequently than is absolutely necessary.

At any rate it should be clear that a fair amount of the fuel produced by breeder reactors of the future will always be tied up at something less than peak productivity. Some of it will be in the cure where it originated, contributing only modestly to that reactor’s efficient operation. Another sizable fraction will be “in the pipeline” from its point of origin to another reactor—ill temporary storage, ill transit, being reprocessed, or undergoing fabrication into new fuel elements.
For all these reasons there will he an increasing need to mine uranium.

For the Future: A Mixed System

Even after breeder technology is well established, converter reactors will continue to operate. The power plants being built today represent, vast investments themselves. Undoubtedly they will he operable for many years, and as their initial capital costs are amortized they will certainly form a solid and economical basis for most utility networks. If the demand for electricity continues to skyrocket it will be impractical to retire large plants that can still function effectively, even though they may be technically obsolete compared with the advanced breeders.

There is still another reason for predicting a mixed system of nuclear power plants to the future. Breeder reactors cost more than converters to build and to fuel, but the plutonium they produce can be sold for a good price. The price of any commodity in a free market,* however, depends on supply and demand; and if our country switched to breeders almost completely, there might come a time when the resale price of plutonium would drop enough to make nuclear converters economically attractive.

*Initially the federal government offered to purchase all plutonium produced in reactors from uranium leased or sold by the AEC at a guaranteed price of a little over $9 per gram. However as private enterprise began to accept a larger role in fuel marketing, federal authorities allowed the buy- back agreement to expire January I, 1971.
†Atomic Energy in Use, Division of Public Information, U.S. Atomic Energy Commission, contains a brief discussion of fusion reactor, that could draw their fact from the earth’s oceans.

Of course there is also the possibility that technological advances will change the entire outlook.† But it seems safe to predict that we will continue to use uranium in fairly large quantities for the foreseeable future and that the cheaper uranium ores will eventually disappear. So the next few decades – and certainly the 21st century – may see a strong tendency to supplement our uranium supply through the use of thorium. And this raises the question: Where will it come from?

The Lemhi Pass in Montana and Idaho contains large reserves of thorium.

FROM MINERAL TO METAL

Thorium is found in more than 100 minerals, but there are only a few in which it is the chief constituent. One is thorite, the glassy substance in which Berzelius discovered thorium. The other is thorianite. Neither of these minerals has ever been very important in the commercial production of the element, however. Instead, most of the world’s supply comes from a sandy material called monazite – of which only a small percentage is thorium.
Thorium is also found in the uranium ores commonly mined in Canada, although the market for thorium hasn’t justified processing these ores for it. It’s estimated that over 35,000 tons of thorium oxide remain in the tailings dumped outside Canadian mines, but it’s rather doubtful that it will ever be used. This ore contains certain thorium isotopes that would produce an inordinately high amount of 232U when irradiated in a reactor.
Monazite is essentially a phosphate of various rare earths. It occurs in small quantities in granite and some other common rocks. Like most thorium-bearing minerals. Monazite is insoluble in water, and as the rocks containing it erode, its grains tend to concentrate along nearby river beds or beaches. It has been found, for example, in the midst of old gold placer deposits in the Black Bills of South Dakota, as well as in South Africa. Large masses of monazite have also been observed in Madagascar, Quebec, the Carolinas. Idaho, and Southern California.

The most important commercial deposits of monazite right now are probably the beach sands of southwestern India, Ceylon. Brazil, and eastern Australia. Because there has been only a small market for thorium until now, nobody is really sure what further resources might he brought to light by a more intensive search. Large thorium reserves have been found in the Lemhi Pass area of northern Idaho and Montana in recent years. This area alone has a potential of several hundred thousand tons of Th02 recoverable at $5 to $10 per pound. However in view of the small market, little work is being done to locate reserves for mining.

Monazite is mined primarily for the rare earth elements it contains. After the sands have been washed, sifted, and sometimes concentrated by magnetic processing, the thorium is removed as a by-product. The usual method is to treat the ore mixture with either acid or caustic soda, then to separate the various components by precipitating or dissolving them selectively.

A dredge gathers sand containing thorium.

If the thorium will be used in a nuclear application, of course, additional purification is required. Some rare earths are such good ”neutron grabbers” that they have been used in reactor control rods. A few parts per million of materials like that in the thorium would noticeably cut down its ability to breed 233U; so they must be eliminated by further solvent extraction, precipitation, or other refining techniques.

The final step in producing thorium metal may he taken in several ways, but the most widely used method involves a large vacuum container called a “bomb”. This is a sealed steel pot, lined with high-purity limestone and heated to 600`( about 1150°F). The pot is loaded with a purified thorium compound (normally the fluoride or oxide), powdered calcium, and zinc chloride. The furnace’s heat causes the chemical reduction of the two compounds by the calcium and this generates more heat.

The free thorium and zinc combine in an alloy, which settles to the bottom of the container. Later the alloy is heated to 1150C(more than 2000F) in another vacuum chamber, and the zinc is allowed to boil off, leaving pure thorium metal in a spongy mass shaped like (and called) a “derby”. Electrolysis, arc melting, or vaporization – and – condensation then produce the silvery while ingots, which arc the raw material for at least some of the thorium forms to be used in advanced nuclear reactors.
Some of the physical constants for thorium are listed in the table along with the corresponding data for thorium dioxide and thorium carbide two compounds favored for many reactor designs.

Like uranium and plutonium, thorium is a pyrophoric metal, which means that it can ignite spontaneously. This normally takes place, however, only when the metal is finely divided, say, in powder form or in small particles machined from a larger piece. The level of radioactivity from ordinary thorium is low enough so that it causes no great concern in industry, although the potential danger from materials associated with it (especially thoron and radon gases) requires some caution in handling. All in all, thorium is much easier to work with than plutonium and certainly no more difficult than natural uranium. It is also simpler to handle than enriched uranium.

Research in thorium alloys uses a consumable arc-melting furnace at AEC installation in Richland, Washington.

Nevertheless, it’s interesting to look briefly at the methods used to fabricate thorium fuel elements, and to consider the more substantial technical problems involved in re-fabricating the output of a thorium converter or breeder reactor into elements of the third fuel itself–uranium-233.

FABRICATING FUEL ELEMENTS

Even though thorium and its compounds are not especially difficult to handle, there is an obvious challenge in producing fuel elements that must remain intact inside a “radioactive furnace”. It may be necessary to control the diameter of a ceramic pellet to within a few ten-thousandths of an inch to make sure that it will conduct heat properly to an adjoining metal surface. Fuel compacts must have a very high average density and be almost perfectly uniform throughout in order to prevent undesirable “hot spots”, which arise either from fuel concentrations or from tiny voids.
To cite just one more difficulty, some of the products of nuclear fission are gases, therefore a fuel element must normally be designed and manufactured to retain these gases as they form, without swelling or warping.

Thorium, remember, is not the fuel. To make sure that this fertile material gets its fair share of the neutrons released by nuclear fission inside the reactor, it is ordinarily interspersed with the uranium or plutonium fuel as closely as possible. In the case of metallic fuel, an alloy is used. If a ceramic compound is called for, the fissionable and fertile materials may be blended before being put into the final fuel form. In either case, the thorium is an integral part of the fuel element.

So long as we are dealing with thorium that is going into a reactor
for the first time, the fabrication methods are straightforward. On the other hand, if we are recycling thorium, which has been recovered from a spent reactor core, the situation is quite different. The problems and procedures in that case are related to those of fabricating new fuel elements from 233U and they’ll he treated separately, starting on page 26

Preparing for the Harvest

Thorium can be used in reactors in at least four forms:
1. Metal slugs
2. Molten salt
3. Oxide compounds (in pellets, rods, or “pebbles”)
4. Carbide compounds (in a similar variety of shapes)

In the first type of application, a uranium-thorium, plutonium-thorium, or uranium-plutonium-thorium alloy may be cast or extruded. Its cladding may be aluminum, stainless steel, or some zirconium alloy. The melting point of thorium and its alloys has limited the use of this sort of fuel element to experimental and test reactors in which the operating temperature is kept fairly low.

Thorium is “canned” in aluminum to provide neutron targets production of uranium-233.

The second application doesn’t involve fuel elements in the conventional sense, and so it will be discussed only in the section of this booklet dealing with various reactor types. The third and fourth forms- oxide and carbide compounds -are interesting and complex enough to warrant describing their preparation in some detail. Both thorium oxide and thorium carbide are ceramics, or pottery. Their extraordinarily high melting points make them ideal for use in nuclear power plants, where high fuel element temperatures are generally a key to greater efficiency in generating electricity.

Oxide fuel materials have certain other advantages. Unlike carbides, they do not react chemically with high-temperature water. They can withstand irradiation quite well, and their dimensions are not greatly affected by a reactor environment. At the same time, however, oxides don’t conduct heat very well, and this limits the size and controls the shape of practical fuel elements.

The first commercial power plant to use thorium in its nuclear core was Unit No. 1 at the Indian Point Nuclear Power Station in New York. It used stainless-steel tubes filled with pellets of urania-thoria (a mixture of uranium dioxide and thorium dioxide). The thoria was prepared by dissolving a common commercial salt (thorium nitrate tetrahydrate) in water and then precipitating the oxide by adding oxalic acid. The resulting powder was dried and hardened at a very high temperature, then mixed with urania in the desired proportions before being sintered, or heat-pressed, into pellet form. Millions of these pellets -each about the size of an ordinary cold capsule- were loaded by hand into the stainless-steel tubes, which were eventually bundled together into the fuel elements for Indian Point.

The first fuel load at the Indian Point Station commercial power reactor contained a mixture of uranium dioxide and thorium dioxide.

Since then, a process has been developed in which urania-thoria powder is packed into tubes similar to those described above. These tubes are vibrated as they are filled to produce a very dense compact. For best results in vibratory packing, however, it’s essential that the individual grains of powder be very dense themselves and that they be almost uniform in size. Several ways of producing such feed material have been tried, but the most popular and most successful so far is the sol-gel process developed by the U. S. Atomic Energy Commission’s Oak Ridge National Laboratory in Tennessee.

Fuel bundles for the Elk River boiling-water reactor also used urania-thoria pellets sealed inside stainless steel tubes.

The sol-gel method starts with the same commercial compound, thorium nitrate tetrahydrate. But instead of adding it to water and precipitating, the material is put into a rotating drum and treated with superheated steam (350°-450°C) to drive off the trapped water and nitric acid. The very fine thoria powder, which is left, retains a small amount of nitrate ions. When this material is mixed with uranyl nitrate in a dilute acid solution, the two solids unite to form a gelatine- like substance, which does not settle to the bottom but remains suspended in the liquid.

The whole business is pumped into flat trays about 3/4 inch deep and heated to 80°-95°C for a number of hours. When all the liquid has evaporated, the solid solution of urania and thoria left in the bottom of the tray resembles a dried up mud flat. After heat treatment at 1150°C in an inert atmosphere, the urania-thoria crumbles into coarse black particles, which look like tiny chunks of coal when magnified. (See figures on page 25.) Their density is more than 98% of the theoretical maximum. Ball-milling and screening will make them uniform in size.

The sol-gel process for carbide production is quite similar, except that carbon is added to the suspension before evaporating. The dried gel is heat-treated at a somewhat higher temperature (up to 2100°C) in a carbon monoxide atmosphere. In this case the end product is thorium-uranium carbide, with a particle density of from 95 to 99% of the theoretical maximum.

Urania-thoria in evaporation trap (above) during the sol-gel process. On the left is a magnified view of the particles after calcining.

Although the carbides have a lower melting point than thorium dioxide, they are better for high-temperature reactors because carbide fuel won’t react with a graphite moderator. It can be embedded directly into graphite blocks instead of being sealed inside metal tubes.

Something Old, Something New

One enormous improvement in controlling the size and shape of either oxide or carbide particles uses a simple principle that has been employed by man for centuries and by nature for countless millenia: Rain falls in droplets instead of in a continuous sheet because surface tension in a falling liquid tends to pull it into individual spheres. Musket balls used to he manufactured accordingly by dropping molten lead through a screen from the top of a tall tower. In one variation of the sol-gel process, the sol is sprayed through small nozzles into a tapered column through which an organic liquid is being pumped upward. The surface tension of the slowly settling gel particles pulls each one into a”microsphere”. The result is that the uranium-thorium carbide (or
urania-thoria) ends up looking like a bunch of black marbles under the microscope instead of tiny coal chunks (see figure).
Because of their shape, microspheres can be given uniform
individual coatings rather easily. In effect, each particle becomes a complete microscopic fuel element in itself, with its own protective cladding to keep fuel and fission products sealed inside.

The great beauty of this extra trick – like the rest of the basic sol-gel process – is that it can be done remotely. The elimination of manual steps is especially important in any process involving thorium, because the eventual use of recycled fuel will mean dealing with much higher radiation levels.

There are three approaches to the fabrication of 233U into new fuel
elements after it has been bred in the thorium cycle. One combines an extra step of processing with speedy operation to fabricate the fuel elements with a minimum of shielding. At the other extreme, it is far more costly (but sometimes necessary) to carry out the entire operation in very heavily shielded “hot cells”. And, as you might expect, there is still another approach, which seeks a compromise between the two. All three techniques are illustrated on page 27.

Microspheres of thorium-uranium carbide.

Three Approaches to Uranium-233 Fuel Element Fabrication
(A) simple glove boxes.

(B) Completely remote handling.

(C) A combination of A and B.

Whichever method is to be used in refabrication, the reprocessing of the spent fuel itself must be carried out remotely, even after the spent fuel has been in storage for 3 to 6 months, to allow short-lived fission products to decay. Watching through periscopes from behind 6-foot-thick walls, operators of remote-controlled cranes unload fuel elements under 20 feet or more of water. When dealing with metal-encased fuel, the end fittings are cut away and the fuel-bearing sections are chopped into short segments. These are then dumped into concentrated nitric acid that dissolves the fuel and leaves the empty metal tubes intact. Solvent extraction, heating, reduction, precipitation, and further heat treatment produces a mixture of thorium and uranium oxides, which can be formed into pellets for additional processing.
Graphite fuel, on the other hand, may be treated in one of two ways: (1) it may be crushed and ground into a fine powder before the fuel is separated from the graphite by dissolving it in acid, or (2) the fuel compacts may he broken up and burned; this process drives off the carbon as carbon dioxide gas.
At any rate, the fuel residue at this point consists of different radioisotopes. In addition to the fission products there may be: (1) 235U, which didn’t get a chance to fission while in the reactor: (2) 238U and 232Th, which are also “left over”: (3) 233U, which has been formed from the original thorium: and (4) the “bad actor” – 232U. Besides the normal assortment of thorium isotopes, there is also an unusually large amount of thorium-228 – the first daughter of 232U in the bothersome decay chain that produces dangerous gamma emitters.
The thorium can be separated from the rest of the mixture by fairly simple chemical means, but the presence of the 228Th makes it too hot to handle economically for some time.* Thorium-228 has a radioactive half-life of almost 2 years, but most engineers believe that the recycled thorium could be used safely after 10 or 12 years of underground storage (5 or 6 half-lives) if that seemed desirable. By the end of a decade, the 228Th content would be about one-half of one-half of one-half of one-half of one-half of what it was at the time it came from the reactor. The most persistent of its dangerous daughters have half-lives of only a few days, and therefore offer no great problem once their birth rate has been cut down sufficiently.

*It is possible, of course, to separate one isotope of an element from another. However, while this might be practical in some cases, it certainly is not for a relatively cheap material like thorium.

Haste May Avoid Waste

Uranium-232 itself has a half-life of almost 74 years, which means that 228Th builds up again quite slowly after it has been removed from a mixture of the two. This fact is what makes lightly shielded fabrication facilities for 233U fuel elements possible.
The first pilot plant for such direct fabrication (at Lynchburg, Virginia) proved its feasibility in the mid-1960s. The sol-gel process was carried out and the fuel pellets were loaded into tubes inside glove boxes. This meant that the technicians watched the operation through windows; and instead of using automatic manipulators they handled the materials directly, wearing thick gloves built into the outer walls of the various chambers. It took only 2 or 3 days to complete rods from the raw feed material, and these were assembled into the finished fuel bundle by a machine operating at the bottom of a deep pool.

Such a pool also offers a convenient storage spot for fuel assemblies, because the high gamma radiation gradually appears as more 232U decays. Shipping arrangements for 233U fuel are no more difficult than those for transferring spent cores to a reprocessing plant in the first place.
Uranium-233 is normally formed much more readily in a power reactor than the unwanted impurity 232U. Uranium-232 represents only a few hundred of each one million atoms. But, as we have pointed out, this may be too many for comfort. To simplify experiments with 233U, some batches have been produced with only a few parts of 232U per million, but this convenience involves a penalty. The method involves segregating the fertile thorium from the uranium fuel in a reactor and limiting the time the thorium stays there. The reasoning behind the technique is that 232U tends to be produced mostly by very energetic neutrons- the kind produced by the fissioning of nearby atoms before they can he moderated. Obviously if the thorium (and the 233U it produces) are kept far from the center of nuclear activity, not much 232U will he formed. On the other hand, this means that the 233U won’t do much to extend the operating life of that particular core either.
One thing is certain. The problem of gamma radiation in 233U will increase each time fuel is recycled through a reactor; and even those who developed the light-shielding technique admit that it will not always be able to do the complete job. The first pilot-scale demonstration of semi-remote fabrication took place at Oak Ridge, also in the mid-1960s. In this case the work was conducted in a converted hot cell, which actually became a sort of multistory glove box. Viewing was permitted only through high-density glass portholes, and the walls were made of thick layers of steel and concrete. Lead-lined gloves were used for some operations, but remote manipulators were employed in others, and the whole process was automated to a greater degree. In spite of these stricter precautions, however, this technique still requires the extra step of separating all thorium, front the feed material before the main job begins. And there is still a time limit (in this case a few weeks) between the time a hatch of fuel material is purified and the time the fabrication of the fuel elements must be complete.

Look! No Hands!

Completely remote fabrication facilities – in which everything is done automatically or indirectly – have been built now both in this country and abroad. They are clearly more expensive to construct and maintain, but such plants have advantages, too. They eliminate some of the preparatory steps, they can he operated at whatever work pace happens to be convenient, and they don’t need periodic intensive cleanings. Finally, they make it possible to handle the whole fuel cycle in one location. A reactor core can be dismantled, the fissionable materials removed, and a new core produced in one uninterrupted process. Some nuclear engineers believe that very large power plants will eventually carry out all these steps right at the plant site. This would reduce transportation costs to a minimum and bring the original Sunday supplement dream of a nuclear power plant, which “feeds itself” and operates continuously, closer to reality.

As in every other aspect of thorium utilization, the final determining factor will almost surely be economics. Other means of turning the third fuel into a producer of “golden eggs” have also been proposed. Most of them involve advanced reactor designs, and so it seems most appropriate to conclude this booklet about thorium with a brief discussion of how various reactors that use thorium work, and how others may work in the future.

THORIUM REACTORS—NOW AND TOMORROW

In the Beginning

It was in 1961 that 233U was first used to produce a chain reaction in a commercial installation. The little-publicized event took place near Los Angeles in a critical assembly* devised to study the concept of an Advanced Epithermal Thorium Reactor, which was intended ultimately to use liquid sodium as a coolant. More than 7 years passed before any reactor used 233U exclusively as its fuel, and this was another rather exotic type – the Molten Salt Reactor Experiment at Oak Ridge, Tennessee.

But perhaps the most significant “first” in the development of the thorium fuel cycle took place between these two dates in an ordinary light-water reactor – the kind that has formed the backbone of the early nuclear power program in this country. The initial core installed at the Indian Point reactor in 1962 used pellets of urania-thoria, and a small percentage of the energy released by the core of this pressurized-water reactor during its 442 full-power days of operation came from 233U, which was bred from the thorium and burned in place. The core was removed for reprocessing late in 1965, and – although it had performed well from an engineering standpoint – it was replaced by a more conventional core (which contained no thorium) for economic reasons.

*A critical assembly is fissionable material arranged in just the right amount and geometry to produce a self-sustaining chain reaction for test purposes: no effort is made to produce power.

The first “critical assembly” using Th or in in consisted simply of uranium and thorium wafers arranged in a lattice of aluminum drawers.

Thorium can be used as a fertile material in almost any sort of reactor, and its inclusion has been considered at least in most of them, but actual operating experience is still limited. Thorium oxide was used in the Borax IV boiling-water reactor experiments at the National Reactor Testing Station in Idaho; and in the boiling-water reactor at Elk River, Wisconsin.

Two types of thorium-bearing fuel for gas-cooled reactors. On the left are gear-shaped compresses of uranium thorium carbide for a power plant near Denver, Colorado. Below is a filled graphite sphere for a German reactor.

Gas-Cooled Reactors

Much of the hope for wider use of thorium and 233U now seems to lie with gas-cooled reactors; and it’s in this type that the third fuel has reached its greatest acceptance. It has been used, in fact, in the gas-cooled reactors of three Western Countries: the United States, the United Kingdom, and the Federal Republic of Germany.

The oldest of these is the small but highly successful Dragon Reactor Experiment, which first reached criticality in mid-1964 and achieved its designed power level of 20 thermal megawatts about a year and a half later. No attempt has been made to generate electricity from the Dragon’s “hot breath”, since this reactor was built in England by the multi-nation Organization for Economic Cooperation and Development simply to provide general physics information and the means of testing new fuel elements for advanced design concepts.

The Dragon core is relatively small-only a bit over 5 feet tall and about 3 1/2 feet in diameter. The uranium fuel and thorium fertile material are both in the form of coated particles embedded in graphite, and both oxides and carbides have been tried. The individual fuel compacts are cylindrical in shape with a hole in the center so they can be stacked on central guide rods inside graphite tubes.

Helium gas cools the reactor, emerging from the core at 750°C. High outlet temperatures like this offer a considerable advantage in operating turbines, compared with water- cooled reactors, which typically produce steam at less than 400°C.
The nuclear power plant that pioneered the use of thorium in Germany is considerably larger than Dragon, but it is also something of a test bed. Located at Julich, about 25 miles west of Cologne, the Arbeitsgemeinschaft Versuchsreaktor (AVR) is intended to provide working experience with the “pebble-bed concept”. In this design, microspheres of thorium-uranium carbide are packed inside hollow graphite spheres, which are simply shoveled into a 10-foot-diameter cylinder-along with a good many more spheres of plain graphite-to form the core.

These pebbles are fairly large; each fuel element is just a trifle smaller than a tennis ball. The AVR power plant stacks them to a depth of about 10 feet inside the core cylinder to develop enough heat to generate 15 megawatts of electricity ( 15 Mwe).
The ratio of thorium to uranium in AVR is about 5-to-1, and the form of the fuel elements should make it relatively easy to handle the mixture of residue fuel and newly bred fuel after they have been taken from the reactor. Furthermore, the number of pebbles (and thus the amount of fuel in the core) can be changed quite easily from time to time as a control measure if that’s desirable. Interestingly enough, the main control rods of the AVR are very similar to those in reactors of more conventional design; they can be plunged up and down amid the pebbles almost as if they were moving through liquid.

Britain’s Dragon Reactor.

The helium in AVR reaches an even higher temperature-850°C-than that in Dragon. Its performance will he watched closely all over Western Europe, because its success may lead to the construction of a very large plant of this type under the auspices of Euratom.* The full-scale plant is referred to as THTR-Thorium High-Temperature Reactor. Europeans hope that it can combine low-cost power with a way of stretching their nuclear fuel supply.
The United States has not lagged in the development of high-temperature, gas-cooled reactors using the thorium fuel cycle. Late in the 1950s, more than 50 utilities joined with the U. S. Atomic Energy Commission in research and development that led to the construction of the 40-Mwe Peach Bottom Atomic Power Station in Pennsylvania. Low-power operation began at Peach Bottom in 1960. and full-power operation followed a year later.

*Euratom is the European Atomic Energy Community. This organization promotes nuclear growth in Europe. Its members arce Belgium, France, Federal
Republic of Germany, Italy, Luxembourg, and The Netherlands.

West Germany’s pebble-bed power plant.

Highly enriched uranium carbide-a material in which more than 90% of the uranium atoms are 235U-is used along with thorium carbide throughout the core. Both the fuel and the fertile material are in the form of carbon-coated microspheres embedded in graphite. As in Dragon, the Peach Bottom fuel compacts are generally cylindrical in shape, and they fit around a central spindle and are simply stacked, one on top of the other, inside a graphite sleeve.
The most ambitious thorium converter to date is the Fort St. Vrain Nuclear Generating Station under construction about 33 miles north of Denver, Colorado. It will be almost 10 times as powerful as the one at Peach Bottom, and it should begin operation early in the 1970s. It will be similar to Peach Bottom in design, but the microspheres of uranium and thorium carbide in the fuel elements will be coated with silicon carbide instead of plain carbon. An even larger version of such a plant (about 1000 Mwe) has also been proposed, with a number of modifications that would turn it into a full-fledged 233U breeder.

Other Paths to the Future

As you might expect, some of the original ideas concerning the use of thorium have changed as more practical experience was obtained. An early U. S. program to develop a thorium reactor using heavy water* as its moderator and a waxy organic compound as the coolant, for example, has been dropped. The Large Seed Blanket Reactor (LSBR) concept is being studied as a long-term possibility.
Even in its heyday, nobody ever thought that the LSBR could breed fuel as efficiently as the liquid-metal-cooled fast breeder reactors. But the plain fact is that we are much more familiar with reactors that use ordinary water as a coolant and operate with thermal neutrons. It was natural to assume that they would be cheaper and easier to build.

The basic idea of the LSBR was to ”seed” a core of thorium with chunks of pure 235 U. Because its breeding ratio was expected to be only a little more than 1.0 (i.e., an average of only slightly more than one atom of 233U would be formed per fission), the LSBR would have a very long doubling time. The reactor couldn’t breed enough new fuel to make it a supplier for other reactors, but the 233U it bred and burned in place would make each one of its own cores last a long time – about 9 years. As long as so much of a reactor’s fuel costs is involved in fabricating new cores, this long core life could he a marvelous advantage.
The physics and abstract economics behind the LSBR concept were correct, but the engineering and manufacturing proved to be stumbling blocks. The main trouble was that in the mid-1960s there was little immediate hope of producing economical fuel elements that could survive in a water-cooled reactor for nearly a decade. The fuel compounds tended to swell and crack; and the metal cladding wasn’t strong enough to take the long-term stresses. So the idea of building such a system then was scrapped, and efforts were redirected into a more leisurely development program for what is now called LWBR (Light-Water Breeder Reactor).
Less than l% of natural uranium consists of atoms that will fission in a light-water thermal reactor; and there doesn’t seem to be much we can do to change that discouraging statistic. But a successful seed blanket system would enable such a reactor to draw energy from about 50% of the atoms in thorium (which, after all, is at least as plentiful as natural uranium). That’s why we will undoubtedly hear more about the LWBR, or LSBR, or whatever it is eventually called, in the future as soon as further research and development improve its chances of competing economically.

*Heavy water is water (H20) in which atoms of ordinary hydrogen have been replaced by a heavier isotope of hydrogen called “deuterium”. The nucleus of a deuterium atom consists of a single proton, as usual, plus one neutron. Heavy water, sometimes referred to as D20, is more efficient than ordinary water at moderating the speed of neutrons in a reactor, but it’s also much more expensive. One advantage is that it permits the use of unenriched uranium fuel.

Some Canadian officials believe that thorium can increase the importance of heavy-water reactors like the one being fueled here.

Can CANDU Do It?

Canada has by necessity shown much more interest than the United States in nuclear reactors that use natural uranium as fuel and heavy water as their moderator. Canada is one of the world’s richest sources of natural uranium, but it lacks facilities for 235U enrichment.

No heavy-water reactor has yet tried the thorium fuel cycle, but it would seem to he easily adaptable to the type of reactor called CANDU (Canadian-Deuterium-Uranium). Some nuclear engineers and economists, in fact, maintain that if Canada were able to tap the energy potential of her vast thorium resources it would never he necessary to develop true breeder reactors.

The U.S. and Canada have exchanged technical information regularly about the use of heavy water as a moderator, and if any practical benefit from its use in this country appears likely in the future our interest will undoubtedly perk up. The simplicity of the fuel elements in such a reactor is a definite attraction.

View of Molten-Salt Reactor during core assembly shows graphite channels through which liquid fuel and fertile material flow.

The Molten-Salt Reactor

Perhaps the most exciting of all prospects for using thorium is in a reactor that doesn’t use fuel elements at all. It is called the Molten-Salt Reactor, and the figure illustrates generally how it works.

Because neutrons from fissioning atoms fly off in all directions, they won’t start a chain reaction unless they are moving around in a nuclear fuel mass of the right size and shape so that enough of them will meet with suitable nuclear targets. For example, a chain reaction could never take place in a thin sheet of fissionable material no matter how big it was. Too many neutrons from fissioning atoms would leak away into the atmosphere and he lost. The same is true of a single pebble in the Pebble-Bed Reactor described above, although a large number of the pebbles piled together in a cylinder can easily support a chain reaction.
In a Molten-Salt Reactor, a mixture of lithium. beryllium, zirconium, and uranium fluorides is melted to form a liquid that can he pumped through pipes like water. As long as the molten salts are confined to a narrow pipe, no chain reaction can begin. But when the liquid is pumped through hundreds of parallel channels in a graphite framework, it reproduces the mass and geometry of a conventional solid reactor core. A self-sustaining chain reaction can take place.

Thorium fuel cycle.

As early as 1950 Oak Ridge National Laboratory began development of a liquid-fuel reactor, which would have used uranium salts dissolved in water. Later this led to the actual construction of a molten-salt reactor in which uranium fluoride composed exclusively of 233U has now been used. Based on this experience, a large reactor has been designed to use both uranium and thorium; it’s called the Molten-Salt Breeder Reactor (MSBR). As the figure illustrates, molten thorium-fluoride would surround the core and also pass through it, although it wouldn’t mix with the molten fuel.

The job of reprocessing either fuel or fertile material from a reactor like this would obviously he relatively simple. There would be no fuel element cladding to strip off – not even a coating or matrix to dispose of. The cost of fabricating fuel elements would be zero. Refueling would be easy. Gaseous fission products could he drawn off as soon as they formed.

On the other hand, the MSBR has some drawbacks. The fuel can only produce energy while it is inside the core, yet a utility operating the MSBR would have to invest in enough fuel to fill the core and all the external piping that leads to the heat exchanger. Uranium salt may be cheap as a fuel form, but it’s pretty expensive to use as the reactor coolant (which it is in the MSBR).

The Thorium Uranium Recycling Facility ( extreme left) has beat established as part of the Oak Ridge research complex to improve all aspects of the thorium fuel cycle.

There are engineering problems that also remain unsolved. Molten salts are quite corrosive. But analyses point out that an MSBR might he able to double its fuel inventory in as short a period as 13 years. Utilities might be willing to build the comparatively simple sort of reprocessing plant that an MSBR could use right next door to each power plant, and this would provide greater savings. It wouldn’t be necessary to wait many years for each fuel batch; refueling and reprocessing could be done on a more or less continuous basis.

CONCLUSION

In spite of certain advantages in its nuclear properties, it is unlikely that 233U will ever compete with 235U or 239Pu as our principal reactor fuel. As long as this is the case, the demand for thorium (the source of 233U) will remain relatively small – probably rising to no more than a few thousand tons a year as we approach the end of this century.

Nevertheless, a deliberate program to learn more about thorium and 233U is under way both here and abroad. Thorium is being used in a variety of reactors, and more such reactors are likely to be developed and built.

As a fertile material, thorium can stretch the world’s resources of fissionable nuclear fuel, helping to postpone the need for low-grade uranium ores (which are a high cost source of fuel). The economics of nuclear power are complex, but there seems to be little doubt that the power industry of the future will use different kinds of reactors, both converters and breeders – each serving a slightly different purpose, but each complementing the others.

Raw material prices, the cost of fabricating various sorts of fuel elements, the expense of maintaining fuel inventories in a power plant, the construction requirements to meet the surging demand for electricity, the rate at which various reactors consume fuel, and the cost in time and money of reprocessing are all interrelated. Some new discovery or development might change any one of these factors tomorrow with a profound influence on the future importance of thorium.

To go back to the baseball analogies we used earlier, it might he safest to compare thorium to a somewhat inexperienced but promising pinch hitter. He might get to bat only rarely, or perhaps not at all this season, but it’s reassuring to know that he’s available on the bench if we have to call on him.

SUGGESTED REFERENCES

This booklet itself probably represents the first attempt to treat thorium and uranium-233 more than fleetingly on a popular level, and even technical literature on the subject is fairly sparse. For the reader who is interested in delving further into the prospects and problems of the third fuel, however, the following references may be helpful. For up-to-date details about the reactors mentioned in the text, inquiries should be directed to the public information office at the respective site.

Books

Fabrication of Thorium Fuel Elements (An Atomic Energy Commission-American Society of Metals Monograph), L. R. Weissert and G. Schileo, American Nuclear Society, Hinsdale, Illinois 60521, 1968, 208 pp., $11.10.
Sourcebook on Atomic Energy (third edition), Samuel Glasstone, Van Nostrand Reinhold Company, New York 10001, 1967, 883 pp., $9.25.
Thorium: Its Industrial Hygiene Aspects, Roy F. Albert, Academic Press, Inc., New York 10003, 1966, 222 pp., $7.00.
Thorium Production Technology, F. L. Cuthbert, Addison-Wesley Publishing Company, Reading, Massachusetts 01867, 1958, 303 pp., S6.50. (Out of print but available through libraries.)
Reports
Thorium Fuel Cycle (CONF-660524), proceedings of the Second International Thorium Fuel Cycle Symposium, Gatlinburg, Tennessee, May 3-6, 1966, Division of Technical Information, U. S. Atomic Energy Commission, available from Clearinghouse for Federal Scientific and Technical Information, National Bureau of Standards, U. S. Department of Commerce, Springfield, Virginia 22151, 1968, 839 pp., $3.00.
Utilization of Thorium in Power Reactors, Report of a Panel Held in Vienna, June 14-18, 1965, Technical Report Series No. 52, International Atomic Energy Agency, available from the National Agency for International Publications, 317 Fast 34th Street, New York 10016, 1966, 376 pp., $8.00.
Proceedings of the Thorium Fuel Cycle Symposium (TID-7650), Gatlinburg. Tennessee, December 5-7, 1962, Division of Technical Information Extension, U. S. Atomic Energy Commission, 1963, 766 pp., 58.00. (Out of print but available in AFC depository libraries and many metropolitan libraries.)

Joseph M. Dukert is an author and consultant who received his
BA. degree in journalism from the University of Notre Dame in
1951. For 2 years he was Assistant Director of Public Information at that University. Later lie did graduate work in international relations at Georgetown University, and was one of three U. S. Fellowship students during the first academic year of the Ifopkins School of Advanced International Studies in Bologna, Italy.
While lie was Public Relations Director for the Nuclear Division of the Martin Marietta Corporation lie developed a program that interpreted the company’s basic research work in physics, mathematics, metallurgy, and the biosciences to a diverse group of audiences, which included the general public.

He has been executive producer for several movies, one of which received a “Cindy” as the best industrial sales film of the year. Another of his films was chosen as the outstanding television newsreel film of the year at the Rome Film Festival.

He is the author of two popular science books, Atompower (Coward-McCann, 1962) and This is Antarctica (Coward-McCann, 1965), numerous booklets, magazine and newspaper articles, and movie and TV scripts.