Asimov's New Guide to Science

by Isaac Asimov

  Fermi and his associates discovered that they got better results if they slowed down the neutrons by passing them through water or paraffin first. Bouncing off protons in the water or paraffin, the neutrons are slowed just as a billiard ball is when it hits other billiard balls. When a neutron is reduced to thermal speed (the normal speed of motion of atoms), it has a greater chance of being absorbed by a nucleus, because it remains in the vicinity of the nucleus longer. Another way of looking at it is to consider that the length of the wave associated with the neutron is longer, for the wavelength is inversely proportional to the momentum of the particle. As the neutron slows down, its wavelength increases. To put it metaphorically, the neutron grows fuzzier and takes up more volume. It therefore hits a nucleus more easily, just as a bowling ball has more chance of hitting a tenpin than a golf ball would have.

  The probability that a given species of nucleus will capture a neutron is called its cross section. This term, metaphorically, pictures the nucleus as a target of a particular size. It is easier to hit the side of a barn with a baseball than it is to hit a foot-wide board at the same distance. The cross sections of nuclei under neutron bombardment are reckoned in trillion-trillionths of a square centimeter (10–24 square centimeter where I square centimeter is a little less than one-sixth of a square inch). That unit, in fact, was named a barn by the American physicists M. C. Holloway and C. P. Baker in 1942. The name served to hide what was really going on in those hectic wartime days.

  When a nucleus absorbs a neutron, its atomic number is unchanged (because the charge of the nucleus remains the same), but its mass number goes up by one unit. Hydrogen I becomes hydrogen 2, oxygen 17 becomes oxygen 18, and so on. The energy delivered to the nucleus by the neutron as it enters may excite the nucleus—that is, increase its energy content. This surplus energy is then emitted as a gamma ray.

  The new nucleus often is unstable. For example, when aluminum 27 takes in a neutron and becomes aluminum 28, one of the neutrons in the new nucleus soon changes to a proton (by emitting an electron). This increase in the positive charge of the nucleus transforms the aluminum (atomic number 13) to silicon (atomic number 14).

  Because neutron bombardment is an easy way of converting an element to the next higher one, Fermi decided to bombard uranium to see if he could form an artificial element-number 93. In the products of the bombardment of uranium, he and his co-workers did find signs of new radioactive substances. They thought they had made element 93, and called it uranium X. But how could the new element be identified positively? What sort of chemical properties should it have?

  Well, element 93, it was thought, should fall under rhenium in the periodic table, so it ought to be chemically similar to rhenium. (Actually, though no one realized it at the time, element 93 belonged in a new rare-earth series, which meant that it would resemble uranium, not rhenium—see chapter 6. Thus, the search for its identification got off on the wrong foot entirely.) If it were like rhenium, perhaps the tiny amount of “element 93” created might be identified by mixing the products of the neutron bombardment with rhenium and then separating out the rhenium by chemical methods. The rhenium would act as a carrier, bringing out the chemically similar “element 93” with it. If the rhenium proved to have radioactivity attached to it, this would indicate the presence of element 93.

  Otto Hahn and Lise Meitner, the discoverers of protactinium, working together in Berlin, pursued this line of experiment. Element 93 failed to show up with rhenium. Hahn and Meitner then went on to try to find out whether the neutron bombardment had transformed uranium into other elements near it in the periodic table. At this point, in 1938, Germany occupied Austria, and Meitner, who, until then, as an Austrian national, had been safe despite the fact that she was Jewish, was forced to flee from Hitler’s Germany to the safety of Stockholm. Hahn continued his work with the German physicist Fritz Strassman.

  Several months later, Hahn and Strassman found that barium, when added to the bombarded uranium, carried off some radioactivity. They decided that this radioactivity must belong to radium, the element below barium in the periodic table. The conclusion was, then, that the neutron bombardment of uranium changed some of it to radium.

  But this radium turned out to be peculiar stuff. Try as they would, Hahn and Strassman could not separate it from the barium. In France, Irène -Curie and her co-worker P. Savitch undertook a similar task and also failed.

  And then Meitner, the refugee in Scandinavia, boldly cut through the riddle and broadcast a thought that Hahn was voicing in private but hesitating to publish. In a letter published in the British journal Nature in January of 1939, she suggested that the “radium” could not be separated from the barium because no radium was there. The supposed radium was actually radioactive barium: it was barium that had been formed in the neutron bombardment of uranium. This radioactive barium decayed by emitting a beta particle and formed lanthanum. (Hahn and Strassman had found that ordinary lanthanum added to the products brought out some radioactivity, which they assigned to actinium; actually it was radioactive lanthanum.)

  But how could barium be formed from uranium? Barium was only a middleweight atom. No known process of radioactive decay could transform a heavy element into one only about half its weight. Meitner made so bold as to suggest that the uranium nucleus had split in two. The absorption of a neutron had caused it to undergo what she termed fission. The two elements into which it had split, she said, were barium and element 43, the element above rhenium in the periodic table. A nucleus of barium and one of element 43 (later named technetium) would make up a nucleus of uranium. What made it a particularly daring suggestion was that neutron bombardment only supplied 6 million electron-volts, and the main thought of the day concerning nuclear structure made it seem that hundreds of millions would be required.

  Meitner’s nephew, Otto Robert Frisch, hastened to Denmark to place the new theory before Bohr, even in advance.of publication. Bohr had to facc the surprising ease with which this would require the nucleus to split, but fortunately he was evolving the liquid-drop theory of nuclear structure, and it seemed to him that this would explain it. (In later years the liquid-drop theory, taking into account the matter of nuclear shells, was to explain even the fine details of nuclear fission and why the nucleus breaks into unequal halves.)

  In any case, theory or not, Bohr grasped the implications at once. He was just leaving to attend a conference on theoretical physics in Washington, and there he told physicists what he had heard in Denmark of the fission suggestion. In high excitement, the physicists went back to their laboratories to test the hypothesis; and within a month half a dozen experimental confirmations were announced. The Nobel Prize for chemistry went to Hahn in 1944 as a result.

  THE CHAIN REACTION

  The fission reaction released an unusual amount of energy, vastly more than did ordinary radioactivity. But it was not solely the additional energy that made fission so portentous a phenomenon. More important was the fact that it released two or three neutrons. Within two months after the Meitner letter, the awesome possibility of a nuclear chain reaction had occurred to a number of physicists.

  A chain reaction is a common phenomenon in chemistry. The burning of a piece of paper is a chain reaction. A match supplies the heat required to start it; once the burning has begun, this supplies the very agent, heat, needed to maintain and spread the flame. Burning brings about more burning on an ever-expanding scale.

  That is similar to a nuclear chain reaction. One neutron fissions a uranium nucleus, thus releasing two neutrons that can produce two fissions that release four neutrons which can produce four fissions, and so on (figure 10.3). The first atom to fission yields 200 Mev of energy; the next step yields 400 Mev, the next 800 Mev, the next 1,600 Mev, and so on. Since the successive stages take place at intervals of about a 50 trillionth of a second, you see that, within a tiny fraction of a second, a staggering amount of energy will be released. (The actual average number of neutrons produced per f
ission is 2.47, so matters go even more quickly than this simplified calculation indicates.) The fission of 1 ounce of uranium produces as much energy as the burning of 90 tons of coal or of 2,000 gallons of fuel oil. Peacefully used, uranium fission could, in theory, relieve all our immediate worries about vanishing fossil fuels and our mounting consumption of energy.

  Figure 10.3. Nuclear chain reaction in uranium. The gray circles arc uranium nuclei; the black dots, neutrons; the wavy arrows, gamma rays; and the small circles, fission fragments.

  But the discovery of fission came just before the world was plunged into an all-out war. The fissioning of an ounce of uranium, physicists estimated, would yield as much explosive power as 600 tons of TNT. The thought of the consequences of a war fought with such weapons was horrible, but the thought of a world in which Nazi Germany laid its hands on such an explosive before the Allies did was even more horrible.

  The Hungarian-American physicist Leo Szilard, who had been thinking of nuclear chain reactions for years, foresaw the possible future with complete clarity. He and two other Hungarian-American physicists, Eugene Wigner and Edward Teller, prevailed on the gentle and pacific Einstein in the summer of 1939 to write a letter to President Franklin Delano Roosevelt, pointing out the potentialities of uranium fission and suggesting that every effort be made to develop such a weapon before the Nazis managed to do so.

  The letter was written on 2 August 1939 and was delivered to the President on II October 1939. Between those dates, the Second World War had erupted in Europe. Physicists at Columbia University, under the supervision of Fermi, who had left Italy for America the previous year, worked to produce sustained fission in a large quantity of uranium.

  Eventually the government of the United States itself took action in the light of Einstein’s letter. On 6 December 1941, President Roosevelt (taking a huge political risk in case of failure) authorized the organization of a giant project, under the deliberately noncommittal name of Manhattan Engineer District, for the purpose of devising an atom bomb. The next day, the Japanese attacked Pearl Harbor, and the United States was at war.

  THE FIRST ATOMIC PILE

  As was to be expected, practice did not by any means follow easily from theory. It took a bit of doing to arrange a uranium chain reaction. In the first place, you had to have a substantial amount of uranium, refined to sufficient purity so that neutrons would not be wasted in absorption by impurities. Uranium is a rather common element in the earth’s crust, averaging about 2 grams per ton of rock, which makes it 400 times as common as gold. But it is well spread out, and there are few places in the world where it occurs in rich ores or even in reasonable concentration. Furthermore, before 1939 uranium had had almost no uses, and no methods for its purification had been worked out. Less than an ounce of uranium metal had been produced in the United States.

  The laboratories at Iowa State College, under the leadership of Spedding, went to work on the problem of purification by ion-exchange resins (see chapter 6) and, in 1942, began to produce reasonably pure uranium metal.

  That, however, was only a first step. Now the uranium itself had to be broken down to separate out its more fissionable fraction. The isotope uranium 238 (U-238) has an even number of protons (92) and an even number of neutrons (146). Nuclei with even numbers of nucleons are more stable than those with odd numbers. The other isotope in natural uranium—uranium 235—has an odd number of neutrons (143). Bohr had therefore predicted that it would fission more readily than uranium 238. In 1940, a research team, under the leadership of the American physicist John Ray Dunning, isolated a small quantity of uranium 235 and showed that Bohr’s conjecture was true. U-238 fissions only when struck by fast neutrons of more than a certain energy, but U-235 will undergo fission upon absorbing neutrons of any energy, all the way down to simple thermal neutrons.

  The trouble was that in purified natural uranium only one atom in 140 is U-235, the rest being U-238. Thus, most of the neutrons released by fission of U-235 would be captured by U-238 atoms without producing fission. Even if the uranium were bombarded with neutrons fast enough to split U-238, the neutrons released by the fissioning U-238 would not be energetic enough to carry on a chain reaction in the remaining atoms of this more common isotope. In other words, the presence of U-238 would cause the chain reaction to damp and die. It would be like trying to burn wet leaves.

  There was nothing for it, then, but to try for a large-scale separation of U-235 from U-238, or at least the removal of enough U-238 to effect a substantial enrichment of the U-235 content in the mixture. The physicists attacked this problem by several methods, each of them offering only thin prospects of success. The one that eventually worked best was gaseous diffusion. This remained the method of choice, though fearfully expensive, until 1960. A West German scientist then developed a much cheaper technique of U-235 isolation by centrifugation, the heavier molecules being thrown outward and the lighter ones, containing U-235, lagging behind. This process makes nuclear bombs cheap enough for minor powers to manufacture, a consummation not entirely to be desired.

  The uranium-235 atom is 1.3 percent less massive than the uranium-238 atom. Consequently, if the atoms were in the form of a gas, the U-235 atoms would move about slightly faster than the U-238 atoms and thus might be separated, by reason of their faster diffusion, through a series of filtering barriers. But first uranium had to be converted to a gas. About the only way to get it in this form was to combine it with fluorine and make uranium hexafluoride, a volatile liquid composed of one uranium atom and six fluorine atoms. In this compound, a molecule containing U-235 would be less than 1 percent lighter than one containing U-238, a difference that proved sufficient to make the method work.

  The uranium hexafluoride vapor was forced through porous barriers under pressure. At each barrier, the molecules containing U-235 got through a bit faster, on the average; and so with every passage through the successive barriers, the advantage in favor of U-235 grew. To obtain sizable amounts of almost pure uranium-235 hexafluoride required thousands of barriers, but well-enriched concentrations of U-235 could be achieved with a much smaller number of barriers.

  By 1942, it was reasonably certain that the gaseous diffusion method (and one or two others) could produce enriched uranium in quantity; and separation plants (costing a billion dollars each, and consuming as much electricity as all of New York City) were built at the secret city of Oak Ridge, Tennessee, sometimes called Dogpatch by irreverent scientists, after the mythical town in Al Capp’s Li’l Abner.

  Meanwhile, the physicists were calculating the critical size that would be needed to maintain a chain reaction in a lump of enriched uranium. If the lump was small, too many neutrons would escape from its surface before being absorbed by U-235 atoms. To minimize this loss by leakage, the volume of the lump had to be large in proportion to its surface. At a certain critical size, enough neutrons would be intercepted by U-235 atoms to keep a chain reaction going.

  The physicists also found a way to make efficient use of the available neutrons. Thermal (that is, slow) neutrons, as I have mentioned, are more readily absorbed by uranium 235 than are fast ones. The experimenters therefore used a moderator to slow the neutrons from the rather high speeds they had on emerging from the fission reaction. Ordinary water would have been an excellent slowing agent, but unfortunately the nuclei of ordinary hydrogen hungrily snap up neutrons. Deuterium (hydrogen 2) fills the bill much better; it has practically no tendency to absorb neutrons. Consequently the fission experimenters became very interested in preparing supplies of heavy water.

  Up to 1943, it was prepared by electrolysis for the most part. Ordinary water split into hydrogen and oxygen more readily than did heavy water, so that, if a large supply of water were electrolyzed, the final bit of water was rich in heavy water and could be preserved. After 1943, careful distillation was the favored method. Ordinary water had the lower boiling point, so that the last bit of unboiled water was rich in heavy water.

  He
avy water was indeed valuable in the early 1940s. There is a thrilling story of how Joliot-Curie managed to smuggle France’s supply of that liquid out of the country ahead of the invading Nazis in 1940. A hundred gallons of it, which had been prepared in Norway, did fall into the hands of the German Nazis. It was destroyed by a British commando raid in 1942.

  Still, heavy water had drawbacks: it might boil away when the chain reaction got hot, and it would corrode the uranium. The scientists seeking to create a chain-reacting system in the Manhattan Project decided to use carbon, in the form of very pure graphite, as the moderator.

  Another possible moderator, beryllium, had the disadvantage of toxicity. Indeed, the disease, berylliosis, was first recognized in the early 1940s in one of the physicists working on the atom bomb.

  Now let us imagine a chain reaction. We start things off by sending a triggering stream of neutrons into the assembly of moderator and enriched uranium. A number of uranium-235 atoms undergo fission, releasing neutrons that go on to hit other uranium-235 atoms. They in turn fission and turn loose more neutrons. Some neutrons will be absorbed by atoms other than uranium 235; some will escape from the pile altogether. But if from each fission one neutron, and exactly one, takes effect in producing another fission, then the chain reaction will be self-sustaining. If the multiplication factor is more than one, even very slightly more (for example, 1.001), the chain reaction will rapidly build up to an explosion. This is good for bomb purposes but not for experimental purposes. Some device had to be worked out to control the rate of fissions. That could be done by sliding in rods of a substance such as cadmium, which has a high cross section for neutron capture. The chain reaction develops so rapidly that the damping cadmium rods could not be slid in fast enough, were it not for the fortunate fact that the fissioning uranium atoms do not emit all their neutrons instantly. About 1 neutron in 150 is a delayed neutron emitted a few minutes after fission, since it emerges, not directly from the fissioning atoms, but from the smaller atoms formed in fission. When the multiplication factor is only slightly above 1, this delay is sufficient to give time for applying the controls.