The
Discovery of Fission
German chemist and Nobel laureate Otto Hahn and Austrian-Swedish
physicist Lise Meitner are
credited with roles in discovering nuclear fission, the process of splitting an
atom into two or more smaller parts. By the 1930s, Hahn and Meitner
had experimented with radioactive materials for some time. After Meitner left Germany in 1938, Hahn and German chemist Fritz
Strassmann made a breakthrough in their
experimentation methods that enabled them to collect physical data showing that
an atom of uranium would “burst” under certain conditions. They announced their
finding in 1939. Shortly thereafter, Meitner and
British physicist Otto Robert Frisch
published a paper that provided a theoretical explanation for the process, and
named it “fission.” This article by Hahn, which appeared in Scientific
American in 1958, describes how, through a number of twists and turns in
the road, he and his colleagues discovered the fission process in uranium.
The
Discovery of Fission
In 1939
Otto Hahn and Fritz Strassmann announced that uranium
nuclei "burst" when they are bombarded with neutrons. How they came
to this conclusion is a classic example of the nature of science
By Otto Hahn
The editors of Scientific
American have asked me to set down my personal recollections of
the discovery of uranium fission. This I am very pleased to do.
The story must go back to the year 1904, for that was the year in
which I was "transmuted" from an organic chemist into a radiochemist and thus began to learn about radioactive
elements. Armed with a letter of recommendation from the professor with whom I
had taken my doctor's degree in Germany, I went to London in 1904 with two
objectives in mind: namely, to learn English and to study with Sir William
Ramsay, the discoverer of the inert gases. Sir William, who had just become
interested in radium, gave me a dish with a barium salt which he said contained
approximately 10 milligrams of radium, and he asked me if I would purify the
radium by the method of Marie Curie. The barium salt, as it happened, came from
a mineral which contained not only radium but also a good deal of thorium, and
while working with the material I discovered a previously unknown radioactive
substance. I called it radiothorium, "a new element." (Actually it was
an isotope [an atom with the same atomic number as another atom but a different
atomic mass] of thorium, but the existence of isotopes of elements was not yet
known.)
Ramsay urged me to give up organic chemistry and devote myself to
the study of radium. Accordingly, with a view to learning more about the field,
in the fall of 1905 I went to Montreal to study with the then already famous
Professor Ernest Rutherford. There an amicable argument about the stability of
radiothorium with Rutherford's well-known associate, B. B. Boltwood
of Yale University, led me to the hypothesis that another radioactive element,
rather long-lived, must exist in the periodic table between thorium and
radiothorium. And in fact, after returning to Germany to work at the University
of Berlin, I found this substance, which I called mesothorium.
When I attempted to separate mesothorium
from radium, however, it became obvious that this new "element" was
practically indistinguishable chemically from radium. I first tried the
separation method that Madame Curie had worked out for the purification of
radium. In this method, after radium has been precipitated from the mineral
with barium as the carrier substance, the radium is separated from the barium
by fractional crystallization of the chlorides or bromides of the two elements:
the radium is concentrated in the first fraction of crystals that comes out. I
applied this crystallization treatment to the mixture of mesothorium
and radium I had obtained from a thorium-uranium mineral. But mesothorium and radium refused to crystallize in separate
fractions. And try as I would, with every conceivable method of separation, my
efforts to segregate mesothorium from radium came to
naught. I decided that the two substances were remarkably alike, but I did not
have the courage to label them chemically identical.
With the discovery of "isotopy"
a few years later by Frederick Soddy, also a former student of Rutherford and
Ramsay, everything became clear. Mesothorium was now
identifiable as an isotope of radium, and radiothorium as an isotope of
thorium. By this time I had learned also that radiothorium was a decay product
from mesothorium. The long-lived mesothorium
slowly changes, without detectable emission of radiation, to a shorter-lived
form which I called mesothorium 2; the latter is an
isotope of the radioactive element actinium. In turn, mesothorium
2, with a half-life of 6.2 hours, decays by beta-particle emission to
radiothorium.
Some of this information, as we shall see, is directly relevant to
the subject of uranium fission. And in the course of these studies I was
acquiring what was to be of the greatest help to Fritz Strassmann
and myself later in discovering the fission process: namely, a thorough
familiarity with methods of separating radioactive substances.
In the fall of 1907 an important event in my chronicle took place:
Lise Meitner, the Austrian
physicist, joined me at Berlin. She had come from Vienna to attend Max Planck's
lectures in theoretical physics, and, having free time to spare, she visited my
laboratory to work with me. What had been intended as a temporary stay in
Berlin developed into a collaboration of more than 30 years—an association
which was only terminated in the summer of 1938 through the actions of the Nazi
regime. It is perhaps needless to say that the friendship has continued.
The working facilities in the overcrowded Chemical Institute of
the University of Berlin were then very modest. Our laboratory was an
unoccupied woodworking shop. But when, in 1911, Kaiser Wilhelm II inaugurated
his Society for the Advancement of Science, we were given a radiochemical
department in the first institute set up by the Society—the Kaiser Wilhelm
Institute of Chemistry. Over the years this department developed into two large
departments, one for radiochemistry under my direction and one for nuclear
physics under Dr. Meitner.
In 1917 we discovered element 91, which we named protactinium. It
was our work on the chemical properties of this substance that gave rise later
to our keen interest in investigating the irradiation of uranium with neutrons.
This part of the story, which leads directly to the discovery of fission,
begins in the year 1932 in Rutherford's laboratory at the University of
Cambridge.
James Chadwick's discovery of the neutron in that year not only
explained the phenomenon of isotopy but also gave
physicists a new and extraordinarily effective means of producing artificial
transmutation of elements. Since neutrons carry no electric charge, they easily
enter the nucleus of an atom. The first to point out the great value of
neutrons for initiating nuclear reactions was Enrico
Fermi in Italy. He and his co-workers—Edoardo Amaldi, Oscar D'Agostino, Franco Rasetti and Emilio Segrè—bombarded
practically all the elements of the periodic table with neutrons and produced a
great number of artificial, radioactive isotopes.
Extending their experiments all the way up to uranium, the
heaviest natural element (atomic number 92), Fermi's
group found that uranium, too, gave rise to new substances, some of which
decayed very rapidly. Usually the capture of a neutron by a nucleus produces an
unstable isotope which emits a beta particle [an electron or a positron] and is
thus transformed into the next higher element. Since Fermi's products from
uranium emitted beta particles, he and his colleagues made the plausible
assumption that the transmutation produced short-lived isotopes of uranium
which then, by beta-decay, gave rise to elements beyond uranium—element 93 and
possibly even 94.
Since no such elements occur in nature, Fermi's conclusion was
widely disputed. Some physicists suggested that his best-identified new
substance—a material with a half-life of 13 minutes—was actually an isotope of
protactinium, rather than a "transuranic"
element. Naturally Dr. Meitner and I were keenly
interested, for we knew a good deal about the chemical properties of
protactinium, our discovery. We decided to repeat Fermi's experiments to try to
determine whether his substance was or was not protactinium.
Fermi's products were of course much too small in amount to be
detected in any way except with a Geiger-Müller
counter. But fortunately we had a convenient means of making a chemical test. I
had myself discovered a beta-emitting isotope of protactinium, derived from
uranium. This isotope, with a half-life of 6.7 hours, gave us a definite
marker, or indicator, of the presence of protactinium in small amounts.
Accordingly we added a certain amount of our protactinium isotope to the "transuranic" products obtained by Fermi's procedure,
and then applied chemical precipitations to the mixture. Very little of the
protactinium (less than one thousandth) came out with Fermi's newly discovered
substances. This proved unequivocally that his substances could not be protactinium.
We were also able to show with equal certainty that they were not isotopes of
thorium or actinium. That they might be lighter elements seemed entirely out of
the question. Ida Noddack did suggest, to be sure,
that we could not be certain they were transuranic
elements unless we excluded all the other elements of the periodic table as
possibilities, but this thought was considered to be wholly incompatible with
the laws of atomic physics. To split heavy atomic nuclei into lighter ones was
then considered impossible. Thus our experiments appeared to establish the
correctness of Fermi's assertion that he had detected "transuranic
elements." Actually, as we were to learn later, all of the radioactive
substances detected in these early experiments were fission products, not transuranic elements.
Over the course of years of work Lise Meitner and I, subsequently joined by Strassmann,
found a great number of radioactive transmutation products, all of which we had
to regard as elements beyond uranium. They could be arranged in a regular
series, for in chemical properties they corresponded to known elements lower in
the periodic table: namely, rhenium, osmium, iridium and platinum.
There was one product which, unlike Fermi's extremely short-lived
"uranium isotopes," had a half-life of 23 minutes. Its life was
sufficiently long for us to establish chemically that it was in fact an isotope
of uranium. Since it emitted a beta particle, it was evident that this isotope
must become an isotope of element 93, which we called eka-rhenium.
We looked for the new element, but were unable to detect it. If we had not been
convinced that we had already identified two other isotopes of element 93—an
erroneous assumption, as it turned out—we would have prepared stronger samples
of the material and made a determined effort to find the disintegration product
of our 23-minute uranium isotope. We should then have had the pleasure of
discovering element 93. Later Edwin M. McMillan and Philip H. Abelson in the U. S. identified an isotope of element 93
with a half-life of 2.3 days, and they named the element neptunium.
Subsequently, when we ourselves obtained a stronger neutron beam for
irradiating uranium, we had no trouble in detecting neptunium.
On the main road toward the discovery of fission, the next step in
this curious chronicle of near-discovery was taken by Irène
Joliot-Curie and P. Savitch in France. They were
greatly puzzled by an artificially produced substance of 3.5 hours' half-life
which they first took for an isotope of thorium, later for actinium, and
finally for a "transuranic element"
strongly resembling lanthanum. They believed that they had succeeded in
separating the substance from lanthanum by fractional crystallization, but
apparently they were misled by irrelevant separations in their mixture.
Actually their substance was undoubtedly lanthanum itself: had Madame
Joliot-Curie and Savitch recognized this, they would
have been on the verge of discovering fission.
In a paper published by the French Academy of Sciences in 1938,
Madame Joliot-Curie and Savitch concluded: "It
appears, therefore, that this substance can only be a transuranic
element, with properties very different from those of the other known transuranic elements." They considered various ways of
fitting the element into a transuranic series, but
the possibilities they discussed seemed to us highly unlikely. Strassmann and I decided to look further into their
results. (Professor Meitner had
gone to Stockholm, having been forced by the Hitler regime to leave Germany in
July, 1938.)
We found that after the "transuranic
elements" had been precipitated and removed, the solution still contained
some radioactive products. Experiments in chemical separation of these
substances now gave a remarkable result. When we used barium as the carrier,
three radioactive isotopes, with different half-lives, came down with the
barium. We were certain that these could not be accidental impurities, because
our barium precipitates were extraordinarily pure. At Strassmann's
suggestion, the carrier we used was barium chloride, which is deposited from
strong hydrochloric acid in beautiful small crystals that are free of any trace
of adsorbed impurities.
Now the precipitates had to be either barium or radium, which is
chemically similar to barium. There was nothing in the knowledge of nuclear
physics at the time to suggest that barium could possibly be produced as a
result of the irradiation of uranium with neutrons. Therefore we could only
conclude that the products must be isotopes of radium.
Still, it was a strange affair to be producing radium from uranium
under the conditions of our experiments. In order to reach radium (element 88)
from uranium (92) we had to assume that the parent element decayed by emitting two
alpha particles. [An alpha particle is two protons and two neutrons bound
together.] But we had irradiated the uranium with low-energy (thermal)
neutrons, and slow neutrons had never before been observed to produce
alpha-particle transmutations, nor was it possible to understand how they could
do so. In this case irradiation with slow neutrons was apparently yielding more
intense nuclear reactions than had ever been produced by bombardment of a
substance with fast neutrons.
We continued with our experiments and were able to distinguish no
fewer than four isotopes of "radium" produced artificially from
uranium. We called them radium I, II, III and IV. Their half-lives were less
than one minute (an estimate), 14 minutes, 86 minutes and approximately 300 hours,
respectively.
Now since we were dealing with very weak preparations, we
undertook to separate the radium isotopes from the carrier substance, barium,
to obtain thinner layers of material, so that their radioactivity could be more
easily measured. We used the method of fractional crystallization with which,
30 years earlier, I had separated the radium isotope mesothorium
from barium.
Our attempts to concentrate our artificial radium isotopes by this
method came to nothing. We then tried a more effective process that we had
developed, using chromates of the substances instead of their chlorides or
bromides. All in vain.
It next occurred to us that the "radium" might
conceivably be failing to separate from the barium because it was present only
in extremely small amounts—so few atoms that they could be detected only with
the Geiger-Müller counter. To test this possibility,
we put some natural, definitely identified radium through the same experiments.
We meticulously purified this radium and diluted it to intensities as weak as
those of the artificial "radium" isotopes in the aforementioned
experiments. But this time the definitely known radium, in spite of its small
amount, separated from barium upon crystallization as one should expect.
Finally we turned to "indicator" experiments such as I
had used to check whether Fermi's product was protactinium. We mixed one of our
supposed artificial radium isotopes with a known natural isotope of radium and
then tried to separate the mixture from barium by fractionation as before. The
experiments and their results were rather involved, and I shall not attempt to
describe them here in detail. But the findings were quite clear: in each case
most of the natural radium in the mixture separated out, but the artificial "radium"
(e.g., Ra III or Ra IV) stayed with the barium in toto,
according to the radioactivity measurements. In short, our artificial
"radium" could not be separated from barium for the simple reason
that it was barium!
In January, 1939, we published an account of these
"experiments that are at variance with all previous experiences in nuclear
physics." In interpreting the experiments we expressed ourselves very
cautiously, partly because the series of tests had not yet been quite
finished—they took several weeks. But our caution was not due to any mistrust
of our results. Indeed, we already had a strong check of our conclusion, for we
had identified a decay product of one of our "radium" isotopes as
lanthanum, which meant that the parent had to be not radium but barium.
Our overcautiousness stemmed primarily
from the fact that as chemists we hesitated to announce a revolutionary
discovery in physics. Nevertheless we did speak of the "bursting" of
uranium, as we called the surprising process that had yielded barium, far down
in the periodic table. In this first paper we also speculated on what the other
partner of the splitting of the uranium atom might be. Not being physicists, we
thought of uranium's atomic weight (238) rather than the number of its protons
(92). Subtracting the atomic weight of barium (137) from that of uranium, we
guessed at 101 as the atomic weight of the other fragment. This could be an
isotope of technetium or of one of the medium-heavy metals.
Immediately after our paper appeared, Meitner
and Otto R. Frisch came out independently with their historic publication
showing how Niels Bohr's model of the atom could
explain the cleavage of a heavy nucleus into two nuclei of medium size. Meitner and Frisch named the process "fission."
Subtracting the nuclear charge of barium (56) from that of uranium (92), they
identified the other fission product as element 36, the inert gas krypton. And
indeed, Strassmann and I had actually detected this
element, or rather, its decay product, strontium, among the products of the
neutron irradiation of uranium. Now in quick succession we discovered other
fission products, determined their chemical properties and traced their
subsequent transformations. We also found that thorium, like uranium, underwent
fission when bombarded by fast neutrons.
This, in broad outline, is the story of the discovery of fission.
The fact that the process is accompanied by the release of enormous amounts of
energy was soon ascertained by Meitner and Frisch,
and, independently, by John R. Dunning, Fermi and co-workers in the U. S. and
by Frédéric Joliot-Curie in France. Strassmann and I conjectured, but did not prove
experimentally, that neutrons were liberated in the process. This was later
proved by Joliot-Curie, H. von Halban and L. Kowarski. The liberation of neutrons of course made it
possible to harness the vast amount of energy released by the fission of
uranium in the form of a "chain reaction," and this possibility was
pointed out as early as mid-1939 by, among other workers, S. Flügge of the Kaiser Wilhelm Institute of Chemistry.
Strassmann and I did not
concern ourselves with the harnessing of the energy of fission. During the war
we continued, together with two or three co-workers, to investigate the complex
fission processes and to publish our findings. By the beginning of 1945 we had
made up a table containing approximately 100 fission products and their
transformations. But the science of fission was then a classified subject in
the U. S., and our results were not published there until November, 1946.
Source: Reprinted with permission. Copyright ©
February 1958 by Scientific American, Inc. All rights reserved.
Microsoft ® Encarta ® Reference Library 2003. ©
1993-2002 Microsoft Corporation. All rights reserved.
The term fission was first used by
the German physicists Lise Meitner and Otto Frisch
in 1939 to describe the disintegration of a heavy nucleus into two lighter
nuclei of approximately equal size. The conclusion that such an unusual nuclear
reaction can in fact occur was the culmination of a truly dramatic episode in
the history of science, and it set in motion an extremely intense and
productive period of investigation.
The story of the discovery of nuclear
fission actually began with the discovery of the neutron
in 1932 by James Chadwick in England (see above). Shortly thereafter, Enrico Fermi and his associates in Italy undertook an
extensive investigation of the nuclear reactions produced by the bombardment of
various elements with this uncharged particle. In particular, these workers
observed (1934) that at least four different radioactive species resulted from
the bombardment of uranium with slow neutrons. These newly discovered species
emitted beta particles and were thought to be isotopes of unstable "transuranium elements" of atomic
numbers 93, 94, and perhaps higher. There was, of course, intense interest
in examining the properties of these elements, and many radiochemists
participated in the studies. The results of these investigations, however, were
extremely perplexing, and confusion persisted until 1939 when Otto
Hahn and Fritz
Strassmann in Germany, following a clue provided
by Irène Joliot-Curie and Pavle
Savic in France (1938), proved definitely that
the so-called transuranic elements were in fact
radioisotopes of barium, lanthanum, and other elements in the middle of the
periodic table.
That lighter elements could be formed
by bombarding heavy nuclei with neutrons had been suggested earlier (notably by
the German chemist Walter Noddack in 1934), but the
idea was not given serious consideration because it entailed such a broad
departure from the accepted views of nuclear physics and was unsupported by
clear chemical evidence. Armed with the unequivocal results of Hahn and Strassmann, however, Meitner and
Frisch invoked the recently formulated liquid-drop model of the nucleus (see
above) to give a qualitative theoretical interpretation of the fission process
and called attention to the large energy release that should accompany it.
There was almost immediate confirmation of this reaction in dozens of
laboratories throughout the world, and within a year more than 100 papers
describing most of the important features of the process were published. These
experiments confirmed the formation of extremely energetic heavy particles and
extended the chemical identification of the products.
The chemical evidence that was so
vital in leading Hahn and Strassmann to the discovery
of nuclear fission was obtained by the application of carrier and tracer
techniques. Since invisible amounts of the radioactive species were formed,
their chemical identity had to be deduced from the manner in which they
followed known carrier elements, present in macroscopic quantity, through
various chemical operations. Known radioactive species were also added as
tracers and their behaviour was compared with that of the unknown species to
aid in the identification of the latter. Over the years, these radiochemical
techniques have been used to isolate and identify some 34 elements from zinc
(atomic number 30) to gadolinium (atomic number 64) that are formed as fission
products. The wide range of radioactivities produced
in fission makes this reaction a rich source of tracers for chemical, biologic,
and industrial use.
Although the early experiments
involved the fission of ordinary uranium with slow neutrons, it was rapidly
established that the rare isotope uranium-235
was responsible for this phenomenon. The more abundant isotope uranium-238
could be made to undergo fission only by fast neutrons with energy exceeding 1 MeV. The nuclei of other heavy elements, such as thorium
and protactinium, also were shown to be fissionable with fast neutrons; and
other particles, such as fast protons, deuterons, and alphas, along with gamma
rays, proved to be effective in inducing the reaction.
In 1939, Frédéric
Joliot-Curie, Hans von Halban, and Lew Kowarski found that several
neutrons were emitted in the fission of uranium-235, and this discovery led to
the possibility of a self-sustaining chain reaction. Fermi and his coworkers recognized the enormous potential of such a
reaction if it could be controlled. On Dec. 2, 1942, they succeeded in doing
so, operating the world's first nuclear
reactor. Known as a "pile," this device consisted of an array of
uranium and graphite blocks and was built on the campus of the University of
Chicago.
The secret Manhattan Project,
established not long after the United States entered World War II, developed
the atomic bomb. Once the war had ended, efforts were made to develop new
reactor types for large-scale power generation, giving birth to the nuclear
power industry.
The fission process may be best
understood through a consideration of the structure and stability of nuclear
matter. Nuclei
consist of nucleons
(neutrons and protons), the total number of which is equal to the mass number
of the nucleus. The actual mass of a nucleus is always less than the sum of the
masses of the free neutrons and protons that constitute it, the difference
being the mass equivalent of the energy of formation of the nucleus from its
constituents. The conversion of mass to energy follows Einstein's equation, E
= mc2, where E is the energy equivalent of a mass, m,
and c is the velocity of light. This difference is known as the mass
defect and is a measure of the total binding
energy (and, hence, the stability) of the nucleus. This binding energy is
released during the formation of a nucleus from its constituent nucleons and
would have to be supplied to the nucleus to decompose it into its individual
nucleon components.
A curve illustrating the average
binding energy per nucleon as a function of the nuclear mass number is shown in
Figure
1
Figure 1: The average binding energy per nucleon as a function
of the mass number, A (see...
. The largest binding energy (highest stability) occurs near mass number 56--the mass region of the element iron.Figure
1Figure
1: The average binding energy per nucleon as a function of the mass number, A (see...
On
the basis of energy considerations alone,Figure 1
Figure
1: The average binding energy per nucleon as a function of the mass number, A (see...
Figure
2: The potential energy as a function of elongation of a fissioning
nucleus. G</I...
Figure
2: The potential energy as a function of elongation of a fissioning
nucleus. G</I...