Quantum generations, p.26
Quantum Generations, page 26
THE RISE OF NUCLEAR PHYSICS
THE ELECTRON-PROTON MODEL
SHORTLY AFTER the nuclear model of atomic structure had been accepted, several physicists started speculating about the structure of the atom’s tiny nucleus. The general view was that proposed by Rutherford, namely, that the nucleus was made up of electrons and positive unit particles, the latter identical with hydrogen nuclei and often termed “positive electrons” or H-particles; or, from 1920 on, protons. It appeared evident that the nucleus included electrons, for the positive particles were clearly in need of some negative electricity to prevent the nucleus from exploding. Moreover, it was known since 1913 that the beta electrons had their origin in the nucleus and not in the outer layers of electrons. This had been argued by Bohr, among others, who in his 1913 work pointed out that “the fact that two apparently identical elements [isotopes] emit β-particles of different velocities, shows that the β-rays as well as the α-rays have their origin in the nucleus” (Bohr 1963, 53).
According to the nuclear atomic model, the mass number A and the atomic number Z would depend on the number of protons (p) and electrons (e) as A = p and Z = p – e. On the other hand, nuclei of radioactive bodies also gave rise to alpha particles, which therefore were often assumed to be additional nuclear constituents. For a nucleus with a alpha particles, the equations would then be A = 4a + p and Z = 2a + p – e. This hypothesis enjoyed general acceptance between 1915 and 1932. In fact, not a single physicist seems to have doubted the nuclear electron hypothesis and what has rightly been called the two-particle paradigm, namely, that all matter consisted of electrons and protons (although some of these might exist in bound forms, as alpha particles or other combinations). How, then, were the two or three nuclear species arranged in the nucleus? Given the almost complete lack of experimental evidence, it was a hopeless task to construct reliable nuclear models in the 1910s and 1920s. Nonetheless, a surprisingly large number of physicists (and chemists, too) were undeterred by the difficulties and speculated more or less freely on nuclear structures. Most of these models were pure speculations, often based on loose numerological arguments, but a few were of a more serious nature. For example, in 1918 the Munich physicist Wilhelm Lenz constructed a model of the alpha particle in accordance with the rules of quantum theory, namely, four protons revolving in an equatorial plane with one electron at each pole. Sommerfeld referred approvingly to this model in his Atombau.
The most eminent of the earliest generation of nuclear physicists was also the most prolific of the model makers, and not the least speculative. Inspired by his earlier alpha scattering experiments, Rutherford suggested in his 1920 Bakerian lecture that other particles in the nucleus than electrons, alphas, and protons might exist. Rutherford argued that there was evidence for a light helium nucleus, consisting of three protons bound by one electron in Rutherford’s notation), and perhaps also a heavy hydrogen isotope consisting of two protons and one electron. And why not? the nucleus also contained a neutral particle made up of one electron and one proton, a “neutron,” according to Rutherford. It was also at this occasion that Rutherford introduced the name “proton.” Rutherford was particularly fascinated by the possibility of neutrons because these would “enter readily the structure of atoms, and may either unite with the nucleus or be disintegrated by its intense field, resulting possibly in the escape of a charged H atom [proton] or an electron or both” (Badash 1983, 886). Rutherford may not have been aware that his “neutrons” had been proposed as early as 1899 by the Australian physicist William Sutherland, who suggested that the ether consisted of doublets of positive and negative electrons. Sutherland’s suggestion was taken over by Nernst in his authoritative textbook Theoretical Chemistry, where it appeared in all editions between 1903 and 1926. Whereas the idea of light helium nuclei was based on weak experimental evidence and abandoned in 1924, the neutron hypothesis had a long life and was taken quite seriously in Cambridge. James Chadwick believed in its existence as much as Rutherford, and tried on several occasions in the 1920s to detect the hypothetical particle. He did not succeed until 1932, and then it turned out that the observed neutral particle was not Rutherford’s neutron after all. Rutherford continued to develop his ideas of nuclear structure, and in 1925 he reached the conclusion that the nucleus consisted of a massive core surrounded by positive and negative satellites (protons and electrons). He found his satellite model important enough to include it in Radiations from Radioactive Substances, written with Chadwick and Charles D. Ellis and published in 1930.
Most of Rutherford’s hypotheses were based on interpretations of experiments in which substances were bombarded with alpha particles. In December 1917 he wrote to Bohr, “I am detecting & counting the lighter atoms set in motion by α particles & the results, I think, throw a good deal of light on the character & distribution of forces near the nucleus. I am also trying to break up the atom by this method” (Stuewer 1986a, 322). The most important of these experiments was made in Manchester in 1919, shortly before Rutherford left for Cambridge to become director of the Cavendish Laboratory. In a reinvestigation of experiments made earlier by Ernest Marsden, Rutherford studied the action of alpha particles on various gases by detecting the scintillations produced by long-range particles formed by the action. With pure nitrogen, he observed what he called an anomalous effect, namely, the production of long-range particles similar to those obtained with hydrogen. “It is difficult to avoid the conclusion,” he wrote, “that these long-range atoms arising from the collision of alpha particles with nitrogen are not nitrogen atoms but probably charged atoms of hydrogen, or atoms of mass 2. If this be the case, we must conclude that the nitrogen atom is disintegrated under the intense forces developed in a close collision with a swift α particle, and that the hydrogen atom which is liberated formed a constituent part of the nitrogen nucleus” (Beyer 1949, 136). Further work done in the Cavendish proved that Rutherford’s conclusion was largely correct. He had achieved the first artificial disintegration of an atomic nucleus and thus opened up a new stage in the history of modern alchemy. The process was 14N + 4He → 17O + 1H, although Rutherford originally interpreted it as 14N + 4He → 13C + 4He + 1H. It was only in 1924, when cloud chamber photographs failed to show the track of any alpha particles from the recoil atoms, that the error was corrected.
During the following years, Rutherford and his Cavendish group continued this kind of experiments in the hope of transforming still more elements. Similar work was done at the Vienna Radium Institute, but not with the same results. Whereas Rutherford and Chadwick did not find evidence for disintegration of elements heavier than potassium, nor for beryllium and lithium, the Vienna physicists Gerhard Kirsch and Hans Petterson (the latter of whom was Swedish) claimed to be much more successful in disintegrating elements. Not only did they report results widely different from those produced in Cambridge, but they also attacked Rutherford’s satellite model of the nucleus. The disagreement evolved into a protracted controversy with some of the same components that characterized the notorious N-ray episode at the beginning of the century. As a result of a visit Chadwick made to the Vienna Institute in 1927, he found that the Austrian-Swedish team did not control their results and that the counting of the scintillations was systematically biased toward the (too) high values they wanted to find. According to an historian: “The counting was done by women, the reasoning being that they could concentrate on the task more intensely than men, having little on their minds anyway, and by Slavic women because their large, round eyes were best suited for counting. The women were told what sort of counting rate was anticipated, and, being anxious to please, they provided it” (Badash 1983, 887).
Nuclear physics in the 1920s was intimately linked with radioactivity, the only source of high-energy projectiles being the alpha and beta particles emitted by naturally occurring radioactive substances. In order to measure the intensity and direction of particles, either scattered or produced by disintegration, simple scintillation devices were normally used. Visual counting of scintillations went back to 1908, when Erich Regener of Berlin University concluded that each alpha particle hitting a phosphorescent screen produced a scintillation. The simple method was used extensively until the early 1930s. Rutherford’s experiment of 1919 did not use apparatus more advanced than that used in Geiger and Marsden’s alpha scattering experiments some ten years earlier. Lack of money and available technology made the Cavendish experiments simple and in agreement with Rutherford’s own predilection toward sealing wax and string methods. Yet it was not all sealing wax and string. One of the most important instruments in the infancy of nuclear physics was the electromagnetic mass spectrograph, developed by Francis Aston from its first version in 1919. By the late 1920s, the mass spectrograph had become a complicated and expensive instrument.
For detection purposes, the cloud chamber began to play in important role in the 1920s. The principle of making tracks of ionized droplets visible by means of sudden expansion was discovered by Charles T. R. Wilson in the Cavendish Laboratory in the late 1890s in connection with meteorological studies. By 1911, Wilson had constructed the first cloud chamber to study the paths of ionizing particles and taken the first cloud chamber photograph. In 1921 T. Shimizu, a Japanese physicist working at the Cavendish, found a means for working the chamber automatically, and the technique was further improved by Patrick M. S. Blackett. Practically all the innovative work on scintillation and cloud chamber techniques was done in the Cavendish Laboratory, which was also the birthplace of the gas ionization chambers or counters. The most effective version of the early ionization counters was designed by Hans Geiger in 1913. With the starting of the war, Geiger returned to Germany to serve in the artillery; after 1918, he continued his work on ionization counters. The modern, highly sensitive Geiger-Müller counter was a German invention. It was developed in 1928 by Geiger and his collaborator at the University of Kiel, Walther Müller. The development in detection methods in the 1920s and later owed much to electronics engineering, especially the use of vacuum tube circuits.
QUANTUM MECHANICS AND THE NUCLEUS
Quantum mechanics was a general theory of atoms and electrons. It was assumed to hold for the atomic nucleus as well, but during the first phase of quantum mechanics, there were no attempts to apply the new theory to nuclear physics. The situation changed in the summer of 1928, when it was shown that alpha radioactivity could be understood in terms of quantum mechanics. The important quantum-mechanical theory of alpha decay was proposed independently by George Gamow in Göttingen and Ronald Gurney and Edward Condon in Princeton. Gamow, a twenty-four-year-old Russian physicist, argued (as Rutherford had done earlier) that the nuclear potential must be strongly attractive for very small distances and attain a maximum height before it merged with the repulsive Coulomb potential. He pictured the alpha particles as preexisting in the nucleus, vibrating or orbiting within the potential well. Classically, the alpha particle would not be able to penetrate the potential, but according to Born’s interpretation of Schrödinger’s wave mechanics, there would be a finite probability that a particle escaped the nucleus with an energy smaller than the maximum height of the potential. This is the famous case of a particle or matter wave penetrating or “tunneling” a potential barrier, a case that today enters all introductory textbooks in quantum mechanics but in 1928 was a new and exciting phenomenon.
Gamow’s use of quantum mechanics enabled him to find the penetration probability and, after translating it into the decay constant, derive a linear relationship between the logarithm of the decay constant and the energy of the emitted alpha particles. This was just the Geiger-Nuttall relationship, which had been known empirically since the 1912 work of Geiger and the English physicist John Nuttall. There had earlier been derivations of the Geiger-Nuttall law for example, by Frederick Lindemann in 1915 but these were pseudo-explanations based on ad hoc assumptions. Gamow’s explanation, as well as the largely identical one proposed by Gurney and Condon, was much more satisfactory because it was based on fundamental theory. The Gamow-Gurney-Condon theory was extremely important, both because it provided a convincing demonstration that quantum mechanics applies to the atomic nucleus and because it formed the basis of other applications of quantum mechanics to nuclear physics.
The statistical nature of radioactivity had been a puzzle ever since it was recognized in the early part of the century. Numerous attempts had been made to provide a causal explanation of the origin of radioactivity, but it was only with quantum mechanics that it was realized that such attempts to make sense of radioactivity’s statistical nature were futile. As Gurney and Condon put it in 1929, referring to these attempts: “This has been very puzzling so long as we have accepted a dynamics by which the behaviour of particles is definitely fixed by the conditions. We have had to consider the disintegration as due to the extraordinary conjunction of scores of independent events in the orbital motions of nuclear particles. Now, however, we throw the whole responsibility on to the laws of quantum mechanics, recognizing that the behaviour of particles everywhere is equally governed by probability” (Kragh 1997a, 357).
During the years following 1928, quantum mechanics was successfully applied to several other problems involving atomic nuclei, among which collision problems were particularly important. For example, in 1928 Nevill Mott, a twenty-three-year-old Cambridge physicist, reproduced Rutherford’s 1911 expression for the scattering of charged particles from a pointlike nucleus. More interestingly, for collisions between two identical particles (such as alpha particles scattered by a helium gas) he predicted a result that differed from Rutherford’s, namely, that for low velocities scattering at 45° would occur about twice as frequently as classically expected. The prediction, confirmed by cloud chamber experiments by Patrick Blackett and Frank Champion in 1931, was one of the first novel predictions of quantum mechanics in the nuclear regime.
The works of Gamow, Mott, and Gurney and Condon did not shed any new light on the structure of the nucleus, which was still assumed to consist of electrons and protons. These were the only known elementary particles and, besides, radioactive nuclei emitted electrons in the form of beta particles. On the other hand, during the late 1920s it was gradually realized that somehow, electrons ought to have no place in the atomic nucleus. They were necessary, but unwelcome. One of the problems of the electron-proton model was that it did not agree with the experimentally determined statistics of nuclei. Studies of the rotational spectrum of the N2+ molecule showed that the spin of the nitrogen nucleus must be one. But if the nucleus consisted of 14 protons and 7 electrons, an odd number of particles with spin one-half, it should itself have spin one-half. The discrepancy between measurements and theoretical expectations was pointed out by Ralph Kronig, who suggested in 1928 that “probably one is therefore forced to assume that protons and electrons do not retain their identity to the extent they do outside the nucleus” (Pais 1986, 301). Neither Kronig nor others could be more concrete at that time. The following year, studies of Raman spectra confirmed the result that the nitrogen nucleus obeyed Bose-Einstein statistics, that is, possessed an integral spin. In Göttingen, Walter Heitler and Gerhard Herzberg amplified Kronig’s conclusion: “[I]t seems as if the electron in the nucleus loses, along with its spin, also its right of participation in the statistics of the nucleus” (ibid., 302).
Even more important than the nitrogen anomaly was the problem of understanding the beta spectrum. In 1914 Chadwick had found that the spectrum of beta radioactivity was continuous, although mixed with a line spectrum. According to Chadwick and the Cavendish physicists, the continuous spectrum was the real one, whereas the discrete lines had their origin in, for example, an inner photoelectric effect in the electron system such as proposed by Charles Ellis in 1922. However, it was possible to account for the spectrum without assuming the beta electrons to be emitted with energies in a continuous range. In Berlin, Lise Meitner suggested that the electrons started out with the same energy but that some of the energy was converted to gamma radiation, which would produce secondary beta rays. Meitner’s alternative led to a protracted controversy with the Cavendish scientists. The controversy was solved only in the late 1920s, when experiments proved incompatible with Meitner’s theory. It was now firmly established that the continuous beta spectrum had its origin in the nucleus. This conclusion, however, was most discomforting from a theoretical point of view. According to quantum mechanics, a nucleus can exist only in discrete energy states; assuming energy conservation, a two-particle decay into a daughter nucleus and a beta electron cannot, therefore, reproduce the continuous spectrum.
Together with the spin-statistics problem and problems of relativistic quantum mechanics, the continuous beta spectrum led to a kind of crisis in parts of the physics community in 1929 31. Niels Bohr’s answer to the crisis was radical, namely, that energy conservation fails in beta decay. In an unpublished note of June 1929, he emphasized “how little basis we possess at present for a theoretical treatment of the problem of β-ray disintegrations.” He continued: “Indeed, the behaviour of electrons bound within an atomic nucleus would seem to fall entirely outside the field of consistent application of the ordinary mechanical concepts, even in their quantum theoretical modification. Remembering that the principles of conservation of energy and momentum are of a purely classical origin the suggestion of their failure in accounting for β-ray emission can on the present state of quantum theory hardly be rejected beforehand.” Bohr continued to advocate violation of energy conservation for at least three years; he was supported by some of the younger physicists, including Gamow and Landau. Gamow was the author of the first textbook ever in nuclear physics, in its modern sense: a book titled Constitution of Atomic Nuclei and Radioactivity and prefaced on May 1, 1931. At that time, nuclear physics was still a nascent field. The book, aimed to give “as complete an account as possible of our present experimental and theoretical knowledge of the nature of atomic nuclei,” filled 114 pages. Gamow referred approvingly to Bohr’s idea of energy nonconservation and wrote, in complete accord with his master in Copenhagen, “The usual ideas of quantum mechanics absolutely fail in describing the behaviour of nuclear electrons; it seems that they may not even be treated as individual particles, and also the concept of energy seems to lose its meaning” (p. 5).
