CONGRESSES AND CONFERENCES

ALL-UNION CONFERENCE ON PROBLEMS IN ATOMIC NUCLEAR PHYSICS

S. Ya. Nikitin, Leningrad

In November of last year the Academy of Sciences of the USSR convened in Kharkov the fourth conference on problems in atomic nuclear physics. At the conference 35 reports were heard, in which the following problems of contemporary nuclear physics were considered: 1) cosmic rays, 2) properties of fast electrons and γ-rays, 3) properties of heavy particles and nuclear reactions, 4) ways of practically using the achievements of nuclear physics, 5) the technique of producing fast charged particles.

The present state of the problem of cosmic rays was elucidated in the introductory report by D. V. Skobeltzyn (FIAN, Moscow), who pointed out that at present the interpretation of all phenomena of cosmic radiation is based essentially on two basic theories. The shower theory must, and apparently is able to, explain the behavior of the soft component. All phenomena caused by the hard component must, according to present ideas, be placed in the meson theory, organically connected with the problem of nuclear forces. However, as Skobeltzyn pointed out, the results of the experiments of K. I. Alekseeva, which established the practical absence of decay electrons, cast doubt on the very foundations of the meson theory. Skobeltzyn then dwelt on the exceptional difficulties of interpreting the nature of the soft component.

In original reports relating to the first item of the conference program, the most essential problems of cosmic radiation were touched upon: the determination of the number of decay electrons accompanying the hard component (report by K. I. Alekseeva), an experimental check of the cascade theory of showers (report by S. N. Vernov), studies of the splitting of atomic nuclei by cosmic rays (report by A. P. Zhdanov), and a study of secondary slow mesons (report by V. I. Veksler).

According to present ideas, mesons—the particles constituting the penetrating component—are stable and may spontaneously decay into an electron and a neutrino. The mean value, determined from various indirect experiments, of the meson lifetime proved to be equal to \(2 \cdot 10^{-6}\) sec. Thus, if the hypothesis of meson decay is correct, the soft component at sea level must consist of δ-particles (of the equilibrium part of the soft component) and decay electrons. Calculations show that, if the above value of the meson lifetime is adopted, then at sea level the decay electrons should constitute 25% of the penetrating component.

In the experiments of K. I. Alekseeva (FIAN, Moscow) the relative intensity of the soft component was measured with respect to the hard component in air and under a filter made of wood, which completely absorbs decay electrons.

The difference between these two quantities should give the relative intensity of decay electrons, since the equilibrium intensities of the soft component (δ-particles) in air and in wood are the same. According to Alekseeva’s data, the effect being sought is \(8\%\) instead of the expected \(25\%\).

The discussion that took place on the question of meson decay showed that at the present time one cannot put forward any reasonably plausible hypothesis explaining the results of Alekseeva’s experiments without at the same time raising the question of abandoning the hypothesis of meson decay.

At present it is generally accepted that the interaction with matter of fast electrons that are part of cosmic radiation is well described by the so-called cascade theory. This theory is able to describe qualitatively all observed effects. However, cascade theory is in sharp contradiction with experimental data when the passage of electrons through substances with a high atomic number is concerned. The most substantial contradiction consists in the fact that when electrons fall on lead there should occur a considerable increase in the number of particles; in experiment, however, no such increase is observed. In the report by S. N. Vernov (FIAN, Moscow) it was shown that this very sharp contradiction is due not to the erroneousness of the basic physical idea of cascade theory, but is connected with an underestimation of the role of losses of energy to ionization and scattering of electrons inside the substance. The attempt made by S. N. Vernov to take these two factors into account leads to a substantially closer agreement of cascade theory with experiment.

The report by V. I. Veksler (FIAN, Moscow) was devoted to the study of slow mesons of cosmic radiation by means of proportional counters. Veksler found that cosmic radiation contains a very large number of slow mesons, having a range of approximately \(1.5\ \mathrm{g/cm^{2}}\). It should be noted that the observed number of slow mesons considerably exceeds the number of mesons of the same component that end their range in the substance, as a consequence of energy loss inside the proportional counter. The presence of slow mesons undoubtedly indicates some completely new kind of interaction of mesons with matter.

The observation of nuclear disintegrations caused by cosmic rays by ordinary methods (for example, with the aid of a Wilson chamber or Geiger-Müller counters) is not possible because of the extremely small probability of these processes. The most effective method for studying this phenomenon is the method of thick-layer photographic plates, developed by L. V. Mysovsky and A. P. Zhdanov. The essence of the method is that cosmic rays cause the disintegration of atomic nuclei in the photographic emulsion. The products of disintegration, in passing through the emulsion, cause the development of its individual grains. Thus, the track of a charged particle on the plate will consist of a whole thread of developed grains.

A. P. Zhdanov (RIAN, Leningrad) demonstrated at the conference several extraordinarily interesting cases of the disintegration of atomic nuclei. The most interesting disintegration was the case of a shower consisting of approximately one hundred heavy charged particles.

The second section of the conference reports was opened by the report of A. K. Walter (UFTI, Kharkov), devoted to the measurement of the radiative losses of energy by fast electrons in lead. As is known, the losses of energy by fast electrons consist of two parts, which play different roles at different electron energies. Ionization losses play the main role at comparatively low energies (for lead, several million volts), whereas the principal mechanism of energy loss by very fast electrons is radiative braking. In A. K. Walter’s experiments the radiative losses were measured by the relative calorimetric method. A beam of electrons accelerated in the tube of a Van de Graaff electrostatic generator entered alternately lithium and lead calorimeters. The wall thickness of the calorimeter was chosen to be somewhat greater than the range of electrons in the given substance. In lithium, practically all

energy of the electrons is spent on ionization, whereas in lead a considerable fraction of the energy is spent on radiation and, consequently, is not registered by the calorimeter. Thus, to a first approximation, the difference in thermal effects in lithium and in lead is a measure of the energy losses of electrons to radiation. The value of the radiation losses in lead found by A. K. Valter is in good agreement with the theory of Bethe and Heitler.

The question of the radiation losses of energy by fast electrons in aluminum, copper, tin, and lead was also the subject of a report by L. A. Artsimovich (LPTI, Leningrad). A special feature of this work by L. A. Artsimovich is the use, for the study of bremsstrahlung, of extremely thin layers of matter. In this arrangement of the experiment the process of bremsstrahlung appears in pure form. The data obtained by L. A. Artsimovich are in good agreement with theory.

In recent years a whole series of works has been devoted to the elastic scattering of fast electrons. However, the data obtained by different authors diverge sharply. The results of some authors are in agreement with theoretical data (for example, A. I. Alikhanov, Sen-Gupta, and others); other authors, on the contrary, obtain values of the effective cross section for scattering that differ from the theoretical value by several times, and sometimes by several tens of times (L. V. Skobeltsyn and E. G. Stepanova, Bothe and Klarmann). A report by M. D. Borisov (UPhTI, Kharkov) was devoted to this very confused question; in it were set forth the results of an investigation of the scattering of electrons by nitrogen nuclei. The investigation was carried out with the aid of a Wilson chamber. The data obtained by Borisov are in good agreement with the theoretical ones.

In the discussion of the question of the scattering of fast electrons, A. I. Leipunsky expressed the supposition that the above-mentioned discrepancies between the results of individual authors are apparently connected with some methodological shortcomings of the Wilson chamber that are still unknown.

A large report by G. D. Latyshev (UPhTI, Kharkov) was devoted to the study of the properties of hard γ-radiation. Latyshev studied with extreme precision the Compton recoil electrons from the γ-rays of Th (C + C″) and RaC. The investigation was carried out with the aid of a spectrometer (after Danysz), in the focus of which was placed a plate of a light substance irradiated by γ-rays. The data obtained on the Compton effect made it possible to measure the relative intensities of the γ-lines. An interesting case was noted by Latyshev with respect to the γ-line Th (C + C″) 3,200 MeV, which had been discovered by Alikhanov and coworkers from the positrons of internal conversion of the γ-rays Th (C + C″). Careful searches for recoil electrons corresponding to this γ-line did not lead to a positive result. Thus the following situation is obtained: the γ-line appears in the spectrum of positrons of internal conversion, but is absent according to the data of the recoil-electron spectrum; this is apparently explained by the fact that a direct transition with emission of a γ-quantum from the excitation level 3,200 MeV to the ground level is forbidden, and therefore the decay of the level 3,200 can occur only through the direct interaction of nuclei with electrons at negative energy levels.

The report by L. V. Groshev (FIAN, Moscow) was devoted to the question of pair production by γ-rays in nitrogen. In L. V. Groshev’s work, some essential features of this phenomenon were studied, such as: the distribution of pair components by energy, the angular distribution of pairs, and the determination of the effective cross section. The value of the effective cross section for pair production obtained by Groshev is in rather good agreement with the theoretical value.

In his report A. B. Migdal (LPTI, Leningrad) subjected to theoretical consideration the ionization of atoms in the case of a sudden change in the charge of the nucleus. Such a process takes place, for example, in radioactive decay or in the fission of a uranium nucleus. As the calculation shows, the probability of ionization of shells in radioactive decay for light nuclei is approximately equal to unity, which, apparently, agrees with experiment.

...data. An estimate of the charge of uranium fragments leads to the conclusion that the charge \(Z\) of these fragments lies approximately within the limits \(16 > Z > 7\).

S. Ya. Nikitin (LPTI, Leningrad) gave a report on a new method of the double magnetic spectrometer for investigating the \(\beta\)-spectra of radioactive elements. The method developed represents a certain development of the Danysh method. With the aid of the new spectrometer, the end of the \(\beta\)-spectrum of RaE was investigated, and it was found that in the immediate vicinity of the upper limit the \(\beta\)-spectrum curve can be represented in the form of two rectilinear segments and that it intersects the abscissa axis at a certain angle. The experimental results are in complete agreement with earlier investigations of the end of the \(\beta\)-spectrum of RaE carried out by Alikhanov, Alikhanyan, and Dzhelepov.

A. I. Leipunskii (UFTI, Kharkov) gave a large survey report on the problem of the splitting of uranium and thorium nuclei by neutrons. As is known, one of the greatest achievements in the field of nuclear physics of the past year consisted in the discovery of a new type of nuclear splitting. When bombarded with neutrons, uranium and thorium atomic nuclei undergo splitting into two nuclei of approximately equal mass with the release of 175 MeV of energy. At the present time it is generally accepted that slow neutrons cause fission of the rare isotope of uranium \(U^{235}\), whereas fission of the principal isotope of uranium \(U^{238}\) and of thorium is caused only by fast neutrons. Further, it may be considered established that good known resonant absorption of neutrons with energy of 25 eV\(^1\) by uranium nuclei does not cause fission, but leads to the formation of the \(\beta\)-radioactive isotope \(U^{239}\). Soon after the discovery of uranium fission it was established that the fission process is accompanied by the emission of fast neutrons. According to the most accurate Fermi data, on average one and a half neutrons are emitted per fission. The circumstance that uranium fission is accompanied by the emission of several neutrons makes it possible to count on carrying out a chain reaction of uranium fission. The new neutrons produced as a result of fission will cause fission of other uranium nuclei, etc. It is quite obvious that such a chain reaction will be accompanied by colossal liberation of heat.

The possibility of obtaining a chain reaction for the first time places the question of the practical utilization of intranuclear energy on a real basis.

A. P. Zhdanov (RIAN, Leningrad) demonstrated at the meeting several photographs of tracks of uranium fission products in a Wilson chamber. A characteristic feature of these tracks is an extremely large specific ionization.

G. A. Petrzhak (RIAN, Leningrad), in the work reported at the meeting, studied the ranges of uranium fragments. He found that uranium fragments constitute two groups differing in their range in air. This makes it possible to conclude that there are two types of uranium fission.

V. G. Khlopin (RIAN, Leningrad) dwelt on the chemical nature of the fragments of the uranium nucleus. Chemical investigations showed that the primary products of uranium fission may be xenon and strontium or krypton and barium. However, the data obtained by V. G. Khlopin and collaborators show that xenon and krypton, at least in some cases, are intermediate products of uranium fission. According to V. G. Khlopin’s data one may consider that there are two types of uranium fission—into antimony and niobium and into molybdenum and tin.

To elucidate the possibility of chain fission of uranium it is necessary to know the effective cross sections for fission of uranium, the number of neutrons emitted in fission, and the effective cross sections of all processes leading to useless, from the point of view of uranium fission, absorption of neutrons.

G. N. Flerov (LPTI, Leningrad) reported the result of his work on determining the number of neutrons emitted in uranium fission, and the physical proof that resonant capture of neutrons with ener—

\(^1\) According to new data, 5 eV (S. N.).

by a 25 eV nucleus of U\(^{238}\) does not lead to fission. As the result of Flerov’s work shows, the number of neutrons emitted in the fission of uranium is equal to \(3 \pm 1\).

In Yu. B. Khariton’s (LPTI, Leningrad) substantive report, the possibilities of a chain reaction in uranium fission were analyzed in detail. The report considered the following two basic questions—the possibility of a chain reaction (bearing in mind the experimental data on the elementary processes connected with this phenomenon) and the kinetics of the reaction (if the latter is to be considered possible).

Calculation shows that obtaining a chain reaction without slowing down the neutrons emitted in uranium fission is possible only on condition that pure metallic uranium is used; moreover, the number of neutrons emitted in fission must be not less than 2.5–3. However, by the most probable present data, the number of neutrons per fission does not exceed 1.5. Thus, obtaining a chain reaction with fast neutrons is, in all likelihood, not possible.

The qualitative picture of a chain reaction with slow neutrons differs from the one just mentioned. Slow neutrons cause fission only of the rare isotope U\(^{235}\). As calculation shows, a chain reaction is also possible with slow neutrons if a hydrogen-containing substance is mixed with the uranium as a neutron moderator. There are a number of possibilities for reducing the useless absorption of neutrons (for example, using deuterium, helium, or carbon as the moderator). The most realistic path for obtaining a chain reaction is, in all probability, enrichment of uranium with the isotope U\(^{235}\). As for the kinetics of the reaction, here the question of greatest interest is whether the chain reaction will lead to an extremely violent explosion, or whether there are certain ways of reducing the rate at which the reaction proceeds.

As calculation shows, the main factor slowing the course of the reaction will be the thermal expansion of the mass of uranium.

At present all the elementary processes determining the possibility of a chain reaction have not yet been studied sufficiently well; therefore the conclusions reached by the calculations of Yu. B. Khariton and Ya. B. Zel’dovich may change substantially, depending on further work on the study of these elementary processes.

I. I. Gurevich (RIAN, Leningrad), in his presentation, gave a very ingenious explanation of the circumstance that extremely large effective cross sections for the capture of slow neutrons by rare-earth elements are observed experimentally. This fact must be interpreted as meaning that the average distance between resonance levels of the nuclei of rare-earth elements has a nonmonotonic variation with atomic weight.

To explain such a dependence of the average distance between levels on atomic weight, Gurevich put forward the hypothesis that each heavy nucleus may exist, depending on the excitation energy, in two different thermodynamic states. Thus we arrive at the hypothesis of a phase transition of nuclear matter, analogous to the well-known phase transitions of matter. This hypothesis quite naturally explains the behavior of the effective cross section for the capture of slow neutrons as a function of atomic weight.

The study of neutron scattering by nuclei is of considerable interest for constructing a theory of the interaction of neutrons with atomic nuclei. In the work by T. Goloborod’ko (UFTI, Kharkov), presented by him at the conference, the scattering of photoneutrons (RaTh + Be) by various substances was investigated. The results of Goloborod’ko’s experiments show that the effective scattering cross section does not vary monotonically with atomic weight, but changes irregularly from element to element.

At present it is impossible to give an unambiguous explanation of this phenomenon.

The question of the interaction of neutrons with nuclei was also the subject of a report by D. V. Timoshuk (UFTI, Kharkov) on the absorption of fast neutrons.

Effective cross sections for the absorption of fast neutrons for a whole series of elements vary extremely sharply from one element to another; this undoubtedly indicates that theoretical consideration of this question must take into account a number of individual features of nuclei.

The problem of nuclear isomerism was presented in a report by L. I. Rusinov (LFTI, Leningrad). As is known, about two years ago L. I. Rusinov and A. A. Iozefovich discovered soft electron radiation in the β-spectrum of radioactive isomeric bromine Br⁸⁰. It was then suggested that this soft radiation consists of internal-conversion electrons.

In the work reported by L. I. Rusinov, the task was set of directly proving the above hypothesis. Further, the paper measured the internal-conversion coefficient of soft γ-radiation, which in practice turned out to be equal to unity. This latter circumstance makes it possible to estimate the difference between the angular momenta of the ground state of the Br⁸⁰ nucleus and the metastable state of the same nucleus.

Two reports by Ya. I. Frenkel (LFTI, Leningrad) concerned the further development of the droplet model of the atomic nucleus proposed by him. In the first report Ya. I. Frenkel analyzed surface oscillations of a nucleus-drop, taking into account the Coulomb repulsion of protons. It was shown that these oscillations can lead to the rupture of the nucleus into two parts, as occurs, for example, in the fission of the uranium nucleus. Ya. I. Frenkel’s second report concerned a theoretical study of the γ-ray spectrum emitted by atomic nuclei. The above-mentioned droplet model of the nucleus was taken as the basis of the consideration.

A report by I. P. Selinov (LFTI, Leningrad) was devoted to a review of results and to the discussion of a number of problems in investigations of stable and radioactive isotopes.

At the conference, the question of the practical use of radioactive substances was discussed for the first time. A major introductory report on this topic was read by Academician V. G. Khlopin (RIAN, Leningrad).

The broadest field of application of radioactive substances is their use for chemical analysis. The radiations emitted by these substances serve as an extraordinarily sensitive and accurate indicator. Radioactive substances can be used with great success in studying the kinetics of chemical reactions, as well as in investigations of the chemistry of complex compounds.

Next, the following ways of using radioactive substances should be indicated: the medical use of radioactive substances, chiefly in the treatment of malignant tumors; biological use, analogous to the use of X-rays; and, finally, the use of hard γ-radiation for radiographing metal castings and reinforced-concrete structures in order to detect internal flaws, cavities, etc.

A. A. Grinberg (RIAN, Leningrad) delivered a report on the use of radioactive elements for the study of complex compounds.

S. Z. Roginskii (LIKHF, Leningrad) reported to the conference on his work on the use of radioactive indicators for studying the properties of liquid solutions.

The introductory report for the last section of the conference program—on the technique of obtaining fast particles—was read by F. F. Lange (UFTI, Kharkov). The speaker gave a detailed survey of all contemporary methods of accelerating charged particles. The speaker’s main attention was devoted to the three methods of acceleration currently basic: the impulse generator, the electrostatic Van de Graaff generator, and the cyclotron. The first two can be used successfully to accelerate electrons, while the cyclotron is the most effective method for accelerating ions, one that has recently undergone exceptionally broad development and dissemination.

B. M. Gokhberg (LFTI, Leningrad) reported on extensive work to develop a new compact electrostatic generator, which is a modification of the Van de Graaff generator in compressed gas.

At present they have built a model of a 1 MV generator. Such generators may find application in medicine for roentgenotherapy, and also in the cable industry for testing purposes.

An extremely interesting report on a new ion source was delivered by V. S. Gott (UPTI, Kharkov). Electrons are used as the ionizing agent in the new ion source; moreover, in order to make maximum use of the electrons’ path length, their paths are curved by the application of a magnetic field. With the aid of the new source it will be possible to obtain extremely large ion currents (on the order of 10 A and more).

L. I. Pivovar (UPTI, Kharkov) spoke about the compact pulsed 1 MV generator developed at UPTI. The generator that has been developed, owing to its compactness and ease of operation, can be used successfully in medical institutions as well as in factory laboratories.

The final report at the meeting was delivered by V. S. Shpinel (UPTI, Kharkov), who described the nature of the cathode radiation of a pulsed generator. The investigations showed that the electron spectrum in the tube of a pulsed generator has a linear character.

The meeting adopted a number of resolutions on organizational questions, in particular on the publication of literature on problems of atomic-nuclear physics, and also on the annual publication of tables of nuclear constants, entrusting the compilation of the first table to I. P. Selinov.

A resolution was also adopted to convene the next meeting in the autumn of 1940.