On Magnetic Spectroscopy and Radio Spectroscopy of the Atomic Nucleus

V. K. Arkad'ev

A little more than a quarter of a century ago we advanced a general point of view[^1] on the study of the electromagnetic properties of matter; we then called spectrology the doctrine of the active and passive spectra of matter, determined by its electrical and magnetic properties. By spectra we meant functions of frequency, period, or wavelength expressing the named properties or any relation of matter, as well as of a body, to radiant energy. From this point of view, the scale of waves cannot be called a spectrum.

By active spectra we meant emission spectra; by passive spectra, absorption spectra of all kinds, dispersion curves, the properties of matter in isolated oscillating electric and magnetic fields, etc. At that time we also introduced the concept of electric and magnetic spectra. The test of time has shown that this was not in vain. At present, between active and passive spectra we must place transformation spectra, i.e., spectra that arise as a result of a change in the frequency of the oscillations of the initial radiation and the appearance of a new frequency composition. These are the spectra of fluorescence, combination scattering, nonlinear distortions, etc.[^3]

Alongside this systematics of the spectrology of waves and oscillations one may place the equally important question of the systematics of the spectra of beams of elementary particles. A typical example of such a spectrum is provided by the mass spectrograph. For more than 50 years researchers have worked with molecular beams: at first these were cathode “rays,” then anode rays, streams of atoms in the experiments of Gerlach and Stern, and molecular beams of many other kinds.

Against the background of this general classification of spectra of waves, oscillations, and beams, we wish here to consider the connection between magnetic spectra of all kinds, from the normal passive spectra of matter to the radio spectra of atomic nuclei. One may think that a systematic classification of spectra in their mutual connection will help clarify the state of these questions, in the presentation of which ambiguities are sometimes encountered.

The diagram (p. 82) presents both systems of spectroscopy. The older spectroscopy of waves and oscillations is the more developed; it deals with the spectra of matter as collections of a large number of elementary particles. It is subdivided into the spectroscopy of active spectra, transformation spectra, and passive spectra.

Passive spectra are curves of the coefficients of refraction (dispersion), absorption, or other characteristics of matter. Absorption may occur when rays pass through a body or when they are reflected from its surface. Passive spectra also include the curves of the dielectric and magnetic permeability of a substance or body situated inside a capacitor or a coil. Permeability is understood, of course, as a complex quantity, consisting of a conservative or real part and a consumptive or imaginary part.

Active, transformation, and passive spectra can often be subdivided into electric and magnetic. By electric spectra we mean spectra determined by the electric properties of matter: dielectric permeability and electrical conductivity, which depend on the mass and electric charge of the individual elementary particles, their moment of inertia, electric moment, damping of motion, their proper period, their number per unit volume, etc. By magnetic spectra we mean spectra determined by magnetic properties: magnetic susceptibility or permeability, which depend on the magnetic moment and the number of centers per unit volume, on their proper period, on the structure of the ferromagnet, on magnetic viscosity and hysteresis, etc.

To the left of this general diagram a scale of waves is given. Depending on the wavelength or frequency at which a spectrum is observed, in the diagram at the corresponding height we have placed cells with the name of the phenomenon or source that determine the spectrum and the corresponding kind of radiation. For brevity, the name of the investigator associated with the given phenomenon or method is sometimes given.

1. The column of electric active spectra is essentially the content of the usual scale of electromagnetic oscillations and waves, as given in 1901 by P. N. Lebedev and developed in 1936 by A. A. Glagoleva-Arkad’eva. The oblique section denotes the overlap, demonstrated by the work of A. A. Glagoleva-Arkad’eva, of waves of atomic-molecular origin (long infrared waves) with microwaves obtained by means of electrical instruments—in this region, with a mass radiator. Plasma and the Sun are indicated as sources of radio waves that had not previously been shown on the scale.

General spectroscopy of waves and oscillations

Scale of waves and oscillations

• Gamma rays
• X-rays
• Ultraviolet
• Visible
• Infrared rays
• Infrared waves
• Microwaves
• Radio frequencies
• Audio frequencies
• Infralow frequencies

Active spectra

Magnetic

• Hagen and Rubens, \(\mu = 1\), 1903
• Magnetic antenna?

Electric

• Atomic-molecular sources, Moscow
• Excitation by spark: Arkadiev, 1923
• Electrical apparatus
• Zeeman effect: Stark, 1913
• Plasma, 1926; Solntse, 1944

Transformation spectra

Electric

• Reflection of a photon from an electron, 1922
• Combination scattering: Mandelstam–Landsberg, Tamm, Raman, 1928
• Luminescence: Vavilov, Levshin, 1922–1940
• Distorting crystal, 1946
• Spark spectrography, 1940
• Ferroelectrics: Kurchatov, 1930

Magnetic

• Ferromagnetics; transverse induction, 1944
• Griffiths, 1946
• Zavoisky, 1945
• Paramagnetic resonance, 1938
• Nuclear induction, 1946
• Kurchatov phenomenon, 1950

Passive spectra

Magnetic

• Ziskin, 1885; Hagen and Rubens, 1903
• Magnetic resonance: Arkadiev, 1912
• Magnetic antenna, 1930
• Spectra of viscosity: Velichko, 1935
• Lavrentiev, 1949

Electric

• Lutz, Wulf, 1912; Holweck, 1920
• Rozhdestvenskii, 1911; Vavilov, 1920–1922; Kravets, 1911; Rubens, 1903
• Echo-Ochs, 1921
• Grane, 1924; Moffe, 1912
• Electron diffraction, 1923

Spectroscopy of beams of elementary particles

• \(e/m\) of the electron, 1897
• Mass spectrograph, 1919
• Geiger and Müller, 1924
• Rabi and collaborators, 1937
• Vavilov–Cherenkov radiation, 1934

1. In the column of electrical passive spectroscopy we find spectra, methods, and the associated names that occur on the ordinary scale of waves: Laue, Holweck, and Wulff in the region of gamma rays and X-rays; Vavilov, Rozhdestvenskii, Terenin, and Kravets in the region of light rays (dispersion and absorption); Rubens in the region of infrared rays. Now one should include here the name of the apparatus—the echo box, proposed by us in the 1920s. This is a method for measuring microwave absorption in a closed cavity. In the region of decimeter waves we encounter the names of many Soviet scientists who studied the properties of liquids; in the region of infralow frequencies very many worked on dielectrics, and in constant fields—A. F. Ioffe and his school. By constant fields on the scale of periods we understand fields of such duration as reaches the duration of manipulations in laboratory experiments.

2. In the column of transformation spectra there is quite understandably the presence of phenomena of spectral change under the influence of external magnetic and electric fields (the Zeeman effect and the Stark effect), as well as the change of frequency in the collision of a photon with an electron (1922), in combination scattering (the Landsberg, Mandelstam, and Raman effect, 1928), and in the phenomena of phosphorescence and fluorescence studied in detail by Vavilov and Levshin with their collaborators. Here we also find crystals that distort oscillations produced by klystrons and generate waves up to 5 mm in length (1946), the transformation of centimeter waves into light waves in the screens of Arkad’ev and Penner (1940) for illumination by microwaves, and at lower frequencies we encounter nonlinear distorters of electric fields—ferroelectrics, and of magnetic fields—ferromagnets. This is a region of phenomena in which, in our country, I. V. Kurchatov (1930) and G. V. Dobrovol’skii (1943) worked not a little. In recent years Bernstein, Gorelik, and Zhukova have explained by the nonlinearity of a ferromagnet the phenomenon they discovered of “transverse induction” (1944). The emf arising in this case decomposes into a spectrum, which Gorelik calls the magnetic transformation spectrum[^2].

3. Let us now proceed to a more detailed consideration of magnetic spectra.

The absence of an influence of ferromagnetic properties on the optical spectra of substances was first established in the 1880s by Zeeman, and later by Drude. Investigating the reflection of light from metallic surfaces, one can find that the phenomena proceed as if the magnetic permeability of steel and nickel were equal to unity. Later Rubens extended these investigations into the infrared region, measuring the reflectivity of metals for waves of 4, 8, and 12 microns. It depends on their specific electrical conductivity and on their magnet-

...permeability. Since the former, in waves with a wavelength of 12 μ and more, does not depend on the wavelength, absorption upon reflection can provide a measure of magnetic permeability.

In the region of longer waves the reflection coefficient differs so little from unity that Rubens had to replace measurements of the absorptive capacity of metals by measurements of their emissive capacity, which, according to the laws of radiation, must be proportional to the absorptive capacity. Thus, he replaced the study of passive electromagnetic spectra by the study of active spectra. His measurements of the emissive capacity of all metals, including ferromagnetic ones, gave such values of emissive capacity that, for waves up to 25.5 microns, they require taking the magnetic permeability to be equal to unity. According to our terminology, this is so far the only true magnetic active spectrum obtained up to the present time.

The study of the radiation of magnetic antennas*) in the centimeter-wave region may serve as a method for investigating the magnetic properties of their material. With some qualification, and perhaps even without qualification, this method may be regarded as the study of the magnetic emission spectrum, i.e. the active magnetic spectrum.

1. Very numerous and varied investigations of the magnetic properties of matter have been carried out by observing their passive spectra. We have seen that in the far infrared region the permeability of ferromagnetic metals is equal to unity. However, in the 1890s it was firmly established that in meter waves the permeability of iron and nickel has large values.

Its gradual increase from the infrared toward longer waves was first discovered by me (1908–1912) in centimeter waves by studying the passive spectra of iron and nickel in the wavelength range from 1.3 cm to 73 cm. In this region a magnetic resonance was then found, i.e. sharply expressed dispersion and absorption of wave power with wavelengths of 2–6 cm, depending on the material and its treatment. The presence of the resonance was later confirmed by the investigations of Birks (1947), and a considerable decrease of permeability in decimeter waves was confirmed by the work of Hoag and John (1932), Zenger and Potapenko (1933), and Lindman (1938). A fall of the permeability to 1 at a wavelength of 1.3 cm was observed by E. Maxwell (1946).

*) A magnetic antenna is usually the name given to the rod of magnetodielectric, magnetized by decimeter waves, which we first used in 1930³.

This is the normal magnetic passive spectra characterizing a substance (iron, nickel) in its natural state.

In experiments with a magnetic antenna demagnetized at the focus of a parabolic mirror, in 1930 the possibility was demonstrated of overcoming hysteresis by waves \(60\ \text{cm}\) long. Demagnetization in waves down to \(23\ \text{cm}\) was carried out by M. M. Chetverikova.

The study of passive magnetic spectra of soft iron and permalloy in the region of ultrahigh and sonic frequencies, down to infrared, reveals here the spectra of magnetic viscosity predicted by us in 1928.[^4] They were discovered by O. I. Veletskaya in 1935 and later studied by V. M. Goytannikov, A. I. Pilshchikov, and S. S. Lavrent’ev (1940–1945).[^5] Our theory of the magnetic spectra of viscosity was also confirmed in studies of ferrites at radio frequencies.

1. The magnetic spectra that arise in a substance under the action of an external strong magnetic field, we classify as magnetic transformation spectra. In 1936 such magnetic spectra in paramagnets were observed by Gorter. These spectra are most often obtained by changing the intensity of the external field at a constant frequency of the high-frequency field, and also by changing the frequency of the latter (from 0.15 to \(5\ \text{MHz}\)) at an unchanging constant field. These were viscosity spectra obeying the laws of our magnetic spectra of viscosity of ferromagnets, obtained in 1928.[^4] The resonance of paramagnets was first observed by E. K. Zavoisky (1947), who used an alternating field of centimeter waves (\(16\ \text{cm}\)).

Resonance peaks in ferromagnets under analogous conditions were observed by Griffiths in 1946, to whom the Anglo-American literature persistently ascribes priority in the discovery of the magnetic resonance of ferromagnets*). Applying fields from 500 to 5000 gauss, he measured \(\mu\rho\)—the product of permeability by specific resistance. His curves, plotted in arbitrary units, reveal peaks of \(\mu\rho\) at wavelengths of 1.22 and \(3.18\ \text{cm}\) (cobalt, iron, nickel).

) Meanwhile, articles on magnetic resonance of ferromagnets entitled “Oscillations and Resonance of Elementary Magnets” and “Magnetic Resonance” were published by us in Russian and foreign languages in Reports of the Academy of Sciences of the USSR (1927), in Comptes Rendus (1926), and in the collection Contemporary Problems of Electromagnetism* (Moscow, 1931), not to mention a number of our articles on the theory of the field in ferromagnets published since 1913. From this it can be seen that the theoretical and experimental study of the magnetic resonance of elementary carriers of magnetism was undertaken by our national science long before the dates indicated in the Anglo-American literature.

For the first time, the action of an external field on the normal magnetic spectra of the viscosity of a ferromagnet was observed by I. M. Kirko in Riga (1950). Applying fields from units up to 156 oersteds to steel rods, he found that the dispersion and absorption bands regularly shift into the region of higher frequencies; moreover, it was established that the laws of viscous spectra, first verified on ferromagnets by Veletskaya, remain valid: the height of the absorption band is equal to half the drop in permeability and lies at the frequency of the middle of the drop.

I. M. Kirko and his collaborators—V. A. Yanushkovskii, B. O. Groskaufman and Ya. Ya. Daube—carried out these measurements at frequencies from 50 to 20,000 hertz\(^6\).

To this section we also assign Bloch’s nuclear-induction method (1946), used in 1947 by K. V. Vladimirskii\(^7\) and improved in Moscow by S. D. Gvozdover\(^8\) with collaborators. It consists in the following: a strong magnetic field is applied to a dielectric, causing the magnetic axes of the atomic nuclei to precess. By applying a perpendicular alternating field of the same frequency as that at which the precession occurs, the latter is amplified and, consequently, the magnetic flux is changed. Its change is recorded in the receiver coil.

Thus, one can determine the relation between the applied constant field and the precession frequency, from which the magnetic characteristics of the atomic nucleus are calculated.

1. In the field of molecular-beam spectroscopy, an old phenomenon—the deflection of cathode rays in electric and magnetic fields—made it possible for the first time to study the electron and to determine the ratio of its charge to its mass (1897). The mass spectrograph (1919) is based on this technique. In this same section we find the Vavilov–Cherenkov effect (1934)—the appearance of light waves as a result of the motion in a medium of an electron with a velocity greater than the velocity of light in that medium. In the section of passive spectra of molecular beams we have placed electron diffraction and the electronography based on it.

In 1921 the first experiment was carried out measuring the deflection of silver atoms flying in a strong inhomogeneous constant magnetic field, which made it possible to determine the magnetic moment of atoms and to observe the spatial orientation of the magnetic axis of the atom along the field and against the field.

In 1937 Rabi and collaborators acted on molecules flying in a magnetic field with an alternating radio-frequency magnetic field whose period coincided with the period of precession of the magnetic moment of the nucleus of the given isotope. In such a case the projection of the nuclear moment changed, and the beam arrived at another place on the beam spectrum. This made it possible to measure the magnetic moments of the nuclei of a large number of atoms.

Here only an approximate outline is given of the systematization of general beam spectroscopy. The existing rich material must be classified according to deeper criteria and covered more broadly. Here it is presented only to the extent needed to show the connection between methods of radiospectroscopy that lie in entirely different areas of investigation (the diagram, the kinked arrow).

1. The basic theoretical concepts in the field of magnetic spectroscopy were given in our 1913 work,^9 where, for the first time, the theory of the electromagnetic field in a ferromagnetic metal was developed. The complex magnetic permeability following from this theory immediately became an indispensable attribute of all work in this field, as did the two types of apparent permeability, \(\mu_k\) and \(\mu_n\), or \(\mu_R\) and \(\mu_L\), introduced by us at that time. All the ideas on the propagation of waves in a polycrystalline substance, expressed then, were developed in subsequent works, both ours and foreign ones.

At that time, in particular, the hypothesis was put forward that dispersion of electromagnetic waves in a metal should arise if their wavelength in the metal is comparable with the sizes of its crystallites, now domains, and, consequently, with other structural inhomogeneities in the form of grooves and cracks. In 1930 E. I. Kondorskii confirmed this assumption, showing the significant influence of a longitudinal cut in a ferromagnetic conductor on its resistance to alternating current. In 1931 this hypothesis was again advanced by us, and the assumption was expressed that microscopic grooves and scratches on the surface of a metal may be the cause of magnetic dispersion of waves comparable with the depth and width of these grooves. I. M. Kirko, in his works carried out in Riga, verified this idea^10 on models of a ferromagnetic body with such grooves: he measured, in an alternating field, the permeability of rods, one half of which was covered with a screw thread. The measurements showed that the actually cut half of the rod exhibits magnetic dispersion if the depth of the grooves is comparable with the wavelength in the metal. In this case there is a drop in magnetic permeability and a maximum of magnetic absorption, as in a medium with magnetic viscosity.

To explain the magnetic transformation spectra described above upon magnetization of a material with magnetic viscosity, Kirko^6 developed a theory based on the hypothesis of the formation of eddy currents not only in domains, as Arkad’ev had assumed as early as 1918, but also in individual crystalline grains, whose dimensions may be considerably larger than those of domains. As a result of this phenomenon, magnetic viscosity may be observed at considerably lower frequencies than was calculated in 1935 by Arkad’ev, in 1938 by Becker, in 1941 by K. M. Polivanov, and in 1946 by Kittel.

References

1. V. K. Arkadiev, Magnetic Spectroscopy, Publishing House of the Scientific-Technical Department of the Supreme Council of National Economy, Moscow (1924).
2. G. S. Landsberg, Izv. AN SSSR, physical series 14, 174 (1950).
3. V. K. Arkadiev, Electromagnetic Processes in Metals, Part II, Moscow–Leningrad, p. 246 (1936); Vestn. MGU 12, 95 (1947).
4. V. K. Arkadiev, Contemporary Problems of Electromagnetism, Moscow–Leningrad, p. 55 (1931).
5. V. K. Arkadiev, ZhTF 13, 324 (1943); Journ. of Phys. 9, 5, 375 (1945).
6. I. M. Kirko, Proceedings of the Institute of Physics and Mathematics of the Academy of Sciences of the Latvian SSR, II issue 3, 9, Riga (1950).
7. K. V. Vladimirsky, DAN 58, 1625 (1947).
8. S. D. Gvozdover and A. A. Magazanik, ZhETF 20, 705 (1950).
9. V. K. Arkadiev, ZhRFKhO, physical section 45, 312 (1913); Phys. Zeits. 14, 928 (1913).
10. I. M. Kirko, DAN 59, 227 (1948).
11. Ya. G. Dorfman, Magnetic Properties of the Atomic Nucleus, Gostekhizdat (1948).