Modern Radio and Science*

N. D. Papaleksi

To speak at the present time about the role of science in technology is altogether trivial. In any branch of technology one can trace how the achievements of science influenced one or another stage in the development of that branch. However, this influence has been far from the same in different branches. There are such branches of technology as, for example, engineering, which arose together with human culture and developed slowly over millennia, relying chiefly on observations of natural phenomena and accumulated experience and using only natural materials. Although in many respects it thereby attained a high degree of perfection (one need only recall the splendid examples of architectural art in antiquity and the Middle Ages, the stone bridges of the Romans, etc.), nevertheless only in the last 100–150 years—and, moreover, only in connection with advances in science in the field of theory and with the obtaining, on the basis of scientific investigations, of new building materials—has engineering begun to advance very rapidly. There are also other major branches of technology which could arise and develop only when a sufficient body of knowledge in the given field had been accumulated and a theory had been developed. The most striking example of this type of branch of technology may be radio, which owes to science not only its origin but also all its vigorous development, thanks to which, in so short a period for a major branch of technology—barely fifty years of existence—it has acquired exceptional significance in the life of mankind, becoming indispensable in the most diverse fields of culture and of the national economy. Only after Maxwell’s theory of electromagnetic waves had been experimentally substantiated by Hertz’s brilliant experiments, and after the works of W. Thomson, Feddersen, Hertz, Branly, and many others had provided the first methods for generating, radiating, and receiving electric waves, could radio appear; in its creation and development such an outstanding role was played by physicists: A. S. Popov,

* A public lecture delivered on January 6, 1947, at the Moscow House of Scientists.

G. Marconi, F. Braun, M. Wien, L. I. Mandelstam, and others. And throughout all the further development of radio there runs this close and living connection, or rather interconnection, with physics, which led, in particular, to the emergence of an entire branch of physics—radiophysics. But this connection was beneficial not only for radio; as we shall see below, it also proved highly fruitful for science. Having grown and strengthened on a firm scientific foundation, radio engineering not only successfully solves its immediate tasks: radiotelegraphy, radiotelephony, the transmission of images; it not only performs exceptionally important cultural tasks through broadcasting and television; it not only makes it possible to solve the most diverse technical problems in the sphere of the national economy and defense, but in its turn also poses new problems for science, gives researchers new methods and instruments of research, and even opens up new fields of investigation.

The interconnection between radio and science has manifested itself especially vividly in recent times, when, on the one hand, it became quite obvious that the further progress of radio rests upon the successful solution of a number of scientific problems, and when, on the other hand, the development of radio opened up new possibilities for solving scientific problems from other branches of physics and even from other fields of knowledge. It is these scientific problems, on whose solution the further progress of radio depends, and also those whose solution has become possible thanks to the development of radio, that I should like to discuss today.

In order to form a clear picture of the paths of further progress in radio, and also of those scientific problems connected with the development of radio, it seems useful to recall its principal stages. As is well known, to carry out any kind of radio communication it is necessary:

1. First of all, to have a source of radio waves of the required range and power.
2. To transmit these radio waves through space.
3. To receive the desired radio transmission at the required place.

Thus in the field of radio proper—whether radio communication, broadcasting, television, radiolocation, or radio navigation—one can clearly distinguish three principal groups of problems:

1. Problems relating to the generation and radiation of radio waves.
2. Problems connected with the propagation of radio waves in space, and
3. Problems relating to radio reception.

Let us now see how, in the process of the development of radio, its interconnection with science was realized in each of these groups of problems. In the development of radio one can distinguish, with sufficient clarity, the following stages:

The first stage is the heroic epoch of radio, beginning with the experiments of A. S. Popov—the epoch of radiotelegraphy, at first with the aid of

crackling and sounding spark, and then the high-frequency machine and Poulsen arc generator. This was an epoch of gradual conquest of space: from tens and hundreds of meters in the first experiments to many thousands of kilometers across continents and oceans.

Already in this first epoch radio posed a number of major scientific problems. Thus, in connection with Marconi’s first transmission in 1901 of radio signals from England to America across the Atlantic Ocean, the problem of the propagation of radio waves around the Earth arose and immediately acquired great urgency. This major scientific problem, only in recent times, thanks to the investigations of numerous scientists throughout the world, including our own, has come substantially closer to its solution. The point is that radio waves cannot pass through the Earth, which is a conductor. On the other hand, it seemed highly doubtful that radio waves could go around the convexity of the terrestrial globe separating England from America. Here we have, in essence, a problem analogous to the optical problem of diffraction from a light source situated on the surface of a sphere, which presents enormous mathematical difficulties and at that time had not yet been solved. And so, to explain the fact of the transmission of radio waves across the Atlantic Ocean, various hypotheses began to be advanced. Thus, as early as 1902 Kennelly, and at the same time Heaviside, put forward the hypothesis of the existence of upper conducting layers of the atmosphere which, acting as a reflector, cause radio waves emitted upward by a radio transmitter to return back to the earth and thus enable them to overcome the curvature of the Earth. However, this correct hypothesis, because at that time there were no direct experimental confirmations of the existence of such reflecting layers of the atmosphere, for a long time did not meet with general recognition among radio specialists. This was also promoted by the “conception” of surface waves, advanced in 1907 by the well-known radio physicist J. Zenneck, which propagate along the surface of the Earth in the same way as waves of telegraphic transmission propagate over great distances along wires. From this notion of surface waves, which erroneously found theoretical confirmation in incorrect conclusions from Sommerfeld’s rigorous theory, it followed that the greater the wavelength, the better and farther radio waves could be transmitted along the surface of the Earth. Since these conclusions were also in agreement with the increase in range observed in practice as the wavelength increased, this determined the direction of the development of radio in the first epoch toward long waves and the associated tendency to increase the height of antennas.

From the very beginning, in order to increase the range, antennas were raised upward to a great height on balloons. By the end of the first epoch, i.e., after the First World War—in the early twenties—the wavelengths of transcontinental radio stations

were measured in tens of thousands of meters, antenna heights reached half a kilometer and more, and transmitter powers, both spark and machine types, reached many hundreds and even thousands of kilowatts.

Fig. 1.

Thus, the antenna of the super-powerful transmitting radio station in Malabar (on the island of Java), intended for communication with Holland over a distance of about 11,500 km, was suspended on cables fastened

Fig. 2.

to the summits of two mountains (Figs. 1 and 2), with the highest point of the antenna located at a height of 715 m above the ground. Also very indicative in this respect is the evolution of the antenna of the powerful radio station in Nauen (Fig. 3). It should also be noted that the first stage of radio development includes the first practical applications of electronics:

the Braun cathode-ray tube (1897), the progenitor of the now so indispensable electron oscilloscope, and also Fleming’s first diode detector (valve).

When the development of radiotelegraphy had reached a certain technical completion and the need for radiotelegraphic communication had to some extent been satisfied, technical thought naturally turned toward radiotelephony, and this led to undamped oscillations. At the first stage such oscillations were generated by high-frequency machines and by the electric arc; however, the technical solution of the problem of radiotelephony became real only

Fig. 3.

with the appearance of the three-electrode electron tube, which was the further remarkable outcome of scientific research in the field of thermoelectronic phenomena.

From its very first steps the electron tube, or, as it is called, the “radio tube,” won supremacy in the field of radio-signal amplification and radiotelephony; and only later, in connection with the appearance of the principle of feedback, did it soon become indispensable as an exceptionally convenient, flexible, and stable source of electrical oscillations of the most varied types—from purely sinusoidal oscillations (corresponding to monochromatic light) to rectangular and sawtooth waves, used in modern sweep circuits (for example, in radiolocators and in measuring technology).

However, at first in the field of radio communications, especially over long distances, where long waves continued to dominate, low-power tube transmitters could not compete with powerful spark, arc, and machine transmitters. The long-distance action of short waves, discovered by radio amateurs in 1921 and caused, as is now well known, by their reflection from the upper conducting layers of the atmosphere, led to the victory of tube transmitters and marked the beginning of the second stage in the life and history of radio, i.e. the beginning of the epoch of short waves (waves on the order of tens of meters) and of the development of tube radio engineering, as well as the epoch of space waves, i.e. waves arriving from the transmitting station not along the surface of the Earth, but through space, either directly or after reflection from the upper layers of the atmosphere. This immediately brought to the fore the experimental verification of the above-mentioned hypothesis of Kennelly and Heaviside, which was brilliantly confirmed in 1925–1926 by Appleton (by the interference method) and by Breit and Tuve—by the “radio echo” method. Thus there arose a new scientific problem: the investigation of the upper, radio-wave-reflecting layers of the atmosphere, forming the so-called “ionosphere” and having very great importance not only for long-distance radio communications, but also for geophysics and solar physics. The radio methods developed for this purpose for studying the properties of the ionosphere, in particular for determining its effective height by means of radio pulses, i.e. short groups of radio waves with a duration of \(10^{-4}\) seconds and shorter, as is known, formed the basis of radiolocation.

The need to maintain the constancy of the frequency of transmitted radio waves, especially short ones, with very high precision, so that there would be no interference in reception from stations close in frequency, led to the important problem of frequency stabilization. To solve this problem, the piezoelectric properties of crystals (quartz and tourmaline), discovered and studied by the famous physicist Pierre Curie, were used (Cady, 1922).

The second epoch in the history of radio, which was also the epoch of the development of broadcasting and radiotelephony, after these technical problems had been largely solved in connection with the development of electronics, naturally led to a new technical task—television, for the realization of which it was necessary to pass to still shorter waves of the meter range, which received the name “ultrashort” waves. Thus the third stage of the development of radio was outlined—the epoch of ultrashort waves and television. However, into this epoch, which even at the present time has not yet reached its full flowering, in connection with the tasks posed by the Second World War there intruded the fourth stage—the epoch of mastering microradio waves (i.e. waves of the decimeter and centimeter ranges) and pulse radio engineering. The latter proved necessary for solving the problems of reflect—

of radar, i.e. the determination of the position of non-radiating objects by means of sharply directed beams of radio rays.

From this brief survey it is evident that radio’s striving to solve practical problems conditioned, in the first periods of its development, a tendency toward increasing the wavelength. Later, as experimental facts concerning the propagation of radio waves through space accumulated, and also in connection with the emergence of new problems, a tendency toward an ever greater shortening of the wavelength gradually became apparent, leading to the shortest radio waves, adjacent to light waves. In this connection radio engineering began to borrow from optics the methods and means it had developed. Such an “opticalization” of radio undoubtedly left its imprint on its present state and development, signifying also a practical convergence of radio and optics.

It goes without saying that the development of new directions in radio engineering and the mastery of new frequency ranges, brought about by the need to solve new technical problems, by no means signifies the abolition or consigning to the archives of its former achievements and of the frequency ranges mastered earlier. Nor does this mean that many questions of radio communication that arose earlier and are of great importance for satisfying cultural needs, the national economy, and defense have already been fully solved. Such, for example, is above all the problem of uninterrupted and reliable long-distance radio communication, for which even now the role of short, and partly also of long, waves remains predominant. It is also necessary to note that already long ago—approximately since the First World War—radio engineering gradually began to go beyond the limits of radio communication proper. Thus, radio gradually acquired exceptionally great importance in navigation—maritime and aeronautical. From the possibility of determining the direction from which waves arrive there arose radio direction finding, which at first operated effectively at comparatively short distances; radio beacons also appeared; and in recent times more advanced methods of short- and long-distance radionavigation have been developed—interference and pulse methods; radio geodesy has also come into being. Radio over wires, in a certain sense, revolutionized wire communication: on the one hand, thanks to the use of vacuum-tube amplifiers, the range of ordinary telephony increased very greatly, and, on the other hand, radio over wires made possible multiple transmission of telephony over a single wire, as well as phototelegraphy. At present there already exists telephone communication over wires on ultrashort waves, allowing up to 400 conversations to be conducted simultaneously over a single cable. Radio engineering has invaded, on a broad front, other branches of the national economy as well. Induction and dielectric heating with the aid of high-frequency currents has acquired enormous significance and wide application in industry: in metallurgy and in the technology of metal processing—for melting, refining, surfac-

surface hardening, welding, etc.; in the woodworking industry, for example, for drying and gluing plywood; in the food industry—for heating and sterilization; in medicine—diathermy, d’Arsonval currents, and so on. Some idea of the scope and role of high-frequency technology, which grew out of radio engineering, may perhaps be gained from the fact that at present the power of high-frequency generators used in industry exceeds by several times the power of all radio stations for communication and broadcasting.

It is also necessary to point out the enormous importance that radio methods have acquired in science as an irreplaceable instrument of research in its most varied branches: physics, biology, astronomy, etc. Exceptional possibilities for research have opened up thanks to electronic amplifiers, which make it possible to amplify electric currents and voltages tens of millions of times. The introduction of quartz frequency stabilizers made it possible to create high-precision frequency standards and the so-called “quartz clocks,” whose accuracy reaches \(10^{-9}\). Such clocks are at present an indispensable accessory of large observatories. The regular transmission by radio of standard frequencies and time signals makes it possible in all corners of the globe to check the frequencies of local standards, and also to obtain the exact time. The latter is of especially great importance for astronomical, gravimetric, geodetic, and hydrographic measurements, as well as for navigation.

Radio engineering is also of great importance for the creation of cyclotrons and other electron accelerators, which play such an important role in nuclear physics. It must be emphasized that not only has electronics developed to a considerable extent thanks to radio engineering, but many of its achievements, which have entered the most varied fields of science and technology—such as, for example, the most diverse applications of photoelectric cells—would not have been possible without radio engineering.

As we see, the particularly rapid development now taking place in radio engineering and in the electronics and other related fields connected with it has become possible only thanks to the rational use of the achievements of the physical sciences. In turn, radio has posed before physics a number of problems whose clarification has not only contributed to the solution of technical tasks both in radio and in other branches of technology, but has also enriched the corresponding branches of science: the theory of oscillations, in particular nonlinear ones; the theory and investigation of questions of the radiation of electromagnetic waves, as well as their propagation and interference; electronics—theoretical and experimental—and, in particular, questions of electron emission from pure metals and complex cathodes (oxide cathodes), as well as questions of vacuum; the study of fluctuation phenomena, so important not only for questions of radio reception but also for statistical physics.

Many scientific problems posed by radio in the course of its development have been successfully solved; others still await their final solution. On the other hand, new problems have taken shape, on whose solution the further progress of radio depends; and new fields of science have also become apparent in which radio, apparently, is destined

Fig. 4.

to play a very important role, not only in its present state, but even more so in its further development.

What, then, are these problems? In accordance with the main aspects of the process of radio transmission as a whole mentioned above, we shall begin with the problems relating to the generation and radiation of radio waves. As we have seen, at the present stage of radio development there is a definite tendency toward a transition to ever shorter waves, which is due mainly to two circumstances: first, microwaves (centimeter, millimeter, and shorter) can be con-

concentrated into very narrow beams of enormous power, for example, by means of parabolic mirrors such as are used in optics, or by specific radio-engineering methods connected with the possibility of creating systems of coherent radiators. Such, for example, are plane or spatial radiating systems (Fig. 4), consisting of regularly arranged tuned vibrators, as well as slot systems or the so-called “diffraction” antennas (Fig. 5), etc.

Fig. 5.

On the other hand, radio waves of these frequencies can, analogously to sound waves, be directed through hollow tubes, which have received the name “waveguides,” or through a concentric cable. Through waveguides it is possible to transmit microradio waves over great distances with very small losses. A concentric cable, as was already mentioned above, is used, in particular, for multiplex telephony on microradio waves.

The possibility of concentrating microradio waves into short pulses of great power, as is known, ensured the striking successes of radiolocation in the Second World War with the aid of the so-called “radars.” Since further progress in the mastery

the microwave range opens new prospects for various applications of radio not only in questions of radio communication, broadcasting, radar, and television, but also in other fields of technology and especially in science—physics, astronomy, chemistry, biology, about which we shall speak in more detail below, then it is clear that the problem of generating powerful undamped oscillations of possibly higher frequencies (down to fractions of a millimeter and shorter) is one of the most urgent scientific and scientific-technical problems.

The task of obtaining large instantaneous powers on centimeter waves has already advanced considerably at the present time. According to American journals, recently on these waves a power has been achieved in pulses of duration \(10^{-6}—10^{-7}\) seconds up to several thousand kilowatts. Significant results have also been obtained with respect to continuous powers. Thus, on a wave of 50 cm in the so-called resnatron, powers up to 100 kW have been obtained. On shorter waves the situation is still considerably worse.

The problem of generating millimeter and shorter waves can be approached in various ways. On the one hand, since radio waves of the order of fractions of a millimeter already approach the infrared waves contained in the spectrum of an incandescent body, it is conceivable to try to obtain the required microwaves by means of thermal radiation. However, as calculation shows, although an absolutely black body at a temperature of \(6000^\circ\ \mathrm{K}\), corresponding to the effective temperature of the Sun, radiates from one square centimeter about seven and a half kilowatts of radiant power, only \(10^{-14}\) of all the radiation, i.e. \(7\cdot10^{-4}\) erg, falls on microwave radiation with a wavelength of the order of 1 mm or, more precisely, on the portion of the spectrum in this region with an extent of \(10^6\) cycles. With increasing temperature the output of these waves of course also increases; however, the efficiency is so negligible that the method of obtaining microwaves by temperature radiation can hardly be considered at all promising. In this connection it is interesting to note that in recent years various investigators have discovered that the Sun actually emits radio waves, and in the region of centimeter waves they give at the Earth a power flux of the order of \(4\cdot10^{-15}\) W/m\(^2\) per 1 MHz for \(\lambda=10\) cm and, for \(\lambda=1\) cm, \(4\cdot10^{-13}\) W/m\(^2\) per 1 MHz.

In addition to the temperature radiation of an absolutely black body, which gives a continuous spectrum, it appears possible, for obtaining microwaves, to use the temperature radiation of gases, for example, of a high-pressure mercury lamp, which gives separate spectral lines. According to Rubens’s observations in 1921, mercury molecules in this case radiate waves of length 0.218 and 0.343 mm, i.e. waves already passing into microwaves. In contrast to the lines of the visible spectrum, which are caused, roughly speaking, by oscillations of electrons in the atom or of atoms in the molecule, infrared radiation is due to the rotation of molecules, which is why the resulting

In this case the spectra are called rotational. Besides mercury molecules, for obtaining waves in the microwave region it is possible to use the rotational spectra of molecules of other substances (as we shall see below, such spectra are also given by NH₃, H₂O, O₂, etc.); however, in all cases here one is dealing with very small powers. Other possibilities for generating microwaves, analogous to the cases of nonthermal radiation known in optics, are also conceivable: for example, excitation of the luminescence of gases by electron impact, widely used in gas-discharge lamps; bremsstrahlung, used in X-ray tubes, and so on. At present, however, no real ways are yet apparent for obtaining, by all these methods, any appreciable powers.

It may not be without interest, in this connection, to note that as early as 1931 the American radio specialist Jansky discovered radio emission from the region of the Galaxy and stellar nebulae at a wavelength of 14.6 m; subsequently, especially quite recently, this emission was discovered and investigated at other wavelengths by other scientists as well (Reber, Hey, and others). This radiation, apparently, owes its origin to bremsstrahlung arising from collisions of electrons with molecules in the cosmic plasma of interstellar space.

Another possible and more promising path is the tried radio-physical path, along which the development of methods for generating microwaves has proceeded up to the present. In contrast to disordered thermal radiation, here, by means of an externally controlled (or auto-) electron stream, organized, ordered radiation is obtained by excitation of waves in the cavities of volume resonators, caused by the flight of groups or “bunches” of electrons past them, just as the condensation or rarefaction of a jet of air excites sound in a hollow space—an acoustic resonator. In the region of centimeter waves, as was already indicated above, quite significant results have been achieved with “klystrons” and “magnetrons” operating on this principle; however, with further advance toward ever shorter waves this method encounters ever greater difficulties, and moreover not only of a technical nature, which could be overcome by rational design. The point is that with decreasing wavelength the dimensions of the resonators decrease, and this leads not only to mechanical difficulties connected with increasing requirements as to the precision of manufacture of these resonators, but also to an increase in the relative losses on their surface, i.e., to a decrease in what is called in radio engineering their “quality factor.” With the decrease in the dimensions of the resonators there also increases the danger of overheating, which sets a limit to the obtainable continuous power, while the danger of breakdown limits the voltage and, consequently, the pulse power.

The production of microwaves by upward frequency transformation—for example, by distorting the shape of the current curve in a nonlinear resistance (detector) with subsequent selection of the corresponding overtone—encounters the same difficulties, since resonators are also needed for selecting the overtone. Therefore, in recent times research thought has naturally begun to seek new ways of solving the problem of microwave generation. I cannot dwell here on a consideration of various such ways; I shall allow myself only, by way of example, to cite one of the principally possible new methods, which was recently indicated by our young theoretical physicist V. L. Ginzburg. This method is based on the use of electrons obtained in electron accelerators (betatrons, synchrotrons, etc.), whose velocities are close to the speed of light—so-called “relativistic” electrons. The principle of the method is as follows (Fig. 6): if a source of electromagnetic waves of frequency

Fig. 6.

\(f_0\), for example, an electron oscillating with frequency \(f_0\), moves rectilinearly in a direction perpendicular to the direction of its oscillations with velocity \(v\), then, as is known, in a direction making an angle \(\theta\) with the direction of motion, the frequency of the arriving waves will, according to the Doppler principle, be equal to:

\[ f=\frac{f_0}{1-\beta\cos\theta}, \]

where \(\beta=\dfrac{v}{c}\), and \(c\) is the speed of light. In the case of a relativistic electron, i.e. with \(\beta\) close to unity and very small \(\theta\), the denominator will be very small, and consequently, within a solid angle with very small \(\theta\), we shall have radiation with upward frequency transformation by a very large factor. Such an emitter can be realized by forcing a relativistic electron to oscillate in a direction perpendicular to its motion with frequency \(f_0\), namely by passing it between the plates of a capacitor to which an alternating voltage of frequency \(f_0\) is applied. In principle, it is possible here to achieve very large coefficients of upward frequency transformation. Thus, for example, at an electron velocity of 5 MeV, i.e. the velocity which it acquires in traversing a potential difference of 5 million volts, quite attainable in modern electron accelerators, the frequency in the direction of motion is increased 200 times, which, with an initial wavelength equal to 20 cm, will give radiation with a wavelength of 1 mm.

Of course, the radiation energy of a single electron is negligibly small; however, if there is a “bunch” of electrons, i.e. a group consisting of a large number \(N\) of electrons \((10^{-9}—10^{10})\), concentrated in a volume whose dimensions are smaller than the emitted wavelength, and executing one and the same oscillatory motion, i.e. mutually coherent, then their total power will be greater than the power of one electron by \(N\) times, i.e. it may already be considerable. In the creation of such powerful “bunches” of electrons lies, apparently, the chief practical difficulty in realizing this and analogous methods of obtaining powerful micro-radio waves. The example cited shows what possibilities still exist in reserve in modern physics in connection with the development of electronics and quantum physics.

Closely connected with the problem of generating micro-radio waves is the problem of stabilizing their frequency. If in the region of meter and even decimeter waves the constancy of frequency can be achieved with an accuracy of \(10^{-7}—10^{-8}\) by extracting a high harmonic of a lower-frequency generator stabilized by quartz or tourmaline, then for millimeter and still shorter waves such stability has not yet been attainable by this method (because of the difficulty of extracting very high harmonics and the accompanying destabilization). The gradual merging of radio with optics as the wavelength is shortened naturally leads to the idea of using, for the solution of this important task of obtaining a frequency-stable sinusoidal radio oscillation (or, to put it “optically,” monochromatic radiation), methods known from optics. As was indicated by V. L. Ginzburg, it seems possible to use for this purpose the absorption lines of the rotational spectra of molecules, for example ammonia—\(\mathrm{NH}_3\)—and others.

I shall not now dwell on a number of questions connected with the channeling, concentration, and radiation of micro-radio waves, which are important for their practical exploitation. I shall only point out that a peculiar “radio-optics” is developing, based, on the one hand, on the quasi-optical properties of micro-radio waves and, on the other hand, on the peculiarities of their propagation in hollow tubes—“waveguides”—and in dielectric rods.

I now turn to the second group of problems, namely to the problem of the propagation of radio waves in the atmosphere and other media. This problem, being, as we have seen, one of the oldest scientific problems of radio, continues to retain its important significance, since as radio develops ever new tasks arise. Despite the fact that, as a result of numerous works by major physicists, mathematicians, and radio specialists—Poincaré, Sommerfeld, Watson, van der Pol, Weyl, Appleton, Eckersley, and others—among which the works of our scientists V. A. Fock, L. I. Mandelstam, M. A. Leontovich, B. A. Vvedensky, E. L. Feinberg occupy a prominent place, the basic facts in the physics of the process of propaga-

of the propagation of radio waves along the Earth’s surface, even now it cannot be considered that all the questions belonging here have received sufficiently complete scientific elucidation. Thus, in particular, although the question of the propagation of radio waves around the terrestrial globe—taken as a sphere of finite conductivity, surrounded by a normal atmosphere whose pressure varies according to the barometric formula—may be regarded as finally solved by the recent works of Academician V. A. Fock and Academician M. A. Leontovich, nevertheless the problem, as it exists in reality, is still very far from being completely resolved. The point is that the variable ionization of the upper layers of the atmosphere and the presence of the Earth’s magnetic field make the problem of the propagation of long and short waves extremely intricate and difficult. As for the problem of the propagation of ultrashort and microwaves, especially beyond the limits of direct visibility, it is greatly complicated and impeded both by considerable fluctuations in the content of water vapor in the air and by temperature inversions.

As part of the general problem of propagation, the problem of the ionosphere—experimental investigations of which have for many years been carried on continuously and regularly in many countries—now requires especially great attention, since we are approaching the maximum of solar activity, which promises to be particularly intense in the coming years. It is not excluded that, along with tests of radiotechnical methods of ionospheric sounding, observations of the Sun’s radio emission will prove important for the task of radio forecasts; we shall speak of this in more detail below.

In the region of ultrashort and microwaves, the investigation of questions of refraction, scattering, and absorption of them in the troposphere as a function of meteorological conditions is of great importance. This problem, as we shall see below, is also of great significance for meteorology.

Specific demands on the problem of radio-wave propagation are now being made by radio navigation, radiolocation, and especially radiogeodesy, which have developed rapidly in recent years and which are based on knowledge of the exact value of the velocity (group or phase) of radio-wave propagation under actual conditions. It is sufficient, perhaps, to point out that in order to determine, with the aid of radio waves, a distance of 150 km with an accuracy of 5 m, it is necessary to know the value of the propagation velocity under the given conditions with an accuracy of 1/30,000, i.e. 8–10 times more accurately than at present. To estimate the difficulty of satisfying this requirement, it must be borne in mind that although the velocity of light in vacuum, according to Michelson’s most precise measurements, is known with an accuracy of about \(10^{-5}\), the correction for air and its humidity amounts to about \(4 \cdot 10^{-4}\) of the velocity, i.e. exceeds by a factor of 12 the permissible error in measuring the distance. This means that in very precise measurements of distances by means of radio waves it is necessary

take into account changes in humidity and barometric pressure along the path of the wave even in propagation in free space, when the influence of the earth’s surface may be neglected. The use in geodesy, for measuring distances, of radio-interference methods analogous to optical ones—for example, the classical Michelson method for the precise measurement of lengths, which in the microwave region can in principle give very high accuracies—will require still more subtle investigations and still more exact allowance for the influence of various natural factors on the conditions and magnitude of the propagation velocity of these radio waves.

Among the problems belonging to the third group, perhaps the least clear, in the sense of possible ways of solving them, is the problem of amplification of microwaves. If in the region of long, short, and even ultrashort (meter-range) waves, by the use of electronic amplifier tubes and the principle of resonant amplification, it is possible to solve completely the problem of direct (high-frequency) amplification, then in the microwave region (from a few centimeters and shorter) we encounter the same difficulties as in the generation of microwaves; the picture here is very unclear, and as yet no definite possible ways of solving this important problem are visible. It is possible that here certain devices from the field of electron optics, analogous to known optical methods, may lead to the goal.

It must also be emphasized that in the microwave region neither the most sensitive optical method of detecting and measuring electromagnetic radiation—namely, the subjective one (vision), for quite understandable reasons—nor the photoelectric method can be used, because as yet there are no photocells that respond to such small quanta. (It should be noted, however, that at present there already exist photocells that respond to infrared radiation with a wavelength of \(1.4 \cdot 10^{-3}\) mm.) Nor can the very sensitive and universal photographic method be applied. Nor, because of their inertia, can the heat-sensitive thermal methods used in optics for detecting thermal radiation—for example, by means of a bolometer or thermoelement—be used for receiving rapidly varying signals.

Of very great importance for the reception of microwaves are also the questions of frequency stabilization. The point is that the most widely used and, properly speaking, the only sufficiently sensitive method of receiving microwaves at the present time is the heterodyne method, i.e., the simultaneous application to a crystal detector both of the incoming microwaves and of oscillations from a local microwave source of approximately the same frequency. As a result of such heterodyning there is obtained, lower than the received one, a difference frequency of the order of the frequencies of ultrashort or short waves, which

is then amplified by ordinary methods. Thus, particular importance is acquired here by the stability of the frequency both of the received microwaves and of the local heterodyne. In fact, in receiving, for example, waves of length \(3\ \mathrm{mm}\), i.e. with a frequency of \(10^{11}\ \mathrm{cps}\), and with a stability of the order of \(10^{-5}\), which can hardly be surpassed at present for these waves, the oscillations of the difference frequency may reach \(2\cdot 10^{-6}\ \mathrm{cps}\), and this will require from the receiving device a pass band of such width as would allow oscillations of not less than \(2\cdot 10^{-6}\ \mathrm{cps}\). Such fluctuations in the width of the pass band are not yet a great evil in receiving very short pulses of duration of the order of several \(10^{-7}\) seconds, as is the case in radiolocation, where the band width reaches \(5\cdot 10^{-6}\), but for highly selective reception of continuous oscillations or radiotelephony it is very undesirable. The point is that, as is well known, with the widening of the band of frequencies passed by the receiver, the level of interference in it increases. Although in the microwave region there are practically neither atmospheric interferences, so harmful for the reception of long and short waves, nor industrial interferences, which also hinder the reception of ultrashort waves, here, however, interference from thermal radiation makes itself fully felt. It is mainly these, together with the fluctuation noises of the receiver itself, that limit the sensitivity of the microwave receiver.

In order to form an idea of the magnitudes involved here, let us give the following figures: if for \(\lambda = 3\ \mathrm{m}\), with an attainable stability of \(10^{-7}\), it is possible to narrow the receiver pass band to \(100\ \mathrm{cps}\), as a result of which the noise power can be brought down to \(4\cdot 10^{-18}\ \mathrm{W}\), then for \(\lambda = 3\ \mathrm{mm}\) and a stability of \(10^{-5}\) the pass band will have to be taken of the order of \(10^{7}\ \mathrm{cps}\), which, taking \(n = 50\) (\(n\) being the noise factor), will give a noise power of \(2\cdot 10^{-12}\ \mathrm{W}\). For comparison, let us indicate that the sensitivity of the rested eye for \(\lambda = 510\ \mathrm{m\mu}\) is \(4.4\cdot 10^{-17}\ \mathrm{W}\), while the sensitivity of photoelements reaches \(3\cdot 10^{-14}\ \mathrm{W}\).

Thus, one of the urgent and by no means minor problems of microwave reception is the problem of combating the inherent noises of the receiving apparatus.

It is also necessary to note that experimental investigations in the microwave region, especially as the waves are shortened, are to an ever greater degree hampered by the absence of accurate and sensitive methods of measurement in this region, lying at the junction between radio and optics. As is known, in radio engineering the direct objects of measurement are currents and voltages. The field strength of radio waves is measured not directly, but through the electromotive force induced by this field either in a straight-line conductor or in a frame of definite dimensions (radio comparator). The remaining energy quantities, such as the flux of power of radio waves, which are of great importance for characterizing

process are computed from these measurements. With decreasing wavelength, the concept of the strength of the current or the voltage at a given point of the conductor gradually loses its meaning, and it becomes practically impossible to measure these quantities without substantially altering the object of the process being measured. Therefore, as objects of direct measurement, there gradually come to the fore—as in optics—quantities characterizing the radiation field, such as: radiant power flux, radiation intensity, the intensity of the electric and magnetic fields of the wave, the character of the wave polarization, and the coefficients of absorption and reflection. For this kind of measurement in the region of microwaves, thermal instruments are successfully used: bolometers and thermoelements, which, because of their inertia, as was indicated above, are not suitable for receiving these waves. It is quite obvious that the development of precise and sensitive methods of measurement in the microwave region is an extremely important scientific and technical task of the present time.

I now turn to the last part of my rather lengthy report. In our brief historical survey of the principal stages in the development of radio, we have repeatedly had occasion to be convinced that radio not only often posed new scientific problems for physics and, for its part, gave science new methods and instruments of research, but also led to new discoveries in various fields of science. Thus, the attempt to explain the very fact of the reception in America of radio signals from England first led Kennelly and Heaviside to the scientific geophysical hypothesis of the existence of upper conducting layers of the atmosphere, reflecting radio signals and forming the so-called “ionosphere”; and the development of long-wave and short-wave radio engineering made it possible to prove experimentally the existence of these reflecting layers, and then to develop an excellent method for investigating the properties of the upper layers of the atmosphere, which are directly inaccessible to us.

The method of radio sounding of the ionosphere, as is known, helped to elucidate the process of propagation of radio waves over great distances around the globe and made it possible to communicate with the antipodes. It also became an exceptionally flexible instrument for investigating the physical properties of the ionosphere. With the aid of this method, the role of the Sun’s ultraviolet radiation in the ionization of the upper layers of the atmosphere was established, and observations during solar eclipses, especially total ones, make it possible to obtain valuable indications concerning the distribution of active centers of radiation on the Sun itself. Thus, radio methods of investigating the ionosphere are also a very valuable instrument for investigating the nature of processes on the Sun and their character.

The significance of radio methods for geophysics is not limited only to questions concerning the properties of the ionosphere. Radio also provides great assistance to meteorology in the investigation of the troposphere. Not to mention the practically extremely important and at present

a broad use of radiosondes, where the valuable role of radio consists in the automatic transmission of readings from instruments recording temperature, pressure, humidity, and other meteorological elements at various altitudes during ascent in a balloon, it is apparently microwaves that are destined to play an essential role in meteorology.

The point is that already at the very beginning of the use of meter-range radio waves it was established that their range often extends far beyond the limits of direct visibility. A thorough theoretical and experimental investigation of this phenomenon, carried out in recent years, has shown that it is substantially connected with meteorological conditions. It turned out that in cloudy and very windy weather the range of meter and centimeter waves exceeds the distance of direct visibility only very slightly, whereas under other meteorological conditions their range sometimes surpasses this distance many times over; moreover, this effect is the stronger the shorter the wave. Thus, according to observations made in England in July 1941, the range of 10-centimeter waves extended to 270 km, whereas the geometrical distance of direct visibility was less than 64 km. An even more striking long-range action of a radar installation at a wavelength of 1.5 m was observed near Bombay: during the monsoons it extended only to 32 km, whereas in the hot season the radius of action reached 320 km; once it reached 1000 km, and there were not infrequently cases when the signals reached the shores of Arabia at a distance of 1600—2000 km. As theory shows, such anomalous propagation phenomena are connected with temperature inversion, which causes anomalous curvature of radio rays, intensified still further by the decrease in the pressure of water vapor with heating. Here we have a phenomenon analogous to the so-called “optical mirage,” well known to travelers in the desert. However, the dependence of microwave propagation on meteorological factors is not limited to this. According to recently published results of observations by English researchers, the presence in the atmosphere of suspended droplets of water, crystals of ice, and even dust particles has a noticeable influence on the attenuation and scattering of centimeter-range waves. Thus, according to observations, the attenuation of radio waves of 3 cm in passing through rain clouds and falling rain of moderate intensity reached 1 db per km in Britain, and in heavy tropical downpours reached up to 6 db per km. A noticeable reflection of these waves from rain clouds has also been established: in one case, at a wavelength \(\lambda = 10\) cm, the reflection from a rain cloud from which moderate rain was falling (the precipitation amounted to 7 mm/hour) was of the same order as from a small airplane, and at a wavelength \(\lambda = 3\) cm it was 100 times stronger. In addition, selective absorption of millimeter waves in water vapor and oxygen was also discovered. Such a բազմofo—

different dependence of the conditions of propagation of microwaves on meteorological factors undoubtedly opens new paths for meteorological investigations.

The phenomenon of selective absorption of microwaves in water vapor and oxygen is not exceptional. As was already noted above, absorption or emission lines corresponding to the rotational spectra of molecules fall precisely in the microwave region; of these, the spectrum of ammonia, NH$_3$, lying in the region $\lambda = 1.25$ cm, has so far been studied best. As can be seen from Fig. 7, this spectrum consists

Fig. 7.

of very fine lines, which can be used as precise frequency standards. On the other hand, the detection of these spectra and their study are of very great importance for understanding the structure of molecules, and also for establishing the composition of a given substance. Thus microwaves open up a new, very promising field of investigation—radiospectroscopy of molecules.

Microradiospectroscopy can also be used to study the properties of dielectrics, for a large number of which the absorption maximum falls precisely in the microwave region. Such a study of the absorption spectra of dielectrics should prove essential for understanding their structure. This is especially important for studying the structure of synthetic resins and rubbers. In connection with this, it seems to me important to bear in mind that investigation of living tissues by means of microradiospectroscopy will possibly make it possible to use microwaves effectively also for diathermy, and, with the application of pulse technique, perhaps for transillumination and other therapeutic purposes.

For lack of time I have no opportunity to dwell on other areas of physics that have been enriched thanks to the application of ra-

...methods of investigation, for example, of the domain of magnetic phenomena or of the above-mentioned accelerators of charged particles, etc. I turn to the prospects that are opening up in geodesy and hydrography thanks to radio. It is well known to everyone that at present in hydrography, for measuring depths, the so-called “echo sounders” are widely used, based on the reflection of ultrasound from the sea floor. The echo-sounding method, which owes its existence to radio, has produced a genuine revolution in hydrography, since it makes it possible quickly and accurately to determine individual depths, and also to keep a continuous record of depths and thus to give the relief of the bottom. Hydrographic surveys, as is known, consist of measuring the depth at a given point and simultaneously determining the coordinates of this point. If before the appearance of the echo sounder the chief difficulty lay in the impossibility of making measurements of great depths quickly and accurately, then at present the center of gravity has shifted toward the determination of coordinates, which becomes especially difficult when the vessel is removed from the shore beyond the limits of direct visibility. Moreover, for the usual geodetic (optical) methods of determining coordinates, good visibility is required, which greatly limits the time during which measurements can be made. The problems of accurately and rapidly determining the position of a vessel beyond the limits of direct visibility, and, moreover, in any weather, are successfully solved by the application of radio-interference methods of distance measurement developed here in the Union. These methods can also be successfully applied to determining coordinates in aerial photography. Further development of radio-interference methods, with a transition to shorter waves and the attainment of a more accurate value of the velocity of propagation of radio waves, also opens up new possibilities for solving certain problems of higher geodesy. I have in mind the problem of tying together various triangulation systems across large bodies of water that do not permit the use of the usual optical methods of determining distances, as, for example, across the Bering Strait or the Mediterranean Sea. In addition, the transition to decimeter and centimeter waves makes it possible to use radio-interference methods for the direct determination, with great accuracy, of the initial baselines of triangulation networks, which is especially important for geodetic surveys of uninhabited regions.

Exceptionally broad prospects are opened up by the application of radio methods in astronomy and astrophysics. We have already mentioned the transmission by radio of precise time signals, which is very important for astronomy, and quartz clocks. The development of the technique of short and ultrashort waves, in particular radiolocation, has provided new, highly powerful means of investigation and has led to a number of extremely interesting and important discoveries in the field of astrophysics. I shall begin with the application of radiolocation methods to observations of meteors, which makes it possible not only to note the appearance of meteors with extraordinary ease and simplicity, but also to determine the height at which...

they produce noticeable ionization effects. As you know from press reports, such observations, carried out during the meteor shower in the morning hours of November 10, 1946, yielded very interesting results. I have emphasized “morning hours” because, with optical methods of observing meteors, observations can be made only on a clear night, whereas radiomethods can be used both at night and by day, regardless of the state of the atmosphere.

Of exceptionally great interest for the theory of the Moon and its motion, as well as for the study of its surface, is the reflection of radio pulses from the Moon, carried out about a year ago, in January 1946. Leaving aside the sporting significance of the very fact of radio transmission to the Moon and back, i.e. over a distance of about \(3/4\) million km, which is in itself a sufficiently eloquent indication of the progress of radio in recent years, it is difficult to overestimate its scientific importance. The fact is that the high astronomical accuracy now attained in determining the distance to the Moon is obtained as a result of numerous observations requiring favorable atmospheric conditions and involving measurements from two points situated approximately on the same meridian at an enormous distance from one another, as well as complex calculations and the allowance for numerous corrections. The use of the radar method for this purpose makes it possible to measure directly the distance to the Moon from a single point; moreover, the accuracy of a single measurement exceeds the mean astronomical accuracy by an order of magnitude. In addition, this method, in principle, makes it possible to follow continuously the change in the distance to the Moon over time, which is of special value for the theory of the Moon. It should also be noted that the dependence of the intensity and form of the reflected radio pulse on the properties of the reflecting surface can be used to study the structure of the Moon’s surface. At present it is still very difficult to say what new prospects for the study of the Moon will open up with the further development of the radar method as applied to measurements of the distance to the Moon.

The application of radiomethods in astronomy and astrophysics is not limited to the study of the Moon and observations of meteors. The point is that a number of processes on the Sun and in outer space are accompanied by phenomena which, by their nature, could be detected only by radiomethods. Fifteen years ago Jansky, while studying atmospheric radio interference, discovered, at a wavelength of \(14.6\) m, along with radio interference of terrestrial origin—natural and industrial—also radio interference of extraterrestrial origin, coming from sources located far beyond the limits of the Earth’s atmosphere. According to his measurements, the maximum of this interference, which was given the name “cosmic interference,” came from the direction of the constellation Sagittarius, i.e. approximately from the center of the Galaxy. These

the results were confirmed by Reber’s observations in 1940 and 1944 at a wavelength of 1.87 m. In addition to the principal maximum in the direction of the constellation Sagittarius, he observed still less intense maxima from the regions of Cygnus, Cassiopeia, Canis Major, Puppis, and others. These observations, which became possible solely thanks to the enormous achievements of ultrashort-wave radio engineering, aroused extraordinarily great interest on the part of astrophysics, since they open new prospects for the study of processes in cosmic space.

Alongside radio emission from the region of the Galaxy, Reber also discovered radio emission from the Sun at a wavelength of 1.87 m. Further observations by Southworth at wavelengths from 1 to 10 cm, as well as observations on radar installations made during the war in England and Australia in the 4–6 m range, detected solar radiation at these wavelengths as well. It was also noted that the radio emission of the Sun increases unusually—by hundreds of thousands of times—during the passage of sunspots, and it was established that the spots themselves are the sources of this powerful radio emission (Ryle and Vonberg). Analysis of the character of this radiation showed that it is to a considerable degree circularly polarized (Martin, Appleton). From the numerous recently published works devoted to observations of the Sun’s radio emission, their enormous significance for the study of processes on the Sun and of the nature of the solar corona follows beyond doubt. This new field of research, which is still in its infancy, undoubtedly represents an extraordinary interest for solar physics. There is every reason to think that with the application of radio methods to astronomy a new era will open, one whose significance may be compared with the discovery of the Fraunhofer lines and the application of spectroscopy in astrophysics, and which will help us penetrate still more deeply into the mysteries of the universe.

It goes without saying that the scientific disciplines I have mentioned by no means exhaust the prospects that modern radio opens up for science. There is no doubt that with the further development of radio, still new and perhaps broader possibilities for scientific research will open up. But it seems to me that from the little I have set forth here, it is quite evident what astonishing results are produced by the close connection between science and technology in general, and between radio and science in particular.