MODERN ALCHEMY¹

Lord Rutherford

In this lecture I intend to give a brief account of modern research in the field of the transmutation of elements. The title is meant to emphasize the contrast between these investigations and the ancient alchemy which, for almost two millennia, attracted human minds with such exceptional force. Belief in the possibility of the transmutation of matter arose as early as the beginning of our era. The search for the philosopher’s stone, by means of which it would be possible to transform some elements into others and, in particular, to extract gold and silver from ordinary metals, continued unceasingly throughout the Middle Ages. The spread of this idea over a number of centuries was to a considerable extent connected with the philosophical conception of the nature of things, based on the authority of Aristotle. According to this conception, it was assumed that all bodies consist of one and the same primary substance, and that the four basic elements—earth, air, fire, and water—differ from one another only in possessing, to varying degrees, the qualities of cold, heat, dryness, and moisture. By strengthening or weakening one of these qualities, the properties of matter could be changed. Imbued with these views, it seemed obvious to the alchemists that one substance could be transformed into another, if only the proper method of such transformation could be found. In the period when chemistry was in its infancy, when the nature of chemical compounds was little understood, a noticeable change in the external appearance and properties of a substance during a chemical reaction served as confirmation of such views. From time to time, one after another, people appeared who asserted that they had discovered the great secret of transforming metals into gold, but we have every reason to believe that not a single grain of gold was ever obtained in this way. From the standpoint of the present level of our knowledge, it is clear that the transmutation of elements was a hopeless undertaking with the extremely limited means available to the experimenters of that time. As experimental technique developed and knowledge in the field of chemistry grew continuously, the idea of the transmutation of elements was gradually cast aside and lost its influence on the progress

¹ Based on the Henry Sidgwick Memorial Lecture, delivered in November 1936 at Newnham College (Cambridge).

science. However, in the minds of the broad public the old alchemical ideas took firmer root, and even now there are charlatans, or victims of self-deception, who claim to possess recipes for producing significant quantities of gold by means of transformation. These charlatans not infrequently make such convincing use of scientific jargon that at times they disturb the sleep of even the most sober financiers. We shall see below that at present, with the aid of modern methods, it is possible to produce artificially exceedingly insignificant quantities of gold, but even that only by transforming an even more precious element—namely, platinum.

With the growth of knowledge in the field of chemistry, the old ideas about the transformation of elements lost their footing. It was established that matter can be decomposed into 80 or more elements, whose atoms proved to be unchangeable and indestructible. The ordinary physical and chemical agents then at our disposal proved powerless to alter the atoms of the elements in any way whatever. The idea of the immutability of atoms was dealt a severe blow when, in 1902, it was discovered that the atoms of two well-known elements, uranium and thorium, undergo a genuine process of spontaneous transformation, although one proceeding at a very slow rate. This conclusion followed from the discovery of radioactive properties in the two above-mentioned heavy elements, which spontaneously emit several types of penetrating radiation, causing the blackening of a photographic plate and the discharge of electrified bodies. These radioactive properties are a sign of the instability of atoms. From time to time some atom spontaneously explodes, expelling from itself, with enormous force, a rapid α- or β-particle. An α-particle is a charged helium atom with mass 4, ejected at a speed of about 10 thousand km per second. A β-particle is merely another name for the light negatively charged particle—the electron; it is usually emitted at a speed many times greater. Sometimes the process of transformation is accompanied by penetrating radiation of the type of X-rays, known as γ-rays.

Radioactive Transformations

If we take 1 g of the element uranium, then in 1 sec. about 24 thousand atoms in it decay, each of which emits an α-particle. However, the number of atoms in 1 g is so large that it would take about 4500 million years for half of all the atoms to undergo transformation. As a result of the emission of an α-particle of mass 4 from an atom of uranium with atomic weight 238, a new atom with atomic weight 234 is formed. The atoms of this new element are very unstable and rapidly decay, each atom emitting a rapid β-particle. Once begun, this process of transformation passes through a series of successive stages, in which one unstable atom gives rise to another. The well-known element—

radium, comes from uranium and is the fifth product in the series of its transformations.

The activity of a radioactive substance, measured by the intensity of the radiation it emits, decreases with time according to the law of a geometric progression. If the activity is reduced by half in a time \(T\), called the half-life, then in a time \(2T\) it will decrease to \(1/4\) of its initial value, in a time \(3T\)—to \(1/8\), and so on. It is easy to calculate that after a time \(20T\) the activity will amount to less than one millionth of its initial value. This law of decay is universal for all radioactive substances, but the half-life \(T\) for each radioactive substance has its own characteristic value, varying for different substances within extraordinarily wide limits. For example, the half-life of uranium is 4500 million years, that of radium is 1600 years, while for one of the decay products of radium, known as radium C, it is only one millionth of a second. This law of decay expresses the fact that the number of atoms decaying per unit time is, on the average, always proportional to the number of atoms remaining unchanged at the given moment. Such a relation was to be expected if one assumes that the decay of individual atoms takes place according to the laws of random events.

Fig. 1. Series of uranium transformations. The upper number in each circle denotes the atomic weight, the lower—the ordinal number of the element and the charge of the nucleus. The length of the thick arrow indicates the relative length of the range of the \(\alpha\)-particles.

The remarkable chain of transformations of uranium is shown in Fig. 1, where circles denote the nuclei of successively formed atoms. For each substance the half-life is given and the nature of the particles emitted by it (\(\alpha\)- or \(\beta\)-particles) is indicated. A description of the methods by which this sequence of changes was firmly established would take too much time;

however, it is necessary to draw attention to the exceptional simplicity of the relations connecting with one another all the members of the series of transformations.

We now know that the chemical properties of an element are determined by its atomic number, which at the same time indicates the number of natural units of charge in the nucleus of the element’s atoms. Since electricity has an atomic structure, the nuclear charge is always expressed by an integer, varying from 1 for the nucleus of the lightest element—hydrogen—to 92 for the nucleus of the heaviest element—uranium. Inside each circle in Fig. 1 the atomic number of the nucleus and its atomic weight are indicated, the latter expressed in terms of the atomic weight of oxygen, which is taken as equal to 16.

The $\alpha$- or $\beta$-particle liberated during a transformation flies out directly from the nucleus of the atom. Thus the ejection of an $\alpha$-particle, which carries 2 positive units of charge and has mass 4, lowers the atomic number of the nucleus by 2 units and its mass by 4 units. On the other hand, when a $\beta$-particle is emitted, carrying a unit negative charge, the total charge of the nucleus is increased by one unit. Since the $\beta$-particle has a very small mass, upon its emission the mass of the atom does not change in a first approximation. These simple considerations, based on examining the nature of the emitted radiation, are sufficient for a satisfactory explanation of the atomic numbers and masses of all the elements in the long chain of transformations. At the present time it is firmly established that mass and energy are equivalent. Knowing the exact mass of the $\alpha$-particle (the helium nucleus) and the maximum kinetic energy of the emitted $\alpha$- or $\beta$-particle, one can accurately calculate the atomic weights of all the atoms of the series, if only the atomic weight of uranium is known. The final product of the series, which shows no traces of activity, has the same atomic number as lead, but its atomic weight is equal to 206, in contrast to the atomic weight of ordinary lead, 207.2.

At the present time it is well known that most elements are mixtures of several isotopes, i.e., atoms with the same nuclear charge but with different masses. Aston showed that ordinary lead consists of at least three isotopes with atomic weights 206, 207, and 208, of which the predominant one is the isotope with atomic weight 206. The final product of the uranium series, usually called uranium lead, is thus one of the isotopes of ordinary lead (isotope 206). Lead separated from an old uranium mineral consists chiefly of this isotope of lead. Let us also note that in the uranium series there occur 2 radioactive isotopes of lead with atomic number 82, namely: radium B with atomic weight 214 and radium D with atomic weight 210.

It is necessary to mention that a similar long sequence of transformations has been discovered in the elements thorium and actinium. The final product of the thorium series is again an isotope of lead, but with atomic weight 208, and not 206, as for uranium lead.

Lead extracted from a pure thorium mineral consists chiefly of the isotope with atomic weight 208. The final product of the actinium series of transformations is likewise an isotope of lead, but with atomic weight 207. It is remarkable that the final products of all three series of transformations are three different isotopes of lead. The striking changes in the chemical and physical properties of elements during radioactive transformations are well illustrated by the example of the transformation of radium. Radium in pure form is a metal which, in its chemical properties, resembles barium. It decays, emitting α-particles with a half-life of 1600 years, and is transformed into a heavy radioactive gas, now called radon. This gas is chemically inert and in this respect belongs to the well-known group of inert gases, which includes, in particular, helium, neon, and argon. The atoms of the emanation are very unstable in comparison with radium atoms: half of them decay in 3.8 days. The intense radioactivity of this gas can be demonstrated by a simple experiment. An insignificant quantity of this gas, of volume less than \(1/10\ \text{mm}^3\) at normal pressure, is introduced into an evacuated glass vessel whose inner walls are coated with a layer of phosphorescent zinc sulfide. In the same second the vessel begins to shine brightly as a result of the intense bombardment of the zinc sulfide by the enormous number of α-particles emitted by the emanation during its decay.

It should be borne in mind that the energy released in the transformation of a single atom, chiefly in the form of the kinetic energy of α- and β-particles, is enormous in comparison with the energy released per atom by the most powerful explosives. If \(1\ \text{g}\) of a pure radium salt is placed in a closed glass tube, the α-particles emitted by the radium and by the products of its decay will be absorbed either by the radium salt or by the glass walls, and their kinetic energy will ultimately be transformed into heat in situ. Some of the fastest β-particles and most of the γ-rays will be emitted through the glass walls. Owing to the liberation of heat, the tube with radium is always several degrees warmer than the surrounding medium. The emission of heat will slowly diminish with time and will be reduced by half after 1600 years.

α-particles, in passing through matter, lose their velocity and, ultimately, their charge, becoming ordinary helium atoms. The helium obtained in this way can be isolated by dissolving or heating the radium salt. The enormous quantities of heat emitted by a radioactive substance can best be illustrated by the example of a more rapidly decaying substance, for example the radium emanation with a half-life of 3.8 days. As is seen from Fig. 1, the emanation, decaying with the emission of α-particles, gives rise to four rapidly changing products—radium A, radium B, radium C, and radium C′, two of which emit α-particles and two—β-particles. Several hours after the emanation is placed ...

placed in a sealed tube, a kind of equilibrium is established between the emanation and its four short-lived decay products, when the activity of the decay products is determined by the decay of the emanation. After 1–2 months practically all the emanation is transformed into radium D. The half-life of the latter (25 years) is so long in comparison with the periods of its decomposition products—radium E and radium F—that the final decay of these products is determined by the half-life of radium D.

Let us imagine that we have succeeded in obtaining a considerable quantity, say a kilogram, of radium emanation and placing it in a bomb made of refractory material. After approximately 2 hours, the quantity of heat released per unit time will correspond to a power of 20 thousand kW, and, unless very effective cooling is provided, the bomb will melt. This thermal effect will weaken at the same rate as the decay of the emanation proceeds, and will be reduced by half after 3.8 days. After about 2 months most of the emanation will have disappeared, and the bomb will be filled with helium gas, produced from $\alpha$-particles, with a volume 3 times greater than the initial volume of the emanation, while the walls of the vessel will be covered with a layer of precipitate consisting of 946 g of radium D, which is a slowly decaying radioactive isotope of lead with atomic weight 210. If we could continue the experiment for another 200 years, then by the end of that time we would find that the radium D had almost entirely disappeared, and in its place there had appeared an inactive isotope of lead—uranium lead, with atomic weight 206. As a result of the emission of $\alpha$-particles from radium F, the volume of helium would have increased by 3–4 times.

It is interesting to note that the last radioactive element of the series, radium F, usually known as polonium, was the first radioactive element isolated from uranium minerals by Marie Curie in 1897.

Although we can predict with certainty the consequences of the experiment described, we are deprived of any possibility of carrying it out in practice, since, in order to obtain 1 kg of emanation, about 200 t of radium would be required, whereas the entire quantity of radium mined up to the present is probably less than 1 kg. However, we may be grateful for this circumstance, since in carrying out an experiment on such a scale the intense emission of energy from the bomb in the form of penetrating $\gamma$-rays, equivalent to a power of 1000 kW, would undoubtedly prove dangerous to the health of people nearby.

Nevertheless, I believe that such an imaginary experiment will help you to gain an idea of the gigantic scale of the emission of energy in radioactive changes, and also of the astonishing nature of the transformations as a result of which the emanation ultimately turns into helium and uranium lead. These radioactive transformations are spontaneous and are not subject to external influences. Neither strong heat, nor extreme cold, nor in ma-

least degree influence this natural process. We can only observe and study these astonishing transformations, without being able in any way whatsoever to affect them.

Radioactive properties manifest themselves noticeably in the two heaviest elements—uranium and thorium—and only to a very slight degree in a few other elements. Most elements usually show no signs of radioactivity whatever, so that we may quite properly conclude that the atoms of these elements, under ordinary conditions on our earth, are invariably stable. In the last few years methods have been found not only for the artificial transformation of one element into another, but also for obtaining many new radioactive substances, disintegrating according to the same laws as natural radioactive substances. We owe this knowledge to intensive investigations, lasting for a number of years, and to the development of new, powerful methods of attacking this most fundamental problem of physics.

Elementary Particles

The study of radioactive transformations has led us to the discovery of fast $\alpha$- and $\beta$-particles as possible constituent parts of the heavy atomic nucleus. In further investigations of the transformations of ordinary elements, the existence was discovered of several more types of elementary particles liberated in the explosions of atomic nuclei. The most important of these newly discovered particles are the proton, neutron, deuteron (or deuton), and positive electron. A proton is the nucleus of hydrogen with charge 1 and mass 1.0076. The neutron is an uncharged particle with a mass somewhat greater than that of the proton, namely 1.0090. According to present-day views, these two particles—the proton and the neutron—are closely connected with one another. It is supposed that, under the action of the intense forces existing inside the atomic nucleus, a neutron can be transformed into a proton by the removal of an electron from it and, conversely, a proton can be transformed into a neutron by the addition of an electron. Although we do not yet possess direct evidence of such mutual transformations, the general data undoubtedly speak in favor of the conception that there is a definite connection between these two particles. It is natural to suppose that the neutron is a very close combination of a proton and an electron, although up to the present time the explanation of the difference in the masses of these particles encounters well-known difficulties.

The $\alpha$-particle is the nucleus of helium with charge 2 and mass 4.0029. The recently made discovery by Urey that an isotope of hydrogen with mass 2 is always present in a small proportion in ordinary hydrogen has proved very important both for physics and for chemistry. By subjecting ordinary water to repeated electrolysis, one can obtain pure heavy water, in which the hydrogen atom of mass 1 is replaced by its isotope of mass 2. Such water is approximately 11% heavier than ordinary water and has different boiling points.

and freezing. Heavy hydrogen with mass 2 was called deuterium and received the chemical symbol D. When an electric discharge is passed through heavy hydrogen, some of its atoms lose electrons and become positively charged ions. These ions are called “deuterons” (or deutons), whereas the ions of ordinary hydrogen, as we have seen, are called “protons.” It is desirable to have different names for these two ions, since they are often used as fast particles for bombarding matter. We shall see below that fast protons and deuterons, along with α-particles and neutrons, have proved to be extraordinarily effective agents for the transformation of many elements. There is direct experimental evidence that the deuteron, as might have been expected, is a close combination of a proton with a neutron.

In some transformations there also appears a positive electron—the antipode of the negative electron of small mass. This elusive particle was first discovered by Anderson several years ago in experiments with cosmic rays. We can now obtain positive electrons in laboratories in small quantities by passing γ-rays of high quantum energy through matter. In addition, certain light elements eject positive electrons at high speed when bombarded with α-particles. The positive electron has been given the name “positron”; it is believed to possess the same negligible mass as the ordinary negative electron, and an equal but opposite charge.

In some transformations two more light elements are formed, or, more precisely, two new isotopes of hydrogen and helium, namely: \(^3\mathrm{H}\) and \(^3\mathrm{He}\). Both of these isotopes are apparently stable, but neither has so far been detected in ordinary substances. It was initially assumed that \(^3\mathrm{H}\) is present in preparations of heavy water, but subsequent observations did not confirm this.

Methods of Detecting Fast Particles

We have seen that all kinds of radiation from radioactive substances possess the characteristic property of discharging an electrified body. This property is explained by the ability of moving α- and β-particles, when passing through a gas, to form a multitude of positively and negatively charged particles—ions. The primary act of ionization consists in the detachment from an atom or molecule of one of the outer electrons as a result of collision with a fast particle. The ions move through the gas in an electric field, the positive ions going to the negative electrode and conversely. The motion of these two kinds of ions in opposite directions is equivalent to the passage of an electric current through the gas.

In the initial period of research on radioactivity, the action of the various types of radiation was usually studied and compared with the po-

by means of such an electrical method, the measuring instrument being an electroscope or an electrometer. This electrical method is a very convenient means for detecting negligible quantities of radioactive substance and is still widely used in those cases where an easily measurable effect may be expected.

The rapid progress in our knowledge of the transformations of the elements that has taken place in recent years has, to a considerable degree, been due to the discovery of delicate methods for detecting and counting individual particles moving at high velocities, for example protons or α- and β-particles. All these methods are ultimately based on the phenomena of ionization of a gas by fast particles flying through it.

The expansion method (Wilson chamber)

Fig. 2. Wilson chamber. The light piston is suddenly lowered by decreasing the pressure beneath it. The gas \(A\) in the space above the piston expands, cooling in the process so much that the vapor contained in it becomes supersaturated. This vapor condenses in the form of small droplets on the charged particles (ions) present in the gas. The chamber is illuminated through the glass wall, and the resulting droplets are photographed in the light scattered by them, which enters two cameras arranged above.

\(A\) — the space containing gas saturated with water vapor

\(B\) — the pressure in this space suddenly drops by means of opening a valve.

The most remarkable of these methods was invented by Prof. C. T. R. Wilson. It is based on the observation that the ions formed during the motion of a fast particle can, under certain conditions, become centers around which condensation of water vapor occurs. In this case, a visible droplet of water forms around each ion. Since a fast α-particle, in passing through a gas, produces more than 100 thousand pairs of positive and negative ions, the true path of the flying particle becomes visible as a line of water droplets closely following one another. Stereoscopic photographs of tracks taken immediately after expansion clearly show the position of the particles’ paths in space.

The apparatus used for this purpose is called a Wilson chamber. A typical diagram of such a chamber, together with an explanation of the principle of operation, is given in Fig. 2. The chamber has a cylindrical shape: the space \(A\) is saturated with water vapor. The pre-

To the article by E. Rutherford

Fig. 3. Tracks of α-particles emitted by thorium \((C + C')\), separating into two groups with ranges of \(8.6\ \text{cm}\) and \(4.8\ \text{cm}\) in air.

(Photograph by Prof. J. Chadwick).

Fig. 4. Tracks of photoelectrons with a range of about \(1\ \text{cm}\), formed upon absorption of the characteristic \(K\)-radiation of silver in air (energy about \(21\,000\ \text{V}\)). The straight-line track belongs to an electron of much greater energy and was probably caused by cosmic rays.

(Photograph by Prof. C. T. R. Wilson).

Fig. 5. Track of a photoelectron formed upon absorption of a quantum of X-rays with energy \(\sim 40\,000\ \text{V}\). In its initial straight-line section the track undergoes a sharp deflection as a result of a close approach of the electron to an atomic nucleus. The photograph clearly shows the increasing density of ionization and the curvature of the path due to collisions as the particle’s velocity decreases toward the end of the track.

(Photograph by Prof. C. T. R. Wilson)

Let us suppose, for example, that at the moment of expansion an α-particle passes through the gas. Then, if the degree of expansion has been chosen in a suitable way, each ion formed along the path of the α-particle becomes a center of condensation, and thus the path of the particle becomes clearly visible. A photograph of the tracks of α-particles, obtained by the method described, is shown in Fig. 3. The source of the α-particles in this case was a small metallic plate, activated by irradiation with thorium emanation and placed in the expansion chamber. The surface of the plate is covered with an invisible film of active substance containing two sources of γ-rays—thorium C and thorium C′. All the α-particles emitted by thorium C′ have the same velocity and a range in air equal to 8.6 cm. The tracks of α-particles from thorium C, with a shorter range (4.8 cm), are visible in the photograph intermingled with the predominant tracks of the faster α-particles.

The overwhelming majority of α-particles penetrate the gas along straight trajectories, and the end of the track indicates the point at which the velocity of the α-particle has fallen so much that it is no longer able to form ions. A β-particle, in passing through the gas, gives a track revealing certain characteristic differences from the track of the more massive α-particle. First, the track of a β-particle is much less dense, owing to the considerably smaller ionization produced by the β-particle per unit path. This is clearly visible in the photograph of β-particle paths shown in Fig. 4. The rectilinear path of a fast β-particle is marked by a chain of drops so far separated from one another that their number can almost be counted. Second, because of collisions with atoms, the light β-particle deviates from a straight path more often than an α-particle moving with the same velocity. This explains the winding character of the β-particle tracks in the photograph. The noticeable thickening at the end of the track is the result of an increase in the ionization produced by the β-particle as its velocity decreases.

The photograph of the path of a β-particle shown in Fig. 5 is of interest in that it shows a series of adventures that occur to a β-particle as it passes through the gas. The long track running on the left bends sharply, almost at a right angle. This occurs as a result of the collision of the β-particle with the heavy nucleus of one of the atoms. The short tracks branching off from the main path represent the paths of secondary electrons ejected from atoms as a result of collisions with the fast β-particle.

The velocity and energy of a flying β-particle can be determined directly by measuring the curvature of the particle’s path in a uniform magnetic field. If the field is perpendicular to the direction of flight, then the β-particle moves along a circle. If the field is sufficiently strong and the velocity of the β-particle is not too great, then the particle’s track in the gas may describe a complete circle many times in succession. The direction of deflection of the particle in the magnetic field depends on the sign of its charge. If the direction of motion of the particle is known, then

in this way one can at once determine whether the track belongs to a fast positive or negative electron.

Electrical Method (Ionization Chamber and Geiger Counter)

In a number of experiments it is important to be able to count the number of fast particles entering the chamber of an apparatus during a given interval of time. This can most simply be achieved by means of the electrical method of counting (the ionization chamber). The principle of the method is explained in Fig. 6. Suppose, for example, that it is necessary to count $\alpha$-particles. They are directed into the chamber through a thin metal foil $A$ and are stopped by striking a parallel, insulated plate $B$. A voltage sufficient to carry to the electrodes the ions formed between the plates, which are usually separated from one another by a distance of 3–5 mm, is applied to the electrodes. Each entry of an $\alpha$-particle into the chamber causes a small increase in the potential of plate $B$, which is automatically increased more than 100 million times by a series of amplifiers specially adapted for this purpose. The instantaneous rise in potential at the output is sufficiently large—of the order of 100 V—to produce a deflection of the oscillograph calculated for higher powers and having a very small natural period of oscillation.

Fig. 6. Counter with one chamber.

A photographic record of the deflections of the oscillograph caused by the entry of $\alpha$-particles into the chamber is shown in Fig. 7. Each vertical line represents the amplified electrical effect of one $\alpha$-particle, while the continuous horizontal band is the natural motion of the oscillograph film in the absence of $\alpha$-particles entering the chamber. With the aid of a rapidly moving photographic film, individual particles can be recorded even if 1000 particles per minute enter the chamber. In an analogous way fast protons and deuterons can also be counted. However, since both these particles carry a unit charge, whereas the charge of an $\alpha$-particle is 2, the ionization produced by $\alpha$-particles is approximately 4 times greater than the ionization caused by a proton or deuteron of the same velocity. Therefore a proton entering the chamber gives a deflection approximately 4 times smaller than an $\alpha$-particle of equal velocity. The recording of protons obtained in this way is also shown in Fig. 7. The difference in the magnitude of the deflections makes it possible

to judge whether a particle carries a single or double charge, and under known conditions it is possible easily to distinguish the tracks of protons and α-particles.

The described method of electrical counting of particles is not applicable to fast β-particles, since in this case the ionization is too small to give a measurable deflection. However, Geiger invented another simple and sensitive method for counting β-particles, which has come into general use. The construction of this counter is extremely simple. It consists essentially of a hollow metal cylinder, closed at both ends by insulating stoppers, through which, along the axis, passes a wire or rod connected to a simple amplifying system. In the cylinder, filled with air or another gas, a certain pressure is established and a voltage is applied to it, almost sufficient for a discharge to begin in the gas. When a β-particle passes through the gas under such conditions, the ionization it produces is amplified many times over, owing to the well-known process of ionization by collision, and an instantaneous discharge passes between the wire and the cylinder. The discharge current is increased by amplifiers, and the β-particles can be counted in the same way as α-particles, either by clicks in a telephone or with the aid of an oscillograph. This instrument, called the Geiger–Müller counter, is a remarkably effective means of counting fast positive and negative electrons entering the cylinder through its walls.

Since γ-rays, passing through the walls of the cylinder and the gas-filled space, produce β-rays, the Geiger counter can also serve as a sensitive means for detecting γ-rays.

In cases where it is necessary to count large numbers of fast particles, whether α- or β-particles or protons, an automatic counting system is often used, in which the number of particles is recorded by an automatic counter. Ingenious methods of automatic particle counting were developed by Wynn-Williams and are widely used in many laboratories.

Since the neutron has no charge, it can pass freely through the outer shell of atoms without forming ions. Sometimes, however, a neutron collides on its path with an atomic nucleus and sets it into rapid motion. This recoil nucleus is capable of forming ions in the gas until it comes to rest. Recoil particles can be detected by methods used for counting α- and β-particles. Usually, for this purpose, the counter chamber is filled with hydrogen, helium, or air. In general, no more than one neutron in 5 thousand entering the counter produces a measurable deflection in the oscillograph. The registration of neutron recoil particles in helium is shown in Fig. 7.

We shall see below that, for counting very slow neutrons, a number of effective methods have been developed, based on the ability-

ness of such neutrons to produce transformations of certain elements, in particular lithium and boron.

Transformation of Elements by Means of α-Particles

After a number of natural transformations of uranium and thorium had been studied, one could hope that someday we would succeed in finding methods for destroying the stable atoms of some ordinary elements. In order to attack this problem with some chance of success, it was necessary to obtain some idea of the structure of the atoms of various elements. We now believe that the atoms of all elements are built according to one type and are closely connected with one another by definite relations. At the center of each atom is an extremely small nucleus, possessing an excess positive charge. In this nucleus is concentrated the greater part of the mass of the atom. The charge of the nucleus changes by one unit in passing from one element to the next and, as we have already seen, is equal to one unit for hydrogen and 92 for uranium. Around the nucleus, at some distance from it, are situated light negative electrons, whose number is equal to the charge of the nucleus. The charge of the nucleus of a given element determines the number and distribution of the outer electrons, so that the properties of the atom, as Moseley first showed, are determined by an integer. Almost all the integers from 1 to 92 correspond to elements known to us.

Some of the outer or planetary electrons may be easily separated from the atom by means of an electric discharge or by other methods, but after a short time their place is taken by other electrons, and the atom returns to its former state. To bring about a stable transformation of an atom, it is necessary to change either the charge of the nucleus, or its mass, or both together. Meanwhile, the atomic nucleus is held together by extremely powerful forces, and therefore from the very beginning it was clear that, in order to split the nucleus, it was necessary to act upon it with very concentrated sources of energy. Twenty years ago, of all particles known to science, the fast α-particles spontaneously emitted by radium and other radioactive substances possessed the greatest energy. The velocity and energy of these particles are so great that they can penetrate deeply into the interior of the atom, and, by observing their deflection or scattering, one can obtain valuable information about the nature and intensity of the deflecting field inside the atom. Indeed, present-day ideas about the nuclear structure of atoms arose as a result of the study of the scattering of α-particles through large angles as they pass through matter. Consider, for example, the path of an α-particle passing in the immediate vicinity of the nucleus of a heavy atom. Since the α-particle carries 2 units of positive charge and the nucleus itself has a large positive charge, repulsive forces arise between the nucleus and the α-particle.

Fig. 7. Recording of ionizing particles on an oscillograph tape. The two upper photographs were obtained under identical conditions, and the difference in the sizes of the recoils is explained by the different ionizing power of $\alpha$-particles and protons. The lower photograph shows recoils produced by recoil atoms of helium, which were set in motion by impacts of neutrons with an energy of 2 million V. The marked time intervals are seconds.

$a$—$\alpha$-particle, $b$—proton.

Fig. 9. Tracks of $\alpha$-particles in oxygen. One track at the end splits as a result of a collision with an oxygen nucleus, the short branch belonging to the recoiling oxygen nucleus, and the longer one to the deflected $\alpha$-particle. Measurements of the angles of deflection of the two branches showed that the momentum and energy in this collision are conserved.

(Photograph by Prof. P. M. S. Blackett)

... forces, which increase strongly near the nucleus. Therefore an α-particle describes a curvilinear path around the nucleus, and if the forces of interaction obey Coulomb’s law, then this path has the form of a hyperbola, whose asymptotes coincide with the directions of approach and recession of the α-particle. An α-particle can undergo a considerable deflection as the result of a single collision with a nucleus. The orbits of α-particles flying at various distances from the center of the nucleus are shown in Fig. 8, where the relative dimensions of the heavy nucleus are indicated by a black circle.

An α-particle flying straight toward the center of the nucleus turns back at some distance from it; this minimum distance to which the α-particle can approach the given nucleus is shown in Fig. 8 by the circle drawn around the nucleus. The closer to the center of the nucleus the direction of impact of the α-particle passes, the greater the angle of its scattering. Measurements of the number of α-particles scattered at various angles while passing through matter gave results completely consistent with calculations based on these assumptions.

Fig. 8. Orbits of α-particles flying near a heavy nucleus.

The fraction of α-particles scattered through a given angle depends on the square of the charge of the nucleus and increases rapidly as the velocity of the α-particle decreases. It should be borne in mind that the target area represented by the nucleus is so small that an α-particle only rarely passes close enough to the nucleus to undergo a significant deflection. An example of such a strong deflection of an α-particle in passing through oxygen is shown in Fig. 9. In the collision the α-particle was deflected to the left, and the recoil track of the nucleus is visible on the right. Until now we have considered only “elastic” deflections, obeying the laws of mechanics. Indeed, in such cases the colliding nuclei behave like tiny perfectly elastic billiard balls. No transformation of elements takes place here. It is clear, however, that in the case of a “head-on” collision of a fast α-particle with a light nucleus carrying a small charge, the repulsive forces will be relatively small and may allow the α-particle to approach very close to the nucleus and perhaps even penetrate into it. In this latter case the entire structure of the nucleus would be disrupted in the strongest possible way, which could lead to its disintegration. On the basis of these considerations, atoms of several

light elements were subjected to bombardment by a very large number of α-particles. Continuing this experiment, in 1919 I obtained experimental evidence that a small number of nitrogen atoms, when bombarded, disintegrated, emitting fast hydrogen nuclei, now known under the name of protons. In the light of later investigations, the general mechanism of this transformation is quite clear. From time to time an α-particle actually penetrates into the nitrogen nucleus, forming for one instant a new nucleus of the type of a fluorine nucleus with mass 18 and charge 9. This nucleus, which does not exist in nature, is extremely unstable and immediately disintegrates, ejecting a proton and turning into a stable oxygen nucleus with mass 17. The phases of this transformation process are shown below in the form of a relation resembling a chemical equation. The left-hand side of the equation contains the elements entering into the reaction, and the right-hand side—the final products of the transformation. The two numbers before each symbol denote the mass and charge of the nucleus of the given element. As is seen from the equation, the total charge of the nuclei is conserved in the transformation, as is their mass, if one takes into account the equivalence of mass and energy. For this purpose the symbol \(E\) is introduced on the right-hand side of the equation; it denotes the mass equivalent to the sum of the kinetic energies of the proton and the oxygen nucleus, less the initial energy of the α-particle.

\[ {}^{14}_{7}\mathrm{N} + {}^{4}_{2}\mathrm{He} \to {}^{18}_{9}\mathrm{F} \to {}^{17}_{8}\mathrm{O} + {}^{1}_{1}\mathrm{H} + E. \]

The transformation occurs on a negligible scale, for only one α-particle out of 50 thousand approaches the nucleus closely enough to be captured by it. By photographing the tracks of several hundred thousand α-particles in a Wilson chamber filled with nitrogen, Blackett established several distinct cases of transformation of the nitrogen nucleus. One of these photographs is presented together with an explanatory diagram in Fig. 10. In the photograph the recoil path of the proton, with a large range, and the short track of the recoil nucleus are clearly visible.

A transformation of an analogous type occurs with a whole series of light elements when they are bombarded by α-particles, and in all cases a fast proton is liberated. During the last few years the mechanism of these transformations has been subjected to careful study, which has yielded a number of important results. It has turned out that the emitted protons consist of two or more groups, each of which possesses a definite velocity. The different energy of the protons of these groups is apparently the result of the emission of energy from the exploding nuclei in the form of γ-rays. There is also clear evidence of the existence in the nucleus of definite energy levels, or “resonance” levels, which leads to the selective capture of α-particles of a definite velocity.

For the article by E. Rutherford

Fig. 10. Splitting of nitrogen by α-particles. Of the large number of α-particles passing through nitrogen, one particle effected the transformation of the nitrogen nucleus into the nucleus \({}^{17}\mathrm{O}\), with the emission of a proton possessing a large energy

\[ \left({}^{14}_{7}\mathrm{N}+{}^{4}_{2}\mathrm{He}\to{}^{17}_{8}\mathrm{O}+{}^{1}_{1}\mathrm{H}\right) \]

(Photograph by Prof. P. M. S. Blackett.)

S

Fig. 11. Tracks of recoil protons arising when methane is bombarded by neutrons with an energy of 2.4 million V. The neutron source was placed at S, where a target of heavy hydrogen was bombarded by accelerated deuterons

\[ \left({}^{2}_{1}\mathrm{H}+{}^{2}_{1}\mathrm{H}\to{}^{1}_{0}n+{}^{3}_{2}\mathrm{He}\right) \]

(Photograph by P. I. Dee and C. W. Gilbert.)

The Discovery of the Neutron

We have already seen that the proton appears as a product of the transformation of a number of light elements when they are bombarded with α-particles. Upon a more detailed study of these transformations, yet another particle was discovered, the significance of which is very great. When the light element beryllium, of mass 9, is bombarded with α-particles, protons are not formed, but Bothe found that a radiation is emitted with a penetrating power exceeding even the maximum penetrating power of radium γ-rays. The Curie-Joliot spouses discovered certain specific features in the absorption of this type of radiation. Finally, in 1932 Chadwick showed that the principal part of this radiation does not at all belong to the type of γ-rays, but consists of a stream of fast uncharged particles with a mass approximately equal to the mass of the hydrogen atom. These particles, called neutrons, possess very distinctive properties, since, owing to the absence of charge, the neutron passes freely through atoms and produces no ionization along its path. The mechanism of the transformation in which the neutron is formed apparently consists in the following: from time to time an α-particle is captured by a beryllium nucleus of mass 9, momentarily forming a nucleus of \(^{13}\mathrm{C}\) with a large excess of energy. This nucleus immediately disintegrates into a stable nucleus and a neutron, the excess energy of the reaction being released in the form of the kinetic energy of the two particles—the final products of the reaction. A very convenient and stable source of neutrons can be obtained by mixing about 100 mg of pure radium salt with beryllium powder in a sealed tube. As a result of the bombardment of beryllium by α-particles, about half a million neutrons per second are obtained, most of which pass through the walls of the tube. Intensive neutron sources can also be obtained by using radium emanation instead of radium salt. In this case the emission of neutrons weakens with time at the same rate as the decay of the emanation.

The idea of the possible existence of neutrons as constituent parts of the atomic nucleus had been discussed long before their experimental discovery. It may be of some interest to quote the statement made by the author on this question in the Bakerian Lecture delivered at the British Royal Society in 1920:

“If our supposition is correct, it seems very probable that one electron is capable of binding together two H nuclei, and perhaps also one H nucleus. The first supposition entails the possibility of the existence of an atom with a mass approximately equal to 2, carrying a unit charge. Such an atom must be regarded as an isotope of hydrogen. In the other case, the possibility is assumed of the existence of an atom with mass 1 and zero nuclear charge. An atomic structure of this kind by no means appears impossible. According to modern views, a neutral

the hydrogen atom is regarded as a nucleus with unit charge, to which one electron is attached at some distance, and the spectrum of hydrogen is explained by the motions of this distant electron. However, under certain conditions the possibility is not excluded of a closer combination of the electron with the H nucleus and of the formation of a kind of neutral doublet. Such an atom would possess very peculiar properties. Its external field would be practically zero everywhere, with the exception of the region immediately adjacent to the nucleus, thanks to which it could pass freely through matter. The presence of such atoms would probably be difficult to detect with the aid of a spectroscope, and it would be impossible to keep them in a hermetically sealed vessel. On the other hand, they should easily penetrate into the interior of the atom and may either combine with the nucleus or disintegrate in its intense field, the result of which would probably be the ejection from the atom of H or of an electron, or of both at once.”

At first it was assumed that neutrons could be obtained by passing an electric discharge through hydrogen. Experiments carried out in this direction gave a negative result. It now seems obvious that neutrons cannot be obtained in this way by means of voltages of the usual order.

Chadwick and I, many years ago, also performed experiments with the aim of establishing whether neutrons are formed when aluminium is bombarded by fast α-particles, but we obtained negative results. No one could have predicted the conditions under which this remarkable particle was finally discovered.

We have seen that the presence of a neutron can be detected if, on its path, it undergoes an elastic collision with a nucleus. If, for example, a neutron passes through hydrogen, then a head-on collision of the neutron with an H-nucleus sometimes occurs. In this case the energy of the neutron is transferred to the nucleus, which begins to move with a velocity equal to the velocity of the incident neutron. In a glancing collision only part of the neutron’s energy is transferred to the nucleus. Fig. 11 shows a photograph of neutron recoil particles in methane, obtained by Dee and Gilbert with the aid of a Wilson chamber. When a stream of fast neutrons is passed through hydrogen or a hydrogen-containing substance, for example water or paraffin, many neutrons are rapidly slowed down by these collisions until, finally, their energy becomes comparable with the energy of thermal motion of the surrounding molecules. This method of obtaining very slow neutrons has proved very useful in many experiments. Such slow neutrons pass with insignificant absorption through thick layers of many substances, for example iron and lead, but are strongly absorbed by certain elements, in particular boron, cadmium, and gadolinium. The absorption of neutrons by gadolinium is so great that a layer of this substance only a fraction of a millimeter thick absorbs practically all slow-

neutrons. Such strong absorption of slow neutrons by certain elements is undoubtedly a consequence of their capture by the nuclei of the elements, as a result of which the latter undergo transformation. Sometimes the capture of a neutron by a nucleus imparts to the nucleus such instability that it breaks up into parts. In other cases this capture can transform one isotope of an element into another, with a mass greater by one unit, or form an unstable or radioactive isotope that decays with the emission of a positive or negative electron.

As Feser, Harkins, and also Fermi and his collaborators have shown, neutrons, and especially slow ones, are an extraordinarily effective means for the transformation of elements. Owing to the absence of charge, slow neutrons can freely penetrate heavy nuclei, whereas a charged particle requires a large kinetic energy in order to approach closely the heavy atomic nucleus despite the action of the repulsive forces of its electric field. I shall illustrate the effectiveness of the neutron as a means of transforming atoms by the example of the light elements lithium and boron. Taylor and Goldhaber recently developed a photographic method for studying the neutron transformations of certain elements. A special photographic plate is impregnated with a solution of a compound containing lithium or boron and is irradiated for several days with a source of slow neutrons. After the plate is developed, the tracks of fast particles can be distinctly examined under a powerful microscope. In the case of irradiation of lithium, its isotope of mass 6 captures a neutron and then splits into an α-particle ($^{4}\mathrm{He}$) and an isotope of hydrogen of mass 3 ($^{3}\mathrm{H}$). Using a powerful microscope, one can clearly see on the plate the combined tracks of these two particles, ejected in opposite directions. For boron, two types of transformation are observed. In one case the boron isotope of mass 10 captures a neutron and then decays into a lithium nucleus of mass 7 and an α-particle ($^{4}\mathrm{He}$); in the other case the unstable nucleus splits into two α-particles and a $^{3}\mathrm{H}$ nucleus. Photographs of tracks obtained in this way are shown in Fig. 12. In the photographs three tracks are clearly visible, diverging from a single point, the longest of them belonging to the $^{3}\mathrm{H}$ nucleus with unit charge. These readily observable transformations of lithium and boron by slow neutrons have proved very useful as a means of detecting and counting slow neutrons. In some cases the chamber of the instrument is filled with gaseous boron fluoride, and in others the chamber walls are coated with compounds of boron or lithium.

Artificial Production of Radioactive Substances

We now turn to a very important discovery made by the Joliot-Curie spouses in 1933. They found that, when certain light elements are bombarded with α-particles, there are formed

radioactive elements, decaying according to the same laws as natural radioactive substances, but emitting in the process of decay not α- or β-particles, but fast positive electrons. As an illustration let us give one example. If boron is subjected for some time to bombardment by α-particles and then examined, it proves to be radio-

Slow     Fast

Magnification × 9000

\[ {}^{10}_{5}\mathrm{B}+{}^{1}_{0}\mathrm{n}\to{}^{4}_{2}\mathrm{He}+{}^{4}_{2}\mathrm{He}+{}^{3}_{1}\mathrm{H} \]

Magnification × 1250

\[ {}^{10}_{5}\mathrm{B}+{}^{1}_{0}\mathrm{n}\to{}^{7}_{3}\mathrm{Li}+{}^{4}_{2}\mathrm{He} \]

Fig. 12. Tracks of particles in a photographic emulsion. A photographic plate impregnated with boron was irradiated with slow neutrons. The products of the splitting of boron, which occurred according to the equations given, gave tracks in the form of a whole chain of individual blackened grains of the emulsion.

(Photograph by G. J. Taylor and M. Goldhaber)

active, namely, emitting a stream of positrons. Its activity falls with time in a geometric progression, decreasing by half in 11 min. The nature of the transformation and its phases are evident from the following equation:

\[ {}^{10}\mathrm{B}+{}^{4}\mathrm{He}\to{}^{14}\mathrm{N}\to{}^{13}\mathrm{N}+\text{neutron}. \]

Owing to an excess of energy the nucleus \({}^{14}\mathrm{N}\) is very unstable and is instantaneously destroyed, turning into the more stable nucleus \({}^{13}\mathrm{N}\). The latter then slowly turns into the stable nucleus \({}^{13}\mathrm{C}\), emitting a positron \(e^{+}\):

\[ {}^{13}\mathrm{N}\to{}^{13}\mathrm{C}+e^{+}. \]

The production of this “radio-nitrogen” is confirmed by the fact that, when collected, it behaves like a radioactive gas with the chemical properties of nitrogen.

It is interesting to note that the same radioactive gas can be obtained in an entirely different way. If carbon is bombarded with fast protons, the following reaction takes place:

\[ {}^{12}\mathrm{C}+{}^{1}\mathrm{H}\to{}^{13}\mathrm{N}. \]

The radio-nitrogen \({}^{13}\mathrm{N}\) obtained in this way, in its radioactive

and in chemical properties it is identical to the gas formed when boron is bombarded with $\alpha$-particles.

In a similar way, aluminum bombarded with $\alpha$-particles gives rise to radioactive phosphorus with atomic weight 30, having a half-life of 3.2 min. Radiophosphorus, emitting a positron, is transformed into a stable silicon nucleus with atomic weight 30.

In the last few years a large number of radioactive substances have been obtained by bombarding elements not only with $\alpha$-particles, but also with fast protons and deuterons. Fermi and his collaborators also showed that slow neutrons are a very effective means of forming radioactive substances even from the heaviest elements. At the present time more than 50 such radioactive elements are known, and in the majority of cases they decay with the emission of negative electrons ($\beta$-particles). Even the heaviest elements—uranium and thorium—are transformed when bombarded with slow neutrons and in each case produce a series of new radioactive substances, but the precise interpretation of these transformations is still in the process of being worked out.

Methods of Artificial Transmutation

Up to now we have been dealing with transformations produced by $\alpha$-particles, which themselves are obtained in the process of decay of radioactive substances, and by neutrons arising when beryllium is transformed by $\alpha$-particles. The quantities of radium available to our laboratories are limited, so that the results of transformations produced with the aid of $\alpha$-particles are, generally speaking, small and can be studied only thanks to the exceptionally sensitive methods we have developed for counting individual rapidly moving atoms of matter. Ten years ago it was established that, for a further expansion of our knowledge of the transformations of the elements, much more intense streams of bombarding particles are required. It has long been known that when an electric discharge passes through a rarefied gas, a multitude of charged atoms and molecules is formed. For example, if a discharge is passed through hydrogen, an enormous number of charged H atoms (protons) is formed, as well as charged molecules. Thanks to the recent discovery of heavy hydrogen of mass 2, known under the name deuterium, another projectile has become available to us, namely the deuton, which has acquired an important role in the expansion of our knowledge of the transformations of the elements. Large quantities of protons and deutons can easily be obtained by passing an electric discharge, respectively, through hydrogen and deuterium; but in order to impart to them a high velocity, it is necessary to accelerate them by a strong electric field. This entails the necessity of using apparatus which in some cases is on the scale of engineering structures, as well as voltages on the order of a million volts; in addition, high-speed pumps are required to maintain a good vac-

vacuum, in order to prevent an electrical discharge in the accelerating system. In the Cambridge laboratory a high constant voltage was obtained by multiplying the voltage taken from a transformer by means of a system of condensers and rectifiers. In Fig. 13 is shown a photograph of the installation used by Cockcroft and Walton in their first experiments of this kind, carried out by them at Cambridge. The method of obtaining and analyzing a beam of fast protons and deuterons for the bombardment of matter is explained in Fig. 14. This apparatus, designed by Dr. Oliphant, was used for studying the transformations of light elements. We hope, in the new high-voltage laboratory at Cambridge, to obtain for purposes of acceleration a steady constant voltage of 2 million V, by means of which sparks about 8 m long could be obtained. In the USA Van de Graaff invented a new electrostatic generator for obtaining the necessary high voltage. A machine of this type was used in experiments on the transformation of elements by Tuve, Hafstad, and Dahl in Washington, steady potentials of up to a million volts being obtained. Prof. E. Lawrence at the University of California constructed an ingenious apparatus, called a “cyclotron,” in which charged particles are automatically accelerated repeatedly a large number of times. This method requires the use of gigantic electromagnets and powerful electric oscillators. In Fig. 15 is shown the diagram of the accelerating system of the cyclotron. The homogeneous magnetic field is directed perpendicular to the plane of the drawing, and the proton or deuteron being accelerated describes a spiral path with a constantly increasing radius. The success of this method of multiple acceleration is based on the circumstance that the time required by a particle to complete a full revolution does not depend on its velocity and, consequently, on the radius of rotation, provided only that the mass of the particle remains almost unchanged. Lawrence believes that a proton or deuteron can make 1 thousand revolutions without undergoing appreciable scattering by residual gas. In this way he succeeded in obtaining intense beams of protons and deuterons with energies reaching 6 million V.

Fig. 13. High-voltage installation used in Cambridge by Cockcroft and Walton in their first experiments on the artificial transformation of elements.

This energy is considerably higher than that which we can expect to obtain in the laboratory by the direct application of high voltages. It is believed that in the near future it will be possible to obtain particles with still greater energy by using a more powerful electromagnet and more intense fields.

Fig. 14. When an electric discharge is passed between the oil-cooled anode \(A\) and the steel cathode \(B\), through which water vapor passes at low pressure, a beam of hydrogen ions is formed, emerging through the aperture \(C\) in the cathode. The ions are accelerated by a voltage applied to the electrodes \(C\) and \(E\), which can be raised to 300,000 V. The steel screen \(S\) serves to protect the glass walls of the apparatus \(N\).

The system is evacuated by a high-speed pump \(O3\), provided with a safety valve \(X\). The beam of ions passes through the magnetic field and the particles of the required type are directed to the target \(T\) through the diaphragm \(I\). The thin mica window \(W\) admits all the fast particles formed into the counter chamber. The Faraday cylinder \(F\) is intended for collecting the beam \(P\) when the electromagnet is not switched on.

During the past year, devices have been constructed that make it possible to bring out from the accelerating chamber of the cyclotron a beam of particles accelerated to high velocity, which is a great advantage in a whole series of experiments. Figure 16 shows a photograph obtained by Lawrence of a luminous beam of deuterons with an energy of 6 million V, corresponding to a current of \(6\,\mu\mathrm{A}\).

In this case an almost parallel beam is obtained, emerging through the platinum window at the end of the tube at a distance of 2 m from the acceleration chamber. Such a beam corresponds to the emission of \(3.8 \cdot 10^{13}\) deuterons per 1 sec., which is equivalent to the flux of \(\alpha\)-particles emitted per 1 sec. by approximately 1 thousand g of pure radium.

Each of these methods of obtaining fast particles has certain advantages in solving problems of a definite type.

Fig. 15. Cyclotron. Positive hydrogen ions, formed in a rarefied gas by electrons from a tungsten filament, are accelerated between \(D\)-shaped electrodes to which an alternating high-frequency voltage is applied. The magnetic field, directed perpendicular to the plane of the drawing, makes the ions move in a circle, and for a definite frequency of the alternating e.m.f. they will always arrive at the place where the two \(D\)’s are separated at the moment when the field is again directed so as to accelerate them.

It was initially supposed that, in order to penetrate the nuclei of comparatively light elements, a particle would be required possessing energy of the same order as an \(\alpha\)-particle, i.e. about 7 million V. However, calculations based on the principles of wave mechanics showed that there is a small probability of a particle’s penetration into the nucleus even in those cases when its energy is considerably lower than the energy of an \(\alpha\)-particle. This idea was fully confirmed by later experiments. Cockcroft and Walton first showed that the artificial transmutation of lithium and boron can be achieved by bombardment with protons having an energy of the order of only 100 thousand V1. The processes of transmutation of these elements by protons and deuterons have now been well studied and are of interest in many respects. Let us first consider the process of transmutation of lithium, which, as we know, consists of two isotopes with masses 6 and 7. Recently methods have been found for separating these isotopes, so that experiments can be performed either

To E. Rutherford’s article

Fig. 16. A beam of deuterons with an energy of 6 million V, emerging from the cyclotron.

(Photograph by Prof. E. Lawrence)

Fig. 18. Tracks of α-particles formed in the bombardment of lithium by artificially accelerated deuterons. The α-particles flying in the direction toward the wall of the chamber have a range \(> 10\) cm and are formed in the transformation

\[ {}^{6}_{3}\mathrm{Li} + {}^{2}_{1}\mathrm{H} \to {}^{4}_{2}\mathrm{He} + {}^{4}_{2}\mathrm{He} \]

(the range of the α-particle is 13.4 cm). The photograph also shows tracks of a group of α-particles with a continuous range distribution up to 8 cm. This group is formed as the result of the process

\[ {}^{7}_{3}\mathrm{Li} + {}^{2}_{1}\mathrm{H} \to {}^{4}_{2}\mathrm{He} + {}^{4}_{2}\mathrm{He} + {}^{1}_{0}n. \]

The lithium target subjected to bombardment was placed in a vacuum inside the chamber with mica windows, which is visible at the center of the photograph under the tube from which the stream of fast deuterons emerges downward.

(Photograph by P. I. Dee and E. T. S. Walton).

with \(^{6}\mathrm{Li}\), or with \(^{7}\mathrm{Li}\). When bombarded with protons, from time to time some proton penetrates into the \(^{7}\mathrm{Li}\) nucleus and is captured by it. The resulting \(^{8}\mathrm{Be}\) nucleus is unstable and immediately disintegrates into 2 fast \(\alpha\)-particles, ejected in almost opposite directions. A scheme of this type of transformation is shown in Fig. 17. When a proton is captured by a \(^{6}\mathrm{Li}\) nucleus, \(^{7}\mathrm{Be}\) is formed, which disintegrates into an \(\alpha\)-particle and a helium isotope of mass 3 (\(^{3}\mathrm{He}\)). If the bombardment is carried out not with protons, but with deuterons, then capture of a deuteron by a \(^{6}\mathrm{Li}\) nucleus again leads to the formation of a \(^{8}\mathrm{Be}\) nucleus, but with a larger excess of energy. This nucleus, as in the preceding case, disintegrates into two \(\alpha\)-particles which, however, have a greater velocity than the \(\alpha\)-particles formed when a proton is captured by a \(^{7}\mathrm{Li}\) nucleus. Moreover, these particles are, with one exception, the fastest among the \(\alpha\)-particles observed in all transformations, since their range in air is \(13\ \mathrm{cm}\). When a deuteron is captured by a \(^{7}\mathrm{Li}\) nucleus, \(^{9}\mathrm{Be}\) is formed, which at once disintegrates into three component parts—two \(\alpha\)-particles and a neutron.

Fig. 17. Scheme illustrating the transformation of the lithium isotope of mass 7 under proton bombardment. For every hundred million protons with an energy of 200,000 V, one lithium nucleus is transformed into two helium nuclei. The range of the resulting \(\alpha\)-particles is about \(8.4\ \mathrm{cm}\).

I shall mention here only some of the most important types of transformation of the two lithium isotopes. Table 1 clearly demonstrates the wide variety of transformations that occur when lithium is bombarded by various particles.

The transformations of lithium can be excellently illustrated by photographing the tracks of the particles produced in these transformations in a Wilson chamber. Fig. 18 shows one such photograph, obtained by Dee and Walton, in which the paths of the \(\alpha\)-particles formed in the transformation of lithium by deuterons are visible. In many photographs obtained by this method, the appearance of a pair of particles flying off in almost opposite directions is clearly noticeable.

The transformation of boron \(^{11}\mathrm{B}\) under proton bombardment has been the subject of prolonged study. In this case a \(^{12}\mathrm{C}\) nucleus is formed, which disintegrates into three \(\alpha\)-particles. Dee and Gilbert showed that the principal type of this transformation passes through two phases. First an \(\alpha\)-particle is emitted and a residual \(^{8}\mathrm{Be}\) nucleus is formed, containing an excess of energy, and then, after a very short interval of time, this nucleus splits into two \(\alpha\)-particles. Owing to technical difficulties it is rarely possible to obtain on a photographic plate the tracks of all three \(\alpha\)-particles formed in one ...

transformations. A beautiful photograph of the tracks of such a triplet of $\alpha$-particles is shown in Fig. 19. All three particle tracks, as was to be expected, lie in one plane, and their total kinetic energy corresponds to the energy liberated in the reaction. The transformations of ${}^{10}\mathrm{B}$ and ${}^{11}\mathrm{B}$ under deuteron bombardment are very complex and proceed with the emission of groups of protons with different velocities, as well as of $\alpha$-particles.

An account even of the main results obtained in the bombardment of all the elements by fast particles of various kinds would take too much time. However, I should like to dwell on a few cases of transformation that are of outstanding interest. The simplest possible case of transformation occurs in the bombardment of deuterium ${}^{2}\mathrm{D}$ by deuterons. The union of these two particles should lead to the formation of a ${}^{4}\mathrm{He}$ nucleus, but with a very large excess of energy. This nucleus immediately disintegrates in one of two equally probable ways, shown in Fig. 20. In one case the nucleus splits into a fast proton and a hydrogen isotope of mass 3 (${}^{3}\mathrm{H}$), and in the other into a fast neutron and a helium nucleus of mass 3 (${}^{3}\mathrm{He}$). If the energy of the bombarding deuterons is small in comparison with the energy released in the transformation, then the two particles in both cases will be ejected in almost opposite directions. This is clearly visible in the photograph (Fig. 21), where the long tracks belong to protons, and the much shorter ones to ${}^{3}\mathrm{H}$ nuclei.

Fig. 20. Diagram of the splitting of deuterium under deuteron bombardment. About 2 million deuterons with an energy of 100,000 V, bombarding a target of pure deuterium, produce one transformation by one or the other of the two indicated reactions.

The transformations described can be detected if the deuteron is accelerated by a voltage of only 20 thousand V, but the quantity of substance produced in the transformation, of course, rapidly increases

To the article by E. Rutherford

Fig. 19. A typical case of the disintegration of boron into three α-particles under proton bombardment

\[ \left({}^{11}_{5}\mathrm{B}+{}^{1}_{1}\mathrm{H}\to 3\,{}^{4}_{2}\mathrm{He}\right). \]

In the center of the photograph there is visible the boron target in the form of a thin line, surrounded by a white sphere formed as a result of the scattering of protons from the bombarding beam of α-particles. \(A\) and \(B\) were ejected in almost opposite directions, while the third α-particle \(C\) received a very small store of energy and barely emerged beyond the limits of the beam of scattered protons.

Fig. 21. Three examples of the emission of particles \({}^{1}_{1}\mathrm{H}\) and \({}^{3}_{1}\mathrm{H}\) in opposite directions when a thin target containing deuterium is bombarded by artificially accelerated deuterons

\[ \left({}^{2}_{1}\mathrm{H}+{}^{2}_{1}\mathrm{H}\to {}^{1}_{1}\mathrm{H}+{}^{3}_{1}\mathrm{H}\right). \]

The tracks of the particles \({}^{3}_{1}\mathrm{H}\), visible to the left of the tube with the target, have a range of \(1.6\ \mathrm{cm}\), whereas the length of the range of the protons emitted in the opposite direction is \(15\ \mathrm{cm}\).

with an increase in the energy of the bombarding deuterons. These types of transformations are the most effective of all transformations known to us that occur at low energy of the bombarding particles; they provide us, for experimental purposes, with a powerful homogeneous source of neutrons with an energy of 2.4 million V and a homogeneous group of protons with an energy of about 3 million V.

These interesting transformations, belonging to the simplest type of all possible ones, were first studied by Oliphant and Harteck and led to the discovery of a new isotope of hydrogen with mass 3 and of a new isotope of helium, likewise with mass 3.

The masses of these two isotopes can be computed exactly if the energy released in the transformation is known. It may not be uninteresting to carry out these computations here for the case of \(^{3}\mathrm{H}\). If the law of conservation of energy is obeyed, then the following relation must hold for the masses of the nuclei:

\[ {}^{2}\mathrm{H} + {}^{2}\mathrm{H} = {}^{1}\mathrm{H} + {}^{3}\mathrm{H} + E, \]

\[ 2.0147 + 2.0147 = 1.0081 + {}^{3}\mathrm{H} + 0.0042, \]

where \(E\) denotes the mass equivalent of the energy released in the transformation. The value of \(E\) is calculated from the observed range of the protons in air, amounting to 14.70 cm, which corresponds to an energy of 2.98 million V. From the law of conservation of momentum, which must hold in the splitting, it follows that \(3/4\) of all the energy liberated is released in the form of the kinetic energy of the proton. Thus the total liberated energy \(E\) is 3.97 million V. According to Einstein’s theory, mass and energy are equivalent to one another, and a decrease in the mass of the system by \(dm\) is equivalent to the release of energy in the amount \(c^{2}dm\), where \(c\) is the speed of light. The validity of this relation has been confirmed for a number of cases in which the masses of the atoms participating in the transformation were precisely known. The release of energy in the amount of 3.97 million V is equivalent to a decrease in mass by 0.0042 atomic units. The equation given above remains valid also in the case in which, in both parts of it, the masses of the atoms are substituted for the masses of the nuclei. Under the chemical symbols of the elements are indicated the values of the atomic weights of hydrogen and deuterium found by Aston with the aid of a mass spectrograph. In order for the masses of the elements on both sides of the equation to balance, it is evidently necessary that the mass of \(^{3}\mathrm{H}\) be equal to 3.0171.

In an analogous manner, by determining the energy of the fast neutron emitted in another type of transformation, it was found that the mass of \(^{3}\mathrm{He}\) is equal to 3.0171, i.e., within the accuracy of the measurements, it coincides with the mass of \(^{3}\mathrm{H}\). We have sufficient grounds to suppose that the values of the atomic weights calculated in this way correspond to reality. It is known, for example, that \(^{3}\mathrm{He}\) is also formed when \(^{6}\mathrm{Li}\) is bombarded with protons (Table 1), and the mass of \(^{3}\mathrm{He}\), calculated from this reaction, agrees with that given above.

TABLE I

Transformations of lithium under bombardment by protons \(\left({}^{1}_{1}\mathrm H\right)\), neutrons \(\left({}^{1}_{0}\mathrm n\right)\), deuterons \(\left({}^{2}_{1}\mathrm H\right)\), and \(\alpha\)-particles.

In these reactions there are formed protons, neutrons, atoms of \({}^{3}_{1}\mathrm H\), \({}^{3}_{2}\mathrm{He}\), \({}^{4}_{2}\mathrm{He}\), \({}^{8}_{3}\mathrm{Li}\), \({}^{8}_{4}\mathrm{Be}\), \({}^{10}_{5}\mathrm{B}\), and \(\gamma\)-rays.

The lithium isotope \({}^{8}\mathrm{Li}\) enclosed in a square is radioactive and has a half-life of 0.8 sec. It decays with the emission of fast \(\beta\)-particles. It is interesting to note that, when \({}^{7}\mathrm{Li}\) is bombarded by protons, intense \(\gamma\)-rays with a quantum energy of 17 million V are also emitted—the greatest so far observed in transformations.

Finally, I should like to dwell briefly on several important discoveries made by Prof. Lawrence and his collaborators, who used the cyclotron to obtain very fast deuterons with an energy of 6 million V. When bismuth is bombarded by such fast deuterons, a radioactive isotope of bismuth is formed which in all respects is identical with the well-known radioactive substance radium E. The radioactive substance obtained from bismuth not only emits \(\beta\)-particles and decays with exactly the same period as radium E, but in doing so forms an \(\alpha\)-particle-emitting substance identical with polonium (radium F). The mechanism of the transformation is apparently the following:

\[ {}^{209}_{83}\mathrm{Bi}+{}^{2}_{1}\mathrm{D}\to{}^{210}_{83}\mathrm{Bi}+{}^{1}_{1}\mathrm{H}. \]

As indicated in Fig. 1, radium E is an isotope of bismuth.

with atomic weight 210. This proof of the production of one of the natural radioactive substances by artificial means is of enormous interest and has outstanding significance.

I must mention one more transformation, which may prove very valuable from the technical point of view. When sodium with atomic weight 23 (or ordinary table salt) is bombarded with fast deuterons, a radioactive isotope of sodium with atomic weight 24—radiosodium—is formed, with the emission of a proton. Radiosodium decays with the emission of a β-particle and forms a stable magnesium nucleus with atomic weight 24. The half-life of radiosodium is 15 hours. Along with the β-particle, each radiosodium nucleus apparently emits γ-rays of high energy, possessing the same penetrating power as the γ-rays emitted by radium in equilibrium with the products of its decay. Lawrence has already succeeded in obtaining in this way a powerful source of γ-rays from radiosodium, approximately equivalent in this respect to 1 g of radium. Thus it is possible that such an artificially created source of γ-rays may someday serve as a substitute for radium for therapeutic purposes.

Transformation by Means of γ-Rays

Up to the present time, the bombardment of matter by fast particles has remained the most effective method for studying the transformation of elements, although we have already seen that in the case of heavier elements slow neutrons prove to be extremely effective. In some cases, however, we may count on producing the transformation of elements by applying γ-radiation of high quantum energy. Chadwick and Goldhaber have recently succeeded, by means of γ-rays, in splitting deuterium \({}^{2}\mathrm{D}\) into a proton and a neutron.

In this case the energy of the radiation quantum must be greater than the binding energy of the proton and neutron, which amounts to about 2.3 million V. In a similar way Szilard found that beryllium of mass 9 decays into \({}^{8}\mathrm{Be}\) and a neutron when irradiated with γ-rays of energy slightly greater than 1 million V. This new method of transformation may prove effective in other cases as well, if only we succeed in obtaining sufficiently intense sources of γ-rays of high quantum energy.

General Conclusions

During the last few years progress in the field of our knowledge of the transformations of elements has proceeded very rapidly, and it has been shown that almost all elements can be subjected to transformation by means of the appropriate actions. It is interesting to note that there is sufficiently convincing evidence for the possibility of obtaining an isotope of gold by bombarding platinum

by fast neutrons; but it is not yet clear which particular isotope of platinum takes part in the transformation. In the course of this work more than 50 new radioactive substances were discovered. They are unstable isotopes of elements, perhaps once existing on the sun, but having disappeared as the earth cooled. In all probability, uranium and thorium alone have survived from the whole large group of radioactive elements only because their half-life periods are great in comparison with the age of the earth. Although much work still remains to be done in order to clarify in all details the nature of many transformations, enough data have already been accumulated to indicate that, by means of bombardment with the fast particles at our disposal, it is possible to bring about a large number of the most varied transformations. For elements having many isotopes, the number of possible transformations must be very large. In general it has been established that all conceivable transformations of elements take place which are compatible with the conservation of nuclear charge, and also with the law of conservation of energy, if changes of mass are taken into account. However, the frequency of the various types of transformation of elements may vary within wide limits. In most transformations the unstable nucleus splits into two particles, emitting γ-rays in the process; but several cases are known of transformations of lighter elements in which the exploding nucleus breaks up into three particles. In the course of this work several new, stable isotopes were discovered, in particular $^{3}\mathrm{H}$, $^{3}\mathrm{He}$, $^{8}\mathrm{Be}$, $^{10}\mathrm{Be}$, previously not observed in nature. In addition, as we have already seen, a number of new elementary particles were discovered, including the neutron, the proton, the α-particle, and the positive electron.

Transformations of elements usually occur on an insignificant scale, and only rarely can the quantity of substance formed in a transformation be weighed or seen. However, our methods of detecting and recognizing the moving particles formed in transformations are so extraordinarily sensitive that even the most insignificant transformation in scale produces a very appreciable effect in our measuring instruments. The reliability of our methods for detecting and analyzing fast particles is in many cases higher than the reliability of ordinary chemical methods, even if one assumes that the amount of transformed substance would be large enough to subject it to chemical analysis.

Generally speaking, the amount of substance formed in a transformation by bombardment of a thick layer of an element increases rapidly with increasing energy of the bombarding particles. In some cases no noticeable transformation is observed until the energy of the particle reaches a definite value, and then the scale of the transformation begins to grow rapidly as the energy of the particles is increased above this level. In the case of light elements, such as, for example, deuterium and lithium, the transformation becomes

noticeable when the bombarding particles have an energy of 20 thousand V or even less. But in most cases the energy of the bombarding particles required to produce a noticeable transformation must be higher than this value and increases rapidly as the atomic number of the element being bombarded rises.

We have seen that these considerations are not applicable to neutrons, since in many cases the scale of the transformation is greatest when the neutrons possess a small energy, of the order of fractions of a volt, comparable with the energy of the thermal motion of molecules. There is also reason to suppose that every slow neutron passing, for example, through boron ultimately brings about the transformation of a boron nucleus (see p. 42). It also seems probable that the independent existence of the neutron in our atmosphere must be very short-lived, since it would soon be captured by the nuclei of nitrogen and oxygen, thereby producing the transformation of these elements. Therefore we should not expect any appreciable accumulation of neutrons in our atmosphere over the course of centuries.

We have already spoken of the release of large quantities of energy in the spontaneous transformation of atoms of natural radioactive substances. In some cases of artificial transformation by means of protons and deuterons, the amount of energy emitted per atom is even greater than in the case of radioactive transformations. For example, the energy liberated in the transformation of 1 atom of lithium 6 by deuterons is 22.5 million V, i.e., almost twice as great as the energy emitted in the decay of any radioactive atom. Since the transformation can be produced by a deuteron with an energy of only 20 thousand V, it is clear that in this individual process there is a large gain in energy. But, on the other hand, only one deuteron out of \(10^8\) proves effective, so that in the final account far more energy is expended than is emitted in the transformation. Even if one takes into account that the overall efficiency of the process increases with an increase in the energy of the bombarding particles, there remains little hope of obtaining useful energy from atoms in this way. The exceptional effectiveness of slow neutrons, which cause transformations of individual elements with the release of large quantities of energy, at first glance seems very promising in this respect. However, one must not lose sight of the fact that the neutrons themselves can be obtained only as the result of an extremely disadvantageous process of transformation. The prospects for obtaining useful energy from atoms by means of artificial transformation are thus not favorable.

The atomic nucleus is a whole world, in which, in an exceedingly small volume, there is enclosed a whole series of different particles, in particular protons and neutrons, held together by exceedingly powerful forces unknown to us. At the present time, energetic

attempts to adapt existing ideas to the explanation of the structure of the atomic nucleus, and in a few simple cases some successes have already been achieved. Yet we are still far from understanding the structure of a complex nucleus and the causes that bring about its disintegration under definite conditions. Whereas wave mechanics has proved adequate for explaining the outer electronic structure of the atom, where the electrons are sufficiently distant from one another, this theory cannot be applied with confidence to the complex nucleus, where there is such an exceptional concentration of massive particles in a very small space. To overcome these difficulties, Bohr proposed a more general approach to the problem, in which the nucleus is regarded as an aggregate of indistinguishable particles, capable of oscillating as a whole and possessing strictly definite energy levels. There are many considerations in favor of this new point of view, and its prospects for the future seem more favorable. The large amount of experimental data on the transformation of elements accumulated up to the present should prove to be a very useful support in solving this most difficult and fundamental problem.

The information we have acquired about the transformation of elements may also be of great help in another direction. It is obvious that inside an incandescent star, such as our sun, where the temperature is very high, protons, neutrons, and other light particles must have velocities of thermal motion sufficiently great to produce transformations of the elements of the sun. Under conditions of such continuous bombardment, there must occur a constant process of building new atoms and splitting others, and within a short time a state of equilibrium must have been reached—at least a temporary one. From what we know about the relative abundance of the elements on the earth, we can form a fairly accurate idea also of what the average composition of the sun was 3,000 million years ago, when the earth separated from the sun. When our knowledge in the field of the transformations of elements advances still further, we shall be able to establish what accounts for the relative abundance of certain elements on our earth, and why elements with even atomic numbers occur, on average, much more often than those with odd ones. Thus we see that the progress of modern alchemy will greatly expand our knowledge not only of the elements themselves, but also of their relative abundance in our universe.

1. In this way it is convenient to express the kinetic energy of fast particles. For example, if an electron moves in a vacuum between two points with a potential difference of 1 million V, then by the end it acquires an energy of 1 million eV, or, in abbreviated form, 1 million V. ↩