Popular Science Monthly/Volume 67/May 1905/Present Problems in Radioactivity

by Ernest Rutherford

Layout 4

THE

POPULAR SCIENCE

MONTHLY

 

---

MAY. 1905.

---

PRESENT PROBLEMS IN RADIOACTIVITY.^[1]

By Professor E. RUTHERFORD,

MCGILL UNIVERSITY.

SINCE the initial discovery by Becquerel of the spontaneous emission of new types of radiation from uranium, our knowledge of the phenomena exhibited by uranium and the other radioactive bodies has grown with great and ever increasing rapidity, and a very large mass of experimental facts has now been accumulated. It would be impossible within the limits of this article to review even briefly the more important experimental facts connected with the subject and, in addition, such a review is rendered unnecessary by the recent publication of several treatises^[2] in which the main facts of radioactivity have been dealt with in a fairly complete manner.

In the present article an attempt will be made to discuss the more important problems that have arisen during the development of the subject and to indicate what, in the opinion of the writer, are the subjects which will call for further investigation in the immediate future.

 

Nature of the Radiations.

The characteristic radiations from the radioactive bodies are very complex, and a large amount of investigation has been necessary to isolate the different kinds of rays and to determine their specific characters. The rays from the three most studied radio-elements, uranium, thorium and radium, can be separated into three distinct types, known as the $\alpha$, $\beta$ and $\gamma$ rays.

The nature of the $\alpha$ and $\beta$ rays has been deduced from observations of the deflection of the path of the rays by a magnetic and electric field. According to the electromagnetic theory, a radiation which is deflectable by a magnetic or electric field must consist of a flight of charged particles. If the amount of deflection of the rays from their path is measured when both a magnetic and an electric field of known strength are applied, the value of the velocity of the particles and the ratio $e/m$ of the charge $e$ carried by the particle to its apparent mass $m$, can be determined. From the direction of the deviation, the sign of the electric charge carried by the particle can be deduced.

Examined in this way, the $\beta$ rays have been shown to consist of negatively changed particles projected with a velocity approaching that of light. The experiments of Becquerel and Kaufmann have shown that the $\beta$ rays are identical with the cathode rays produced in a vacuum tube. This relationship has been established by showing that the value of $e/m$ is the same for the two kinds of rays. In both cases the value of $e/m$ has been found to be about 10⁷ electromagnetic units, while the corresponding value of $e/m$ for hydrogen atoms set free in the electrolysis of water is 10⁴. If the charge on the $\beta$ particle—or electron, as it has been termed—is the same as that carried by the hydrogen atom, this result shows that the apparent mass of the electron at slow speeds is about 1/1000 of that of the hydrogen atom. The $\beta$ particles from the radio-elements are expelled with a much greater speed than the cathode ray particles in a vacuum tube. The velocities of the $\beta$ particles from radium are not all the same, but vary between 10¹⁰ and $3\times 10^{10}$ cms. per second. The swifter particles move with a velocity of at least 95 per cent, of that of light. The emission by radium of electrons with high but different velocities has been utilized by Kaufmann to determine the variation of $e/m$ with speed. He found that the value of $e/m$ decreased with increase of velocity, showing that the apparent mass increased with the speed. By comparison of the experimental results with the mathematical theory of a moving charge, he deduced that the mass of the electrons was in all probability electromagnetic in origin, i. e., the apparent mass could be explained purely in terms of electricity in motion without the necessity of a material nucleus on which the charge was distributed. J. J. Thomson, Heaviside and others have shown that a moving charged sphere increases in apparent mass with the speed and that, for speeds small compared with the velocity of light, the increase of mass $m=2c^{2}/3a$ where $e$ is the charge carried by the body and a the radius of the conducting sphere over which the electricity is distributed. Kaufmann deduced that the value of $e/m=1.86\times 10^{7}$ for electrons of slow velocity. If the mass of the electrons is electrical in origin, it is seen that $a=10^{-13}$ cms., since the value of $e=3.4\times 10^{10}$ electrostatic units. The results of various methods of determination agree in fixing the diameter of an atom as about 10^—8 cms. The apparent diameter of an electron is thus minute compared with that of the atom itself.

The highest velocity of the radium electrons measured by Kaufmann was, as we have seen, 95 per cent, of the velocity of light. The power that electrons have of penetrating solid matter increases rapidly with the velocity, and some of those expelled from radium are able to penetrate through more than 3 mms. of lead. It is probable that a few of the electrons from radium move with a velocity still greater than the highest value observed by Kaufmann, and it is important to determine the value of $e/m$ and the velocity of such electrons. According to the mathematical theory, the mass of the electron increases rapidly as the speed of light is approached and should be infinitely great when the velocity of light is reached. This leads to the conclusion that no charged body can be made to move with a velocity greater than that of light. This result is of great importance and requires further experimental verification. A close study of the high speed electrons from radium may throw further light on this question.

Only a brief statement of our knowledge of electrons has been given in this paper. A more complete and detailed account of both theory and experiment will be given by my colleague, Dr. Langevin, in his address on 'Physics of the Electron.'

 

The a Rays.

The $\beta$ rays are readily deflected by a magnetic field, but a very intense magnetic field is required to deflect appreciably the $\alpha$ rays. The writer showed by the electric method that the rays of radium were deflected both by a magnetic and electric field, and deduced the velocity of projection of the particles and the ratio, $e/m$, of the charge to the mass. The direction of deflection of the a rays is opposite in sense to that of the $\beta$ rays. Since the $\beta$ rays carry a negative charge, the $\alpha$ particles thus behave as if they carried a positive charge. The magnetic deflection of these rays was confirmed by Becquerel and Des Coudres, using the photographic method, while the latter, in addition, showed their deflection in an electric field and deduced the value of the velocity and $e/m$. The values obtained by Rutherford and Des Coudres were in very good agreement, considering the difficulty of obtaining a measurable deviation.

Observer.	Value of Velocity.	Value of e/m.
Rutherford	$\scriptstyle 2.5\times 10^{9}$ cms. per sec.	$\scriptstyle 6\times 10^{3}$ electromagnetic units.
Des Coudres	$\scriptstyle 1.6\times 10^{9}$ cms. per sec.	$\scriptstyle 6\times 10^{3}$ electromagnetic units.

Now the value $e/m$ for the hydrogen atom is 10⁴. On the assumption that the $\alpha$ particle carries the same charge as the hydrogen atom, this result shows that the apparent mass of the a particle is about twice that of the hydrogen atom. If the $\alpha$ particle consists of any known kind of matter, this result indicates that it is the atom either of hydrogen or of helium. The $\alpha$ particles thus consist of heavy bodies projected with great velocity, whose mass is of the same order of magnitude as the helium atom and at least 2,000 times as great as the apparent mass of the $\beta$ particle or electron.

If the a particles carry a positive charge, it is to be expected that the particles, falling on a body of sufficient thickness to absorb them will, under suitable conditions, give it a positive charge, while the substance from which they are projected acquires a negative charge. The corresponding effect has been observed for the $\beta$ rays. The $\beta$ particles from radium communicate a negative charge to the body on which they fall, while the radium from which they are emitted acquires a positive charge. This effect has been very strikingly shown by a simple experiment of Strutt. The radium compound, sealed in a small glass tube, the outer surface of which is made conducting, is insulated by a quartz rod. A simple gold leaf electroscope is attached to the bottom of the glass tube, in order to indicate the presence of a charge. The whole apparatus is enclosed in a glass vessel, which is exhausted to a high vacuum, in order to reduce the loss of charge in consequence of the ionization of the gas by the rays. Using a few milligrams of radium bromide, the gold leaf diverges to its full extent in a few minutes and shows a positive charge. The explanation is simple. A large proportion of the negatively charged particles are projected through the glass tube containing the radium and a positive charge is left behind. By allowing the gold leaf, when extended, to touch a conductor connected to earth, the gradual divergence of the leaves and their collapse becomes automatic and will continue, if not indefinitely, at any rate for as long a time as the radium lasts.

When the radium is exposed under similar conditions, but unscreened in order to allow the a particles to escape, no such charging action is observed. This is not due to the equality between the number of positively and negatively charged particles expelled from the radium, for no effect is observed when the radium is temporarily freed from its power of emitting $\beta$ rays by driving off the emanation by heat. The writer recently attempted to detect the charge carried by the a rays from radium by allowing them to fall on an insulated plate in a vacuum, but no appreciable charging was observed. The $\beta$ rays were temporarily got rid of by heating the radium in order to drive off its emanation. There was found to be a strong ionization set up at the surface from which the rays emerged and the surface on which they impinged. The presence of this ionization causes the upper plate to rapidly lose a charge communicated to it. Although this action would mask to some extent the effect to be looked for, a measurable difference should have been obtained under the experimental conditions, if the a rays were expelled with a positive charge; but not the slightest evidence of a charge was observed. I understand that similar negative results have been obtained by other observers.

This apparent absence of charge carried by the $\alpha$ rays is very remarkable and difficult to account for. There is no doubt that the a particles behave as if they carried a positive charge, for several observers have shown that the a rays are deflected by a magnetic field. It is interesting to notice, in this connection, that Villard was unable to detect that the 'canal rays' carried a charge. These rays, discovered by Goldstein, are analogous in many respects to the $\alpha$ rays. They are slightly deflected by a magnetic and electric field and behave like positively charged bodies atomic in size. The value of $e/m$ is not a constant, but depends upon the nature of the gas in the tube through which the discharge is passed. The apparent absence of charge on the a particles may possibly be explained on the supposition that a negatively charged particle (an electron) is always projected at the same time as the positively charged particle. Such electrons, if they are present, should be readily bent back to the surface from which they came by the action of a strong magnetic field. It will be of interest to examine whether the charge carried by the $\alpha$ rays can be detected under such conditions.^[3] Another hypothesis, which has some points in its favor, is that the a particles are uncharged at the moment of their expulsion, but, in consequence of their collision with the molecules of matter, lose a negative electron and consequently acquire a positive charge. This point is at present under examination. The question is in a very unsatisfactory state and requires further investigation.

It is remarkable that positive electricity is always associated with matter atomic in size, for no evidence has been obtained of the existence of a positive electron corresponding to the negative electron. This difference between positive and negative electricity is apparently fundamental, and no explanation of it has as yet been forthcoming.

The evidence that the $\alpha$ particles are atomic in size mainly rests on the deflection of the path of the rays in a strong magnetic and electric field. It has, however, been suggested by H. A. Wilson that the $\alpha$ particle may in reality be a 'positive' electron, whose magnitude is minute compared with that of the negative. The electric mass of an electron for slow speeds is equal to $2e^{2}/3a$. Since there is every reason to believe that the charge carried by the $\alpha$ particle and the electron are the same, in order that the mass of the positive electron should be about 2,000 times that of the negative, it would be necessary to suppose that the radius of the sphere over which the charge is distributed is only 1/2000 of that of the electron, i. e., about 10^—10 cms. The magnetic and electric deflection would be equally well explained on this view. This hypothesis, while interesting, is too far reaching in its consequences to be accepted before some definite experimental evidence is forthcoming to support it. The evidence at present obtained strongly supports the view that the $\alpha$ particles are in reality projected matter atomic in size. The probability that the a particle is an atom of helium is discussed later.

Becquerel showed that the $\alpha$ rays of polonium were deflected by a magnetic field to about the same extent as the $\alpha$ rays of radium. On account of the feeble activity of thorium and uranium, compared with radium and polonium, it has not been found possible to examine whether the rays emitted by them are deflectable. There is little doubt, however, that the particles of all the radio-elements are projected matter of the same kind (probably helium atoms). The $\alpha$ rays from the different radioactive products differ in their power of penetrating matter in the proportion of about three to one, being greatest for the a rays from the imparted or 'induced' activity of radium and thorium, and least for uranium. This difference is probably mainly due to a variation of the velocity of projection of the a particles in the various cases. The interpretation of results is rendered difficult by our ignorance of the mechanism of absorption of the a rays by matter. Further experiment^[4] on this point is very much required.

It is of importance to settle whether the a particles of radium and polonium have the same ratio $e/m$. Becquerel states that the amount of curvature of the $\alpha$ rays from polonium in a field of constant strength was the same as for the $\alpha$ rays from radium. This would show that the product of the mass and velocity is the same for the $\alpha$ particles from the two substances. The $\alpha$ rays of polonium, however, certainly have less penetrating power than those of radium, and presumably a smaller velocity of projection. This result would indicate that $e/m$ is different for the a particles of polonium and radium. It is of importance to determine accurately the ratio of $e/m$ and the velocity for the rays penetrating two substances in order to settle this vital point.

 

The $\gamma$ Rays.

In addition to the $\alpha$ and $\beta$ rays, uranium, thorium and radium all emit very penetrating rays, known as $\gamma$ rays. These rays are about 100 times as penetrating as the $\beta$ rays and their presence can be detected after passing through several centimeters of lead. Villard, who originally discovered these rays in radium, stated that they were not deflected in a magnetic field, and this result has been confirmed by other observers. Quite recently, Paschen has described some experiments which led him to believe that the ty rays are corpuscular in character, consisting of negatively charged particles (electrons) projected with a velocity very nearly equal to that of light. This conclusion is based on the following evidence: Some pure radium bromide was completely enclosed in a lead envelope 1 cm. thick—a thickness sufficient to completely absorb the ordinary $\beta$ rays emitted by radium, but which allows about half of the $\gamma$ rays to escape. The lead envelope was insulated in an exhausted vessel and was found to gain a positive charge. In another experiment, the rays escaping from the lead envelope fell on an insulated metal ring, surrounding it. When the air was exhausted, this outer ring was found to gain a negative charge. These experiments, at first sight, indicate that the $\gamma$ rays carry with them a negative charge like the $\beta$ rays. In order to account for the absence of deflection of the path of the $\gamma$ rays in very strong magnetic or electric fields, it is necessary to suppose that the particles have a very large apparent mass. Paschen supposes that the $\gamma$ rays are negative electrons like the $\beta$ rays, but are projected with a velocity so nearly equal to that of light, that their apparent mass is very great.

Some experiments recently made by Mr. Eve, of McGill University, are of great interest in this connection. He found by the electric method that the $\gamma$ rays set up secondary rays, in all directions, at the surface from which they emerge and also on the surface on which they impinge. These rays are of much less penetrating power than the primary rays and are readily deflected by a magnetic field. The direction of deflection indicated that these secondary rays consisted, for the most part, of negatively charged particles (electrons) projected with sufficient velocity to penetrate through about 1 mm. of lead. In the light of these results, the experiments of Paschen receive a simple explanation without the necessity of assuming that the $\gamma$ rays of radium themselves carry a negative charge. The lead envelope in his experiment acquired a positive charge in consequence of the emission of a secondary radiation consisting of negatively charged particles, projected with great velocity from the surface of the lead. The electric charge acquired by the metal ring was due to the absorption of these secondary rays by it, and the diminution of this charge in a magnetic field was due to the ease with which these secondary rays are deflected. It is thus to be expected that the envelope surrounding the radium, whether made of lead or other metal, will always acquire a positive charge, provided the metal is not of sufficient thickness to absorb all the $\gamma$ rays in their passage through it.

No conclusive evidence has yet been brought forward to show that the $\gamma$ rays can be deflected either in a magnetic or an electric field. In this, as in other respects, the rays are very analogous to the Röntgen rays.

According to the theory of Stokes, J. J. Thomson and Weichert, Röntgen rays are transverse pulses set up in the ether by the sudden arrest of the motion of the cathode particles on striking an obstacle. The more sudden the stoppage the shorter is the pulse, and the rays, in consequence, have greater power of penetrating matter. In some recent experiments Barkla found that the secondary rays set up by the Röntgen rays, on striking an obstacle, vary in intensity with the orientation of the vacuum tube, showing that the Röntgen rays exhibit the property of one-sidedness or polarization. This is the only evidence so far obtained in direct support of the wave nature of the Röntgen rays.

If Röntgen rays are not set up when the cathode particles are stopped, conversely, it is to be expected that Röntgen rays will be set up when they are suddenly expelled. Now this effect is not observable in an X-ray tube, since the cathode particles acquire most of their velocity, not at the cathode itself, but in passing through the electric field between the cathode and anticathode. It isy however, to be expected theoretically that a type of Röntgen rays should be set up at the sudden expulsion of the $\beta$ particles from the radio atoms. The rays, too, should be of a very penetrating kind, since not only are the charged particles projected with a speed approaching that of light, but the change of motion must occur in a distance comparable with the diameter of an atom.

On this view the $\gamma$ rays are a very penetrating type of Röntgen rays, having their origin at the moment of the expulsion of the $\beta$ particle from the atom. If the $\beta$ particle is the parent of the $\gamma$ rays the intensity of the $\beta$ and $\gamma$ rays should, under all conditions, be proportional to one another. I have found this to be the case, for the $\gamma$ rays always accompany the $\beta$ rays and, in whatever way the $\beta$ ray activity varies, the activity measured by the $\gamma$ rays always varies in the same proportion. Active matter which does not emit $\beta$ rays does not give rise to $\gamma$ rays. For example, the radio tellurium of Marckwald, which does not emit $\beta$ rays, does not give off $\gamma$ rays.

Certain differences are observed, however, in the ionizing action of $\gamma$ and X rays. For example, gases and vapors like chlorine, sulphuretted hydrogen, methyl-iodide and chloroform, when exposed to ordinary X rays, show a much greater ionization, compared with air, than is to be expected according to the density law. On the other hand, the relative ionization of these substances by $\gamma$ rays follows the density law very closely. It seemed likely that this apparent difference between the two types of rays was due mainly to the greater penetrating power of the $\gamma$ rays. This was confirmed by some recent experiments of Eve, who found that the relative conductivity of gases exposed to very penetrating Röntgen rays from a hard tube approximated in most cases closely to that observed for the $\gamma$ rays. The vapor of methyl-iodide was an exception, but the difference in this case would probably disappear if X rays could be generated of the same penetrating power as that of the $\gamma$ rays.

Thus the results so far obtained generally support the view that the $\gamma$ rays are a type of penetrating X rays. This view is in agreement too with theory, for it is to be expected that very penetrating $\gamma$ rays will always appear with the $\beta$ rays.

No evidence of the emission of a type of Röntgen rays is observed from active bodies which emit only $\alpha$ rays. If the a particles are initially projected with a positive charge, such rays are to be expected. Their absence supplies another piece of evidence in support of the view that the a particle is projected without a charge but acquires a positive charge in its passage through matter.^[5]

 

Emission of Energy by the Radioactive Bodies.

It was early recognized that a very active substance like radium emitted energy at a rapid rate, but the amount of this energy was strikingly shown by the direct measurements of its heating effect made by Curie and Laborde. They found that one gram of radium in radioactive equilibrium emitted about 100 gram calories of heat per hour. A gram of radium would thus emit 876,000 gram calories per year, or over 200 times as much heat as is liberated by the explosion of hydrogen and oxygen to form one gram of water. They showed that the rate of heat emission was the same in solution as in the solid state, and remained constant when once the radium had reached a stage of radioactive equilibrium. Curie and Dewar showed that the rate of evolution of heat from radium was unaltered by plunging the radium into liquid air or liquid hydrogen.

It seemed probable that the evolution of heat by radium was directly connected with its radioactivity and the experiments of Rutherford and Barnes proved this to be the case. The heating effect of a quantity of radium bromide was first determined. The emanation was then completely driven off by heating the radium, and condensed in a small glass tube by means of liquid air. After removal of the emanation, the heat evolution of the radium in the course of about three hours fell to a minimum corresponding to one quarter of its original value, and then slowly increased again, reaching its original value after an interval of about one month. The heat emission from the emanation tube at first increased with the time, rising to a maximum value about three hours after its introduction. It then slowly decreased according to an exponential law with the time, falling to half value in about four days.

The curve expressing the recovery from its minimum of the heating effect of radium is complementary to the curve expressing the diminution of the heating effect of the emanation tube with time. The curves of decay and recovery agree within the limit of experimental error with the corresponding curves of decay and recovery of the activity of radium when measured by the α rays. Since the minimum activity of radium, measured by the a rays, after the emanation has been removed is only one quarter of the maximum activity, these results indicate that the heating effect of radium is proportional to its activity measured by the α rays. It is not proportional to the activity measured by the β or γ rays, since the β or γ ray activity of radium almost completely disappears some hours after removal of the emanation.

These results have been confirmed by further observations of the distribution of the heat emission between the emanation and the successive products which arise from it. If the emanation is left for several hours in a closed tube, its activity measured by the electric method increases to about twice its initial value. This is due to the 'excited activity' or in other words to the radiations from the active matter deposited on the walls of the tube by the emanation. The activity of this deposit has been very carefully analyzed, and the results show that the matter deposited by the emanation breaks up in three successive and well marked stages. For convenience, these successive products of the emanation will be termed radium A, radium B and radium C. The time T taken for each of these products to be half transformed and the radiations from each product are shown in the following table:

Product.	T		Radiations.
Radium,			α rays.
Emanation,	4	days,	α rays.
Radium A,	3	mins.,	α rays.
Radium B,	21	mins.,	no rays.
Rαdium C,	28	mins.,	α, β and γ rays.

When the emanation has been left in a closed vessel for several hours, the emanation and its successive products reach a stage of approximate radioactive equilibrium, and the heating effect is then a maximum. If the emanation is suddenly removed from the tube by a current of air, the heating effect is then due to radium A, B and C together. On account, however, of the rapidity of the change of radium A (half value in three minutes) it is experimentally very difficult to distinguish between the heating effect of the emanation and that of radium A. The curve of variation with time of the heating effect of the tube after removal of the emanation is very nearly the same as the corresponding curve for the activity measured by the a rays. These results show that each of the products of radium supplies an amount of heat roughly proportional to its a ray activity. Each product loses its heating effect at the same rate as it loses its activity, showing that the mission of heat is directly connected with the radioactive changes. The results indicated that the product, radium B, which does not emit rays, does not supply an amount of heat comparable with the other products. This point is important and requires more direct verification.

Since the heat emission is in all cases nearly proportional to the number of a particles expelled, the question arises whether the bombardment of these particles is sufficient to account for the heating effects observed. The kinetic energy of the α particle can be at once determined, since e/m and v are known.

The following table shows the kinetic energy of the a particle deduced from the measurements of Eutherford and Des Coudres. The third column shows the number of a particles expelled from 1 gram of radium per second on the assumption that the heating effect of radium (100 gram calories per gram per hour) is entirely due to the energy given out by the expelled α particles.

Observer.	Kinetic.	Number of Particles expelled per Second from 1 Gram of Radium.
Rutherford	$\scriptstyle 5.9\times 10^{-6}$ ergs	$\scriptstyle 2\times 10^{11}$
Des Coudres	$\scriptstyle 2.5\times 10^{-6}$ ergs	$\scriptstyle 5\times 10^{11}$

This hypothesis that the heating effect of radium is due to bombardment of the α particle can be indirectly put to the test in the following way. It seems probable that each atom of radium in breaking up emits one α particle. On the disintegration theory, the residue of the atom, after the a particle is expelled, is the atom of the emanation, so that each atom of radium gives rise to one atom of the emanation. Let q be the number of atoms in each gram of radium breaking up per second. When a state of radioactive equilibrium is reached the number N of emanation particles present is given by $N=q/\lambda$ where λ is the constant of change of the emanation. Now Ramsay and Soddy deduced from experiment that the volume of the emanation released from 1 gram of radium was about one cubic millimeter at atmospheric pressure and temperature. It has been experimentally deduced that there are $3.6\times 10^{19}$ molecules in one cubic centimeter of gas at ordinary pressure and temperature. The emanation obeys Boyle's law and behaves, in all respects, like a heavy gas, and we may in consequence deduce, since $N=3.6\times 10^{16}$ and $\lambda =2.0\times 10^{-6}$, the value $q=7.2\times 10^{10}$.-Now the particles expelled from radium in a state of radioactive equilibrium are about equally divided between four substances, viz., the radium itself, the emanation, radium A and radium C. We may thus conclude that the number of α particles expelled per second from 1 gram of radium in radioactive equilibrium is $2.9\times 10^{11}$. The value deduced by this method is intermediate between the values previously obtained (see previous table), on the assumption that the heating effect is entirely due to the a particles.

I think we may conclude from the agreement of these two methods of calculation that the greater portion of the heating effect of radium is a direct result of the bombardment of the expelled a particles, and that, in all probability, about $5\times 10^{10}$ atoms of radium break up per second.^[6]

The energy carried off in the form of β and γ rays is small compared with that emitted in the form of α rays. By calculation it can be shown that the average kinetic energy of the β particle is small in comparison with that of the a particle. This is confirmed by comparative measurements of the total ionization produced by the α and β rays, when the energy of the rays is all used up in ionizing the gas, for the total ionization produced by the β rays is small compared with that due to the γ rays. The total ionization produced by the γ rays is about the same as that produced by the β rays, showing that, in all probability, the energy emitted in the form of these two types of radiation is about the same. From the point of view of the energy radiated and of the changes which occur in the radioactive bodies, the α rays thus play a far more important rôle in radioactivity than the β or γ rays. Most of the products which arise from radium and thorium emit only α rays, while the β and γ rays appear only in the last of the series of rapid changes which take place in these bodies.

Since most of the heating effect of radium is due to the α rays, it is to be expected that all radioactive substances which emit them will also emit heat at a rate proportional to their α ray activity. On this view, both uranium and thorium should emit heat at about one millionth the rate of radium. It is of importance to determine directly the heating effect for these substances and also for actinium and radio-tellurium.

According to the disintegration theory, the α particle is expelled as a result of the disintegration of the atom of radioactive matter. While it is to be expected that a greater portion of the energy emitted will be carried off in the form of kinetic energy by the expelled particles, it is also to be expected that some energy will be radiated in consequence of the rearrangement of the components of the system after the violent ejection of one of its parts. No direct measurements have yet been made of the heating effect of the α particles independently of the substance in which they are produced. Experiments of this character would be difficult, but they would throw light on the important question of the division of the radiated energy between the expelled α ray particle and the system from which it arises.

The enormous emission of energy by the radioactive substances is very well illustrated by the case of the radium emanation. The emanation released from 1 gram of radium in radioactive equilibrium emits during its changes an amount of energy corresponding to about 10,000 gram calories. Now Ramsay and Soddy have shown that the volume of this emanation is about 1 nubic millimeter at standard pressure and temperature. One cubic millimeter of the emanation and its product thus emits about 10⁷ gram calories. Since 1 c.c. of hydrogen, in uniting with the proportion of oxygen required to form water, emits 3.1 gram calories, it is seen that the emanation emits about three million times as much energy as an equal volume of hydrogen.

It can readily be calculated on the assumption that the atom of the emanation has a mass 100 times that of hydrogen, that I pound of the emanation some time after removal could emit energy at the rate of about 8,000 horse-power. This would fall off in a geometrical progression with the time, but, on an average, the amount of energy emitted during its life corresponds to 50,000 horse-power days. Since the radium is being continuously transformed into emanation, and three-quarters of the total heat emission is due to the emanation and its products, a simple calculation shows that 1 gram of radium must emit during its life about 10⁹ gram calories. As we have seen, the heat emission of radium is about equally divided between the radium itself and the three other α ray products which come from it. The heat emitted from each of the other radioactive substances, while their activity lasts, should be of the same order of magnitude, but in the case of uranium and thorium the present rate of heat emission will probably continue, on an average, for a period of about 1000 million years.

 

Source of the Energy emitted by the Radioactive Bodies.

There has been considerable difference of opinion in regard to the fundamental question of the origin of the energy spontaneously emitted from the radioactive bodies. Some have considered that the atoms of the radio-elements act as transformers of borrowed energy. The atoms are supposed, in some way, to abstract energy from the surrounding medium and to emit it again in the form of the characteristic radiations. Another theory which has found favor with a number of physicists supposes that the energy is derived from the radio-atoms themselves and is released in consequence of their disintegration. The latter theory involves the conception that the atoms of the radio-elements contain a great store of latent energy, which only manifests itself when the atom breaks up. There is no direct evidence in support of the view that the energy of the radio-elements is derived from external sources, while there is much indirect evidence against it. Some of this evidence will now be considered. There is now no doubt that the α and β rays consist of particles projected with great speed. In order that the α particle may acquire the velocity with which it is expelled, it can be calculated that it would be necessary for it to move freely between two points differing in potential by about five million volts. It is very difficult to imagine any mechanism, which could suddenly impress such an enormous velocity on one of the parts of an atom. It seems much more reasonable to suppose that the aa and β particles were originally in rapid motion in the atom and, for some reason, escaped from the atomic system with the velocity they possessed at the instant of their release. There is now undeniable evidence that radioactivity is always accompanied by the production of new kinds of active matter. Some sort of chemical theory is thus required to explain the facts whether the view is taken that the energy is derived from the atom itself or from external sources. The 'external' theory of the origin of the energy was initially advanced to explain only the heat emission of radium. We have seen that this is undoubtedly connected with the expulsion of α particles from the different disintegration products of radium, and that the radium itself only supplies one quarter of the total heat emission, the rest being derived from the emanation and its further products. On such a theory it is necessary to suppose that in radium there are a number of different active substances, whose power of absorbing external energy dies away with the time, at different but definite rates. This still leaves the fundamental difficulty of the origin of these radioactive products unexplained. Unless there is some unknown source of energy in the medium which the radioactive bodies are capable of absorbing, it is difficult to imagine whence the energy demanded by the external theory can be derived. It certainly can not be from the air itself, for radium gives out heat inside an ice calorimeter. It can not be any type of rays such as the radioactive bodies emit, for the radioactivity of radium, and consequently its heating effect is unaltered by hermetically sealing it in a vessel of lead several inches thick. The evidence, as a whole, is strongly against the theory that the energy is borrowed from external sources and, unless a number of improbable assumptions are made, such a theory is quite inadequate to explain the experimental facts. On the other hand, the disintegration theory, advanced by Rutherford and Soddy, not only offers a satisfactory explanation of the origin of the energy emitted by the radio-elements, but also accounts for the succession of radioactive bodies. On this theory, a definite, small proportion of the atoms of radioactive matter every second become unstable and break up with explosive violence. In most cases, the explosion is accompanied by the expulsion of an α particle; in a few cases by only a β particle, and in others by α and β particles together. On this view, there is at any time present in a radioactive body a proportion of the original matter which is unchanged and the products of the part which has undergone change. In the case of a slowly changing substance like radium, this point of view is in agreement with the observed fact that the spectrum of radium remains unchanged with its age.

The expulsion of an α or β particle, or both, from the atom leaves behind an atom which is lighter than before and which has different chemical and physical properties. This atom in turn becomes unstable and breaks up, and the process, once started, proceeds from stage to stage with a definite and measurable velocity in each case.

The energy radiated is, on this view, derived at the expense of the internal energy of the radio-atoms themselves. It does not contradict the principle of the conservation of energy, for the internal energy of the products of the changes, when the process of change has come to an end, is supposed to be diminished by the amount of energy emitted during the changes. This theory supposes that there is a great store of internal energy in the radio-atoms themselves. This is not in disagreement with the modern views of the electronic constitution of matter, which have been so ably developed by J. J. Thomson, Larmor and Lorentz. A simple calculation shows that the mere concentration of the electric charges, which on the electronic theory are supposed to be contained in an atom, implies a store of energy in the atom so enormous that, in comparison, the large evolution of energy from the radio-element is quite insignificant.

Since the energy emitted from the radio-elements is for the most part kinetic in form, it is necessary to suppose that the α and β particles were originally in rapid motion in the atoms from which they are projected. The disintegration theory supposes that it is the atoms and not the molecules which break up. Such a view is necessary to explain the independence of the rate of disintegration of radioactive matter, of wide variations of temperature, and of the action of chemical and physical agents at our command. This must be conceded if the term atom is used in the ordinary chemical sense. It is, however, probable that the atoms of the radio-elements are in reality complex aggregates of known or unknown kinds of matter, which break up spontaneously. This aggregate behaves like an atom and can not be resolved into simpler forms by external chemical or physical agencies. It breaks up, however, spontaneously with an evolution of energy enormous compared with that released in ordinary chemical changes. This question will be considered later.

The disintegration theory assumes that a small fraction of the atoms break up in unit time, but no definite explanation is, as yet, forthcoming of the causes which lead to this explosive disruption of the atom. The experimental results are equally in agreement with the view that each atom contains within itself the potentiality of its final disruption, or with the view that the disintegration is precipitated by the action of some external cause, that may lead to the disintegration of the atom, in the same way that a detonator is necessary to start certain explosions. The energy set free is, however, not derived from the detonator, but from the substance on which it acts. There is another general view which may possibly lead to an explanation of atomic disruption. If the atom is supposed to consist of electrons or charged bodies in rapid motion, it tends to radiate energy in the form of electromagnetic waves. If an atom is to be permanently stable, the parts of the atom must be so arranged that there is no loss of energy by electromagnetic radiation. J. J. Thomson has investigated certain possible arrangements of electrons in an atom which radiate energy extremely slowly, but which ultimately must break up in consequence of the loss of internal energy. According to present views, it is not such a matter of surprise that atoms do break up as that atoms are so stable as they appear to be. This question of the causes of disintegration is fundamental and no adequate explanation has yet been put forward.

 

Radioactive Products.

Rutherford and Soddy showed that the radioactivity was always accompanied by the appearance of new types of active matter which possessed physical and chemical properties distinct from the parent radio-element. The radioactivity of these products is not permanent, but decays according to an exponential law with the time. The activity I at any time t is given by $I_{t}=I_{0}e{-\lambda t}$, where I is the initial activity and A a constant. Each radioactive product has a definite change constant which distinguishes it from all other products. These products do not arise simultaneously, but in consequence of a succession of changes in the radio-elements; for example, thorium in breaking up gives rise to Th X, which behaves as a solid substance soluble in ammonia. This in turn breaks up and gives rise to a gaseous product, the thorium emanation. The emanation is again unstable and gives rise to another type of matter which behaves as a solid and is deposited on the surface of the vessel containing the emanation. It was found that the results would be quantitatively explained on the assumption that the activity of any product at any time is the measure of the rate of production of the next product. This is to be expected since the activity of any substance is proportional to the number of atoms which break up per second, and, since each atom in breaking up gives rise to one atom of the next product together with α or β particles, or both, the activity of the parent is a measure of the rate of production of the succeeding product.

Of these radioactive products, the radium emanation has been very closely studied on account of its existence in the gaseous state. It has been shown to be produced by radium at a constant rate. The amount of emanation stored up in a given mass of radium reaches a maximum value when the rate of supply of fresh emanation balances the rate of change of the emanation present.

If q be the number of atoms of emanation produced per second by the radium, and N the maximum number present when radioactive equilibrium is reached, then $N=q/\lambda$, where λ is the constant of change of the emanation. This relation has been verified experimentally. The emanation is found to diffuse through air like gas of heavy molecular weight. It is unattacked by chemical reagents and in that respect resembles the inert gases of the argon family. It condenses at a definite temperature —150° C. Its constant of change is unaffected between the limits of temperature of 450° C and  —180° C. Since the emanation changes into a non-volatile type of matter which is deposited on the surface of vessels, it was to be expected that the volume of the emanation would decrease according to the same law as it lost its activity. These deductions based on the theory have been confirmed in a striking manner by the experiments of Ramsay and Soddy. The radium emanation was chemically isolated and found to be a gas which obeys Boyle's law. The volume of the emanation observed was of the same order as had been predicted before its separation. The volume was found to decrease with the time according to the same law as the emanation lost its activity. Ramsay and Collie found that the emanation had a new and definite spectrum similar in some respects to that of the argon group of gases.

There can thus be no doubt that the emanation is a transition substance with remarkable properties. Chemically it behaves like an inert gas and has a definite spectrum and is condensed by cold. But, on the other hand, the gas is not permanent, but disappears, and is changed into other types of matter. It emits during its changes about one million times as much energy as is emitted during any known chemical change.

From the similarity of the behavior of the emanation of thorium and actinium to that of radium, we may safely conclude that these also are new gases which have only a limited life and change into other substances.

Fig. 1.

The non-volatile products of the radioactive bodies can be dissolved in strong acids and show definite chemical behavior in solution. They can be partially separated by electrolysis and by suitable chemical methods. They can be volatilized by the action of high temperature and their differences in this respect can be utilized to effect in many cases a partial separation of successive products. There can be little doubt that each of these radioactive products is a transition substance possessing, while it lasts, some definite chemical and physical properties which serve to distinguish it from other products and from the parent element.

The radioactive products derived from each radio-element together with the type of radiation emitted during their disintegration, are shown graphically in Fig. 1.

The radiations from actinium have not been so far examined sufficiently closely to determine the character of the radiation emitted by each product. There is some evidence that a product, actinium X, exists in actinium corresponding to Th X in thorium.^[7] It has not, however, been very closely examined.

The question of nomenclature for the successive products is important. The names Ur X, Th X have been retained and also the term emanation. The emanation from the three radio-elements in each case gives rise to a non-volatile type of matter which is deposited on the surface of the bodies. The matter initially deposited from the radium emanation is called radium A. Radium A changes into B and B into C, and so on. A similar nomenclature is applied to the further products of the emanation of thorium and actinium. This notation is simple and elastic and is very useful in mathematical discussion of the theory of successive changes. In the following table a list of the products is given, together with the nature of the radiation and the most marked chemical and physical properties of each product. The time T for each of the products to be half transformed is also added.

The changes which occur in the active deposits from the emanation of radium, thorium and actinium have been difficult to determine on account of their complexity. For example, in the case of radium, the active deposit obtained as a result of a long exposure to the emanation contains quantities of radium A, B and C. The changes occurring in the active deposit of radium have been determined by P. Curie, Danne and the writer. The value of T for the three successive changes is 3, 21 and 28 minutes, respectively. Radium A gives only a rays, B gives out no rays at all, while C gives out α, β and γ rays. These results have been deduced by the comparison of the change of activity with time, with the mathematical theory of successive changes. The variation of the activity with time depends upon whether the activity is measured by the α, β or γ rays. The complicated curves are very completely explained on the hypothesis of three successive changes of the character already mentioned.

The activity of a vessel in which the radium emanation has been stored for some time rapidly falls to a very small fraction after the emanation is withdrawn. However, there always remains a slight residual activity. The writer has recently examined the activity in

detail. The residual activity at first mainly consists of β rays, and the activity measured by them does not change appreciably during the period of one year. The α ray activity is at first small, but increases uniformly with the time for the first few months that the activity has been examined. These results receive an explanation on the hypothesis that radium C changes into a product D which emits only β rays. D changes into a product E which emits only α rays. This view has been confirmed by separating the α ray product by dipping bismuth-plate into the solution containing radium D and E. The probable period of these changes can be deduced from observations of the magnitude of the α and β ray activity at any time. It has been deduced that radium D is probably half transformed in 40 years and radium E is half transformed in about 1 year. The evidence at present obtained points to the conclusion that radium E is the active constituent present in Marckwald's radio-tellurium and probably also in the polonium of Mme. Curie.^[8]

The changes in the active deposit of thorium have been analyzed by the writer, and the corresponding changes in actinium by Miss Brooks.

The occurrence of a 'rayless change' in the active deposits from the emanation of radium, thorium and actinium is of great interest and importance. As these products do not emit either aa, β or γ rays, their presence can only be detected by their effect on the amount of the succeeding products. The action of the rayless change is most clearly brought out in the examination of the variation of activity with time of a body exposed for a very short interval in the presence of the emanations of thorium and actinium. Let us consider, for simplicity, the variation of activity with time for thorium. The activity (measured by the a rays) observed at first is very small, but gradually increases with the time, passes through a maximum and finally decays according to an exponential law with the time falling to half value in eleven hours. The shape of this curve can be completely explained on the assumption of the two successive changes, the second of which alone gives out rays. The matter deposited on the body during the short exposure consists almost entirely of thorium A. Thorium A changes into B and the breaking up of B gives rise to the activity measured.

If thorium A does not give out rays, the activity of the body at any time t after removal can be easily shown to be proportional to $e{\lambda _{2}t}-e{\lambda _{1}t}$, where $\lambda _{1}$, $\lambda _{2}$ are the constants of change of thorium A and B, respectively. Now the experimental curves of variation of activity are found to be accurately expressed by an equation of this form. A very interesting point arises in settling the values of $\lambda _{1}$, $\lambda _{2}$ corresponding to the two changes. It is seen that the equation is symmetrical in $\lambda _{1}$ and $\lambda _{2}$ and in consequence is unaltered if the values of $\lambda _{1}$ and $\lambda _{2}$ are interchanged. Now the constant of the change is determined by the observation that the activity finally decays to half value in 11 hours. The theoretical and experimental curves are found to coincide if one of the two products is half transformed in 11 hours and the other in 55 minutes. The comparison of the theoretical and experimental curves does not, however, allow us to settle whether the period of change of thorium A is 55 minutes or 11 hours. In order to settle the point, it is necessary to find some means of separating the products thorium A and B from each other. In the case of thorium, this is done by electrolyzing a solution of thorium. Pegram obtained an active product which decayed according to an exponential law with the time falling to half value in a little less than one hour. This result shows that the radiating product thorium B has the shorter period. In a similar way, by recourse to electrolysis, it has been found that the change actinium B has a period of 1.5 minutes. In the case of radium, P. Curie and Danne utilized the difference in volatility of radium B and C in order to fix the period of the changes.

It is very remarkable that the third successive product of radium, thorium and actinium should not give out rays. It seems probable that these rayless changes are not of so violent a character as the other changes, and consist either of a rearrangement of the components of the atom or of an expulsion of an α or β particle with so slow a velocity that it fails to ionize the gas. The appearance of such changes in radioactive matter suggests the possibility that ordinary matter may also be undergoing slow 'rayless changes' for such changes can not be detected in the radio-elements unless the succeeding products emit rays.

It is seen that the changes occurring in radium, thorium and actinium are of a very analogous character and indicate that each of these bodies has a very similar atomic constitution.

While there can be no doubt that numerous kinds of radioactive matter with distinct chemical and physical properties are produced in the radio-elements, it is very difficult to obtain direct evidence in some cases that the products are successive and not simultaneous. This is the case for products which have either a very slow or very rapid rate of change compared with the other product. For example, it is difficult to show directly that radium B is the product of radium A and not the direct product of the emanation. In the same way, there is no direct evidence that radium C is the parent of radium D. At the same time, the successive nature of these products is indicated by indirect evidence.

There can be little doubt that each of the radioactive products is a distinct chemical substance and possesses some distinguishing physical or chemical properties. There still remains a large amount of chemical work to be done in comparing and arranging the chemical properties of these products and in determining whether the successive products follow any definite law of variation. The electrolytic method can in many cases be used to find the position of the product in the electrochemical series. The products which change most rapidly are present in the least quantity in radium and pitchblende. Only the slower changing products like the radium emanation and radium D and E exist in sufficient quantities to be examined by the balance. It is possible that the products radium A, B and C may be obtained in sufficient quantity to obtain their spectrum.

 

Connection between the a Particles and Helium.

The discovery of Ramsay and Soddy that helium was produced by the radium emanation was one of the greatest interest and importance, and confirmed in a striking manner the disintegration theory of radioactivity, for the possible production of helium from radioactive matter had been predicted on this theory before the experimental evidence was forthcoming. Ramsay and Soddy found that the presence of helium could not be detected in a tube immediately after the introduction of the emanation, but was observed some time afterwards, showing that the helium arose in consequence of a slow change in the emanation itself or in its further products.

The question of the origin of the helium produced by the radium emanation and its connection with the radioactive changes occurring in the emanation is one of the greatest importance. The experimental evidence so far obtained does not suffice to give a definite answer to this question, but suggests the probable explanation. There has been a tendency to assume that helium is the final disintegration product of the radium emanation, i. e., it is the inactive substance which remains when the succession of radioactive changes in the emanation has come to an end. There is no evidence in support of such a conclusion, while there is much indirect evidence against it. It has been shown that the emanation which breaks up undergoes three fairly rapid transformations; but after these changes have occurred, the residual matter—radium D—is still radioactive and breaks up slowly, being half transformed in probably about 40 years. There then occurs a still further change. Taking into account the minute quantity of the radium emanation initially present in the emanation tube, the amount of the final inactive product would be insignificant after the lapse of a few days or even months. Thus it does not seem probable that the helium can be the final product of the radioactive changes. In addition, it has been shown that the a particle behaves like a body of about the same mass as the helium atom. The expulsion of a few a particles from each of the heavy atoms of radium would not diminish the atomic weight of the residue very greatly. The atomic weight of the atoms of radium D and E is in all probability of the order of 200, since the evidence supports the conclusion that each atom expels one a particle at each transformation.

In order to explain the presence of helium, it is necessary to look to the other inactive products produced during the radioactive changes. The α particles expelled from the radioactive product are themselves non-radioactive. The measurement of the ratio e/m shows that they have an apparent mass intermediate between that of the hydrogen and helium atoms. If the a particles consist of any known kind of matter they must be atoms either of hydrogen or of helium. The actual value of e/m has not yet been determined with an accuracy sufficient to give a definite answer to the question. On account of the very slight curvature of the path of the a particles in a strong magnetic or electric field, accurate determination of e/m is beset with great difficulties. The experimental problem is still further complicated by the fact that the a particles escaping from a mass of radium have not all the same velocity and in consequence it is difficult to draw a definite conclusion from the observed deviation of the complex pencil of rays.

The results so far obtained are not inconsistent with the view that the a particles are helium atoms, and indeed it is difficult to escape from such a conclusion. On such a view, the helium, which is gradually produced in the emanation tube, is due to the collection of a particles expelled during the disintegration of the emanation and its further products. This conclusion is supported by evidence of another character. It is known that thorium minerals like monazite sand contain a large quantity of helium. In this respect they do not differ from uranium minerals which are rich in radium. The only common product of the different radioactive substances is the a particle and the occurrence of helium in all radioactive minerals is most simply explained on the supposition that the a particle is a projected helium atom. This conclusion could be indirectly tested by examining whether helium is produced in other substances besides radium, for example, in actinium and polonium.

The experimental determination of the origin of helium is beset with great difficulty on all sides. If the a particle is a helium atom, the total volume of helium produced in an emanation tube should be three times the initial volume of the emanation present, since the emanation in its rapid changes gives rise to three products each of which emits α particles. This is based on the assumption, which seems to be fulfilled by the experiments, that each atom of each product in breaking up expels one a particle. This at first sight offers a simple experimental means of settling the question, but a difficulty arises in accurately determining the volume of helium produced by a known quantity of the radium emanation. It would be expected that, if the emanation were isolated in a tube and left to stand, the volume of gas in the tube should increase with time in consequence of the liberation of helium. In one case, however, Ramsay and Soddy observed an exactly opposite result. The volume diminished with time to a small fraction of its original value. This diminution of volume was due to the decomposition of the emanation into a non-gaseous type of matter deposited on the walls of the tube, and followed the law of decrease to be expected in such a case, namely, the volume decreased according to an exponential law with the time falling to half value in four days. The helium produced by the emanation must have been absorbed by the walls of the tube. Such a result is to be expected if the particle is a helium atom, for the a particle is projected with a velocity sufficient to bury itself in the glass to a depth of about 1/100 mm. This buried helium would probably be in part released by the heating of the tube, such as occurs with the strong electric discharge employed in the spectroscopic detection of helium. Ramsay and Soddy have examined the glass tubes in which the emanation had been confined for some time to see if the buried helium was released by heat. In some cases traces of helium were observed.

Accurate measurements of the value of e/m for the a particle and also an accurate determination of the relative volume of the emanation and the helium produced by it would probably definitely settle this fundamental question.

Certain very important consequences follow on the assumption that the α particle is, in all cases, an atom of helium. It has already been shown that the radio-elements are transformed into a succession of new substances, most of which in breaking up emit an a particle. On such a view, the atom of radium, thorium, uranium and actinium must be supposed to be built up in part of helium atoms. In radium, at least five products of the change emit particles, so that the radium atom must contain at least five atoms of helium. In a similar way, the atoms of actinium and thorium (or, if thorium itself be not radioactive, the atom of the active substance present in it) must be compounds of helium. These compounds of helium are not stable, but spontaneously break up into a succession of substances, with an evolution of helium, the disintegration taking place at a definite but different rate at each stage. Such compounds are sharply distinguished in their behavior from the molecular compounds known to chemistry. In the first place, the radioactive compounds disintegrate spontaneously and at a rate that is independent of the physical or chemical forces at our control. Changes of temperature, which exert such a marked influence in altering the rate of molecular reactions, are here almost entirely without influence. But the most striking feature of the disintegration is the expulsion, in most cases, of a product of the change with very great velocity—a result never observed in ordinary chemical reactions. This entails an enormous liberation of energy during the change, the amount, in most cases, being about one million times as great as that observed in any known chemical reaction. In order to account for the expulsion of an α and a β particle with the observed velocities, it is necessary to suppose that the particles are in a state of rapid motion in the system from which they escape. Variation of temperature, in most cases, does not seem to affect the stability of the system.

It is well established that the property of radioactivity is inherent in the radio-atoms, since the activity of any radioactive compound depends only on the amount of the element present and is not affected by chemical treatment. As far as observation has gone, both uranium and radium behave as elements in the usually accepted chemical sense. They spontaneously break up but the rate of their disintegration seems to be, in most cases, quite independent of chemical control. In this respect, the radioactive bodies occupy a unique position. It seems reasonable to suppose that while the radioactive substances behave chemically as elements, they are, in reality, compounds of simpler kinds of matter, held together by much stronger forces than those which exist between the components of ordinary molecular compounds. Apart from the property of radioactivity, the radio-elements do not show any chemical properties to distinguish them from the non-radioactive elements except their very high atomic weight. The above considerations evidently suggest that the heavier inactive elements may also prove to be composite.

 

Origin of the Radio-elements.

We have seen that the radio-elements are continuously breaking up and giving rise to a succession of new substances. In the case of uranium and thorium, the disintegration proceeds at such a slow rate that in all probability a period of about 1,000 million years would be required before half the matter present is transformed. In the case of radium, however, where the process of disintegration proceeds at over one million times the rate in uranium and thorium, it is to be expected that a measurable proportion of the radium will be transformed in a single year. A quantity of radium left to itself must gradually disappear as such in consequence of its gradual transformation into other substances. This conclusion necessarily follows from the known experimental facts. The radium is being transformed continuously into the emanation which in turn is changed into other types of matter. Since there is no evidence that the process is reversible, all the raduim present must, in the course of time, be transformed into emanation. The rate at which radium is being transformed can be approximately calculated either from the number of α particles expelled per second or from the observed volume of the emanation produced per second. Both methods of calculation agree in showing that in a gram of radium about half a milligram is transformed per year. From analogy with other radioactive changes, it is to be expected that the rate of change of radium will be always proportional to the amount present. The amount of radium would thus decrease exponentially with the time falling to half value in about 1,000 years. On this view, radium behaves in a similar way to the other known products, the only difference being that its rate of change is slower. We have already seen that, in all probability, the product radium D is half transformed in about 40 years and radium E in about one year. In regard to their rate of change, the two substances radium D and E, which are half transformed in about 40 years and one year respectively occupy an intermediate position between the rapidly changing substances like radium A, B and C and the slowly changing parent substance radium.

If the earth were supposed to have been initially composed of pure radium, the activity 20,000 years later would not be greater than the activity observed in pitchblende to-day. Since there is no doubt that the earth is much older than this, in order to account for the existence of radium at all in the earth, it is necessary to suppose that radium is continuously produced from some other substance or substances. On this view, the present supply of radium represents a condition of approximate equilibrium where the rate of production of fresh radium balances the rate of transformation of the radium already present. In looking for a possible source of radium, it is natural to look to the substances which are always found associated with radium in pitchblende. Uranium and thorium both fulfill the conditions necessary to be a source of radium, for both are found associated with radium and both have a rate of change slow compared with radium. At the present time, uranium seems the most probable source of radium. The activity observed in a good specimen of pitchblende is about what is to be expected if uranium breaks up into radium. If uranium is the parent of radium, it is to be expected that the amount of radium present in different varieties of pitchblende obtained from different sources will always be proportional to the amount of uranium contained in the minerals. The recent experiments of Boltwood, Strutt and McKoy indicate that this is very approximately the case. It is not to be expected that the relation will always be very exact, since it is not improbable, in some cases, that a portion of the active material may be removed from the mineral by the action of percolating water or other chemical agencies. The results so far obtained strongly support the view that radium is a product of the disintegration of uranium. It should be possible to obtain direct evidence on this question by examining whether radium appears in uranium compounds which have been initially freed from radium. On account of the delicacy of the electric test of radium by means of its emanation, the question can be very readily put to experimental trial. This has been done for uranium by Soddy, and for thorium by the writer, but the results, so far obtained, are negative in character, although if radium were produced at the rate to be expected from theory, it should very readily have been detected.^[9] Such experiments, however, taken over a period of a few months are not decisive, for it is by no means improbable that the parent element may pass through several slow changes, possibly of a 'rayless' character, before it is transformed into radium. In such a case, if these intermediate products are removed by the same chemical process from the parent element, there may be a long period of apparent retardation before the radium appears. The considerations advanced to account for radium apply equally well to actinium, which, in all probability, when isolated will prove to be an element of the same order of activity as radium. The most important problem at present in the study of radioactive minerals is not the attempt to discover and isolate new radioactive substances, but to correlate these already discovered. Some progress has already been made in reducing the number of different radioactive substances and in indicating the origin of some of them. For example, there is no doubt that the 'emanating substance' of Giesel contains the same radioactive substance as the actinium of Debierne. In a similar way, there is very strong evidence that the active constituent in the polonium of Mme. Curie is identical with that in the radio-tellurium of Marckwald. The writer has recently shown that the active constituent in radio-tellurium or polonium is, in all probability, a disintegration product of radium (radium E). The same considerations apply to the radio-lead of Hofmann, which is probably identical with the product radium D. It still remains to be shown whether or not there is any direct family connection between the radioactive substances uranium, thorium, radium and actinium. It seems probable that some at least of these substances will prove to be lineal descendants of a single parent element, in the same way that the radium products are lineal descendants of radium. The subject is capable of direct attack by a combination of physical and chemical methods, and there is every probability that a fairly definite answer will soon be forthcoming.

 

Radioactivity of the Earth and Atmosphere.

It is now well established, notably by the work of Elster and Geitel, that radioactive matter is widely distributed both in the earth's crust and atmosphere. There is undoubtedly evidence of the presence of the radium emanation in the atmosphere, in spring water, and in air sucked up through the soil. It still remains to be settled whether the observed radioactivity of the earth's crust is due entirely to slight traces of the known radioactive elements or to new kinds of radioactive matter. It is not improbable that a close examination of the radioactivity of the different soils may lead to the discovery of radioactive substances which are not found in pitchblende or other radioactive minerals. The extraordinary delicacy of the electroscopic test of radioactivity renders it not only possible to detect the presence in inactive matter of extremely minute traces of a radioactive substance, but also in many cases to settle rapidly whether the radioactivity is due to one of the known radio-elements.

The observations of Elster and Geitel render it probable that the radioactivity observed in the atmosphere is due to the presence of radioactive emanations or gases, which are carried to the surface by the escape of underground water. Indeed it is difficult to avoid such a conclusion, since there is no evidence that any of the known constituents of the atmosphere are radioactive. Concurrently with observations of the radioactivity of the atmosphere, experiments have been made on the amount of ionization in the atmosphere itself. It is important to settle what part of this ionization is due to the presence of radioactive matter in the atmosphere. Comparisons of the relative amount of active matter and of the ionization in the atmosphere over land and sea will probably throw light on this important problem.

The wide distribution of radioactive matter in the soils which have so far been examined has raised the question whether the presence of radium and other radioactive matter in the earth, may not, in part at least, be responsible for the internal heat of the earth. It can readily be calculated that the presence of radium (or equivalent amounts of other kinds of radioactive matter) to the extent of about five parts in one hundred million million by mass would supply as much heat to the earth as is lost at present by conduction to its surface. It is certainly significant that, as far as observation has gone, the amount of radioactive matter present in the soil is of this order of magnitude.

The production of helium from radium indirectly suggests a means of calculating the age of the deposits of radioactive minerals. It seems reasonable to suppose that the helium always found associated with radioactive minerals is a product of the decomposition of the radioactive matter present. In the mineral fergusonite, for example, about half of the helium is removed by heating the mineral and the other half by solution. Thus it does not seem likely that much of the helium formed in the mineral escapes from it, so that the amount present represents the quantity produced since its formation. If the rate of the production of helium by radium (or other radioactive substance) is known, the age of the mineral can at once be estimated from the observed volume of helium stored in the mineral and the amount of radium present. All these factors have, however, not yet been determined with sufficient accuracy to make at present more than a rough estimate of the age of any particular mineral. An estimate of the rate of production of helium by radium has been made by Ramsay and Soddy by an indirect method. It can be deduced from their result that 1 gram of radium produces per year a volume of helium of about 25 cubic mms. at standard pressure and temperature. They, however, consider this to be an under estimate. On the other hand, if the particle is a helium atom, it can be calculated that 1 gram of radium produces per year about 200 cubic mms. of helium.

Let us consider, for example, the mineral fergusonite. Ramsay and Travers have shown that it yields about 1.8 c.c. of helium and contains about 7 per cent, of uranium. It can be deduced from known data that each gram of the mineral contains about one four-millionth of a gram of radium. Supposing that one gram of radium produces ¹⁄₅ c.c. of helium per year, the age of the mineral is readily seen to be about 40 million years. If the above rate of production of helium is an overestimate, the time will be correspondingly longer. I think there is little doubt that, when the data required are known with accuracy, this method can be applied with considerable confidence to determine the age of the radioactive minerals.

 

Radioactivity of Ordinary Matter.

The property of radioactivity is exhibited to the most marked extent by the radioactive substances found in pitchblende, but it is natural to ask the question whether ordinary matter possesses this property to an appreciable degree. The experiments that have so far been made show conclusively that ordinary matter, if it possesses this property at all, does so to a minute extent compared with uranium. It has been found that all the matter that has so far been examined shows undoubted traces of radioactivity, but it is very difficult to show that the radioactivity observed is not due to a minute trace of known radioactive matter. Even with our extraordinarily delicate methods for the detection of radioactivity, the effects observed are so minute that a definite settlement of the question is experimentally very difficult. J. J. Thomson has recently given an account at the British Association meeting of the work done on this subject in the Cavendish Laboratory, and has brought forward experimental evidence that strongly supports the view that ordinary matter does show specific radioactivity. Different substances were found to give out radiations that differed in quality as well as in quantity. A promising beginning has already been made but a great deal of work still remains to be done before such an important conclusion can be considered to have been definitely established.

1. Address given to the International Congress of Arts and Science, St. Louis, 1904.
2. Mme. Curie, 'Thèse presentée à la Faculté des Sciences,' Paris, 1903. H. Becquerel, 'Recherches sur une propriété nouvelle de la matière,' Typographie de Firmin Didot et Cie, Paris, 1903. E. Rutherford, 'Radioactivity,' Cambridge University Press, 1904. F. Soddy, 'Radioactivity,' Electrician Co., London, 1904.
3. Recent experiments have confirmed this point of view. By the use of a strong magnetic field, to remove the slow moving electrons, the charge carried by the $\alpha$ rays has been detected both by J. J. Thomson and the writer.
4. Bragg and Kleeman (Phil. Mag., Dec, 1904) have recently attacked this question and have offered a very satisfactory explanation of the mechanism of the absorption of the $\alpha$ rays by matter.
5. Recent experiments indicate, however, that the $\alpha$ particles are charged at the moment of this expulsion.
6. By measuring the charge carried by the α rays, the writer (Nature, March 2, 1905) has recently deduced that $\scriptstyle 6.4\times 10^{10}$ atoms of radium break up per second.
7. Godlewski (Nature, Jan. 19, 1905) has recently separated actinium X. It is similar in chemical properties to thorium X and loses half of its activity in about 10 days.
8. The writer has recently found that radium E and radio-tellurium have identical rates of decay. Both lose half of their activity in 150 days. This result shows that the active substance in radio-tellurium is a transformation product of radium.
9. In a recent letter to Nature, Soddy states that he has found that there is a slow growth of radium in a uranium solution. A similar result has been noted by Whetham.