Radio-activity/Chapter 6

by Ernest Rutherford

 

CHAPTER VI.

CONTINUOUS PRODUCTION OF RADIO-ACTIVE MATTER.

126. An account will now be given of some experiments which have thrown much light, not only on the nature of the processes which serve to maintain the radio-activity of the radio-active bodies, but also on the source of the energy continuously emitted by those bodies. In this chapter, for simplicity, the radio-activity of uranium and thorium will alone be considered, for it will be seen later that the changes taking place in these two substances are typical of those which occur in all radio-active substances.

We have seen (section 23) that there is some doubt whether the radio-activity of thorium is due to that element itself, or to an unknown radio-active constituent associated with it. This uncertainty, however, will present no serious difficulty when we are discussing the radio-activity of thorium, for the general conclusions are, for the most part, independent of whether thorium is the primary radio-active constituent or not. For simplicity, however, it will be assumed for the present that the radio-activity is due to thorium itself. If future research should definitely show that the radio-activity, ordinarily observed in thorium, is due to a new radio-active element mixed with it, the radio-active processes considered will refer to this new element.

127. Uranium X. The experiments of Mme Curie show that the radio-activity of uranium and radium is an atomic phenomenon. The activity of any uranium compound depends only on the amount of that element present, and is unaffected by its chemical combination with other substances, and is not appreciably affected by wide variations of temperature. It would thus seem probable, since the activity of uranium is a specific property of the element, that the activity could not be separated from it by chemical agencies.

In 1900, however, Sir William Crookes^[1] showed that, by a single chemical operation, uranium could be obtained photographically inactive while the whole of the activity could be concentrated in a small residue free from uranium. This residue, to which he gave the name of Ur X, was many hundred times more active photographically, weight for weight, than the uranium from which it had been separated. The method employed for this separation was to precipitate a solution of the uranium with ammonium carbonate. On dissolving the precipitate in an excess of the reagent, a light precipitate remained behind. This was filtered, and constituted the Ur X. The active substance Ur X was probably present in very small quantity, mixed with impurities derived from the uranium. No new lines were observed in its spectrum. A partial separation of the activity of uranium was also effected by another method. Crystallized uranium nitrate was dissolved in ether, when it was found that the uranium divided itself between the ether and water present in two unequal fractions. The small part dissolved in the water layer was found to contain practically all the activity when examined by the photographic method, while the other fraction was almost inactive. These results, taken by themselves, pointed very strongly to the conclusion that the activity of uranium was not due to the element itself, but to some other substance, associated with it, which had distinct chemical properties.

Results of a similar character were observed by Becquerel^[2]. It was found that barium could be made photographically very active by adding barium chloride to the uranium solution and precipitating the barium as sulphate. By a succession of precipitations the uranium was rendered photographically almost inactive, while the barium was strongly active.

The inactive uranium and the active barium were laid aside; but, on examining them a year later, it was found that the uranium had completely regained its activity, while that of the barium had completely disappeared. The loss of activity of uranium was thus only temporary in character. In the above experiments, the activity of uranium was examined by the photographic method. The photographic action produced by uranium is due almost entirely to the β rays. The α rays, in comparison, have little if any effect. Now the radiation from Ur X consists entirely of β rays, and is consequently photographically very active. If the activity of uranium had been measured electrically without any screen over it, the current observed would have been due very largely to the α rays, and little change would have been observed after the removal of Ur X, since only the constituent responsible for the β rays was removed. This important point is discussed in more detail in section 205. 128. Thorium X. Rutherford and Soddy^[3], working with thorium compounds, found that an intensely active constituent could be separated from thorium by a single chemical operation. If ammonia is added to a thorium solution, the thorium is precipitated, but a large amount of the activity is left behind in the filtrate, which is chemically free from thorium. This filtrate was evaporated to dryness, and the ammonium salts driven off by ignition. A small residue was obtained which, weight for weight, was in some cases several thousand times more active than the thorium from which it was obtained, while the activity of the precipitated thorium was reduced to less than one half of its original value. This active constituent was named Th X from analogy to Crookes' Ur X. The active residue was found to consist mainly of impurities from the thorium; the Th X could not be examined chemically, and probably was present only in minute quantity. It was also found that an active constituent could be partly separated from thorium oxide by shaking it with water for some time. On filtering the water, and evaporating down, a very active residue was obtained which was analogous in all respects to Th X. On examining the products a month later, it was found that the Th X was no longer active, while the thorium had completely regained its activity. A long series of measurements was then undertaken to examine the time-rate of these processes of decay and recovery of activity.

Fig. 47.

The results are shown graphically in Fig. 47, where the final activity of the thorium and the initial activity of the Th X are in each case taken as 100. The ordinates represent the activities determined by means of the ionization current, and the abscissae represent the time in days. It will be observed that both curves are irregular for the first two days. The activity of the Th X increased at first, while the activity of the thorium diminished. Disregarding these initial irregularities of the curves, which will be explained in detail in section 208, it will be seen that, after the first two days, the time taken for the thorium to recover half its lost activity is about equal to the time taken by the Th X to lose half its activity. This time in each case is about four days. The percentage proportion of the activity regained by the thorium, over any given interval, is approximately equal to the percentage proportion of the activity lost by the Th X during the same interval.

Fig. 48.

If the recovery curve is produced backwards to meet the vertical axis, it does so at a minimum of 25 per cent., and the above conclusions hold more accurately, if the recovery is assumed to start from this minimum. This is clearly shown by Fig. 48, where the percentages of activity recovered, reckoned from the 25 per cent. minimum, are plotted as ordinates. In the same figure the decay curve, after the second day, is shown on the same scale. The activity of the Th X decays with the time according to an exponential law, falling to half value in about four days. If I_{0} is the initial activity and I_{t} is the activity after a time t, then

I_{0}/I_{t} = e^{-λt},

where λ is a constant and e the natural base of logarithms. The

experimental curve of the rise of activity from a minimum to a maximum value is therefore expressed by the equation

I_{t}/I_{0} = 1 - e^{-λt},

where I_{0} is the amount of activity recovered when the state of constant activity is reached, I_{t} the activity recovered after a time t, and λ is the same constant as before.

129. Uranium X. Similar results were obtained when uranium was examined. The Ur X was separated by Becquerel's method of successive precipitations with barium. The decay of the separated activity and the recovery of the lost activity are shown graphically in Fig. 49. A more detailed discussion of this experiment is given in section 205.

Fig. 49.

The curves of decay and recovery exhibit the same peculiarities and can be expressed by the same equations as in the case of thorium. The time-rate of decay and recovery is, however, much slower than for thorium, the activity of the Ur X falling to half its value in about 22 days. A large number of results of a similar character have been obtained from other radio-active products, separated from the radio-elements, but the cases of thorium and uranium will suffice for the present to form a basis for the discussion of the processes that are taking place in radio-active bodies.

130. Theory of the phenomena. These processes of decay and recovery go on at exactly the same rate if the substances are removed from the neighbourhood of one another, or enclosed in lead, or placed in a vacuum tube. It is at first sight a remarkable phenomenon that the processes of decay and recovery should be so intimately connected, although there is no possibility of mutual interaction between them. These results, however, receive a complete explanation on the following hypotheses:

(1) That there is a constant rate of production of fresh radio-active matter by the radio-active body;

(2) That the activity of the matter so formed decreases according to an exponential law with the time from the moment of its formation.

Suppose that q_{0} particles of new matter are produced per second from a given mass of matter. The rate of emission of energy due to the particles produced in the time dt, is, at the moment of their formation, equal to Kq_{0}dt, where K is a constant.

It is required to find the activity due to the whole matter produced after the process has continued for a time T.

The activity dI, due to the matter produced during the time dt at the time t, decays according to an exponential law during the time T - t that elapses before its activity is estimated, and in consequence is given by

dI = Kq_{0}e^{-λ(T - t)}dt,

where λ is the constant of decay of activity of the active matter. The activity I_{T} due to the whole matter produced in the time T is thus given by

I_{T} = [integral]_{0}^T Kq_{0}e^{-λ(T - t)}dt
= (Kq_{0}/λ)(1 - e^{-λT}).

The activity reaches a maximum value I_{0} when T is very great,

and is then given by

I_{0} = Kq_{0}/λ;

thus I_{T}/I_{0} = 1 - e^{-λT}.

This equation agrees with the experimental results for the recovery of lost activity. Another method for obtaining this equation is given later in section 133.

A state of equilibrium is reached when the rate of loss of activity of the matter already produced is balanced by the activity supplied by the production of new active matter. According to this view, the radio-active bodies are undergoing change, but the activity remains constant owing to the action of two opposing processes. Now, if this active matter can at any time be separated from the substance in which it is produced, the decay of its activity, as a whole, should follow an exponential law with the time, since each portion of the matter decreases in activity according to an exponential law with the time, whatever its age may be. If I_{0} is the initial activity of the separated product, the activity I_{t} after an interval t is given by

I_{t}/I_{0} = e^{-λt}.

Thus, the two assumptions—of uniform production of active matter and of the decay of its activity in an exponential law from the moment of its formation—satisfactorily explain the relation between the curves of decay and recovery of activity.

131. Experimental evidence. It now remains to consider further experimental evidence in support of these hypotheses. The primary conception is that the radio-active bodies are able to produce from themselves matter of chemical properties different from those of the parent substance, and that this process goes on at a constant rate. This new matter initially possesses the property of activity, and loses it according to a definite law. The fact that a proportion of the activity of radium and thorium can be concentrated in small amounts of active matter like Th X or Ur X does not, of itself, prove directly that a material constituent responsible for the activity has been chemically separated. For example, in the case of the separation of Th X from thorium, it might be supposed that the non-thorium part of the solution is rendered temporarily active by its association with thorium, and that this property is retained through the processes of precipitation, evaporation, and ignition, and finally manifests itself in the residue remaining. According to this view it is to be expected that any precipitate capable of removing the thorium completely from its solution should yield active residues similar to those obtained from ammonia. No such case has, however, been observed. For example, when thorium nitrate is precipitated by sodium or ammonium carbonate, the residue from the filtrate after evaporation and ignition is free from activity and the thorium carbonate obtained has the normal amount of activity. In fact, ammonia is the only reagent yet found capable of completely separating Th X from thorium. A partial separation of the Th X can be made by shaking thorium oxide with water owing to the greater solubility of Th X in water.

Thorium and uranium behave quite differently with regard to the action of ammonia and ammonium carbonate. Ur X is completely precipitated with the uranium in an ammonia solution and the filtrate is inactive. Ur X is separated by ammonium carbonate, while Th X under the same conditions is completely precipitated with the thorium. The Ur X and the Th X thus behave like distinct types of matter with well-marked chemical properties quite distinct from those of the substances in which they are produced. The removal of Ur X by the precipitation of barium is probably not directly connected with the chemical properties of Ur X. The separation is probably due to the dragging down of the Ur X with the dense barium precipitate. Sir William Crookes found that the Ur X was dragged down by precipitates when no question of insolubility was involved, and such a result is to be expected if the Ur X exists in extremely minute quantity. It must be borne in mind that the actual amount of the active constituents Th X and Ur X, separated from thorium and uranium, is probably infinitesimal, and that the greater proportion of the residues is due to impurities present in the salt and the reagents, a very small amount of active matter being mixed with them.

132. Rate of production of Th X. If the recovery of the activity of uranium or thorium is due to the continuous production of new active matter, it should be possible to obtain experimental evidence of the process. As the case of thorium has been most fully investigated, a brief account will be given of some experiments made by Rutherford and Soddy^[4] to show that Th X is produced continuously at a constant rate. Preliminary experiments showed that three successive precipitations were sufficient to remove the Th X almost completely from the thorium. The general method employed was to precipitate a solution of 5 grams of thorium-nitrate with ammonia. The precipitate was then redissolved in nitric acid and the thorium again precipitated as before, as rapidly as possible, so that the Th X produced in the time between successive precipitations should not appreciably affect the results. The removal of the Th X was followed by measurements of the activity of the residues obtained from successive filtrates. In three successive precipitations the activities of the residues were proportional to 100, 8, 1·6 respectively. Thus two precipitations are nearly sufficient to free the thorium from Th X.

The thorium freed from Th X was then allowed to stand for a definite time, and the amount of Th X formed during that time found by precipitating it, and measuring its radio-activity. According to the theory, the activity I_{t} of the thorium formed in the time t is given by

I_{t}/I_{0} = 1 - e^{-λt},

where I_{0} is the total activity of Th X, when there is radio-active equilibrium.

If λt is small,

I_{t}/I_{0} = λt.

Since the activity of Th X falls to half value in 4 days, the value of λ expressed in hours = ·0072. After standing a period of 1 hour about 1/140, after 1 day 1/6, after 4 days 1/2 of the maximum should be obtained. The experimental results obtained showed an agreement, as good as could be expected, with the equation expressing the result that the Th X was being produced at a constant rate.

The thorium-nitrate which had been freed from Th X was allowed to stand for one month, and then it was again subjected to the same process. The activity of the Th X was found to be the same as that obtained from an equal amount of the original thorium-nitrate. In one month, therefore, the Th X had been regenerated, and had reached a maximum value. By leaving the thorium time to recover fully its activity, this process can be repeated indefinitely, and equal amounts of Th X are obtained at each precipitation. Ordinary commercial thorium-nitrate and the purest nitrate obtainable showed exactly the same action, and equal amounts of Th X could be obtained from equal weights. These processes thus appear to be independent of the chemical purity of the substance^[5].

The process of the production of Th X is continuous, and no alteration has been observed in the amount produced in the given time after repeated separations. After 23 precipitations extending over 9 days, the amount produced in a given interval was about the same as at the beginning of the process.

These results are all in agreement with the view that the Th X is being continuously produced from the thorium compound at a constant rate. The amount of active matter produced from 1 gram of thorium is probably extremely minute, but the electrical effects due to its activity are so large that the process of production can be followed after extremely short intervals. With a sensitive electrometer the amount of Th X produced per minute in 10 grams of thorium-nitrate gives a rapid movement to the electrometer needle. For larger intervals it is necessary to add additional capacity to the system to bring the effects within range of the instrument. 133. Rate of decay of activity. It has been shown that the activity of Ur X and Th X decays according to an exponential law with the time. This, we shall see later, is the general law of decay of activity in any type of active matter, obtained by itself, and freed from any secondary active products which it may, itself, produce. In any case, when this law is not fulfilled, it can be shown that the activity is due to the superposition of two or more effects, each of which decays in an exponential law with the time. The physical interpretation of this law still remains to be discussed.

It has been shown that in uranium and thorium compounds there is a continuous production of active matter which keeps the compound in radio-active equilibrium. The changes by which the active matter is produced must be chemical in nature, since the products of the action are different in chemical properties from the matter in which the changes take place. The activity of the products has afforded the means of following the changes occurring in them. It now remains to consider the connection between the activity at any time, and the amount of chemical change taking place at that time.

In the first place, it is found experimentally that the saturation ionization current i_{t}, after the active product has been allowed to decay for a time t, is given by

i_{t}/i_{0} = e^{-λt},

where i_{0} is the initial saturation current and λ the constant of decay.

Now the saturation current is a measure of the total number of ions produced per second in the testing vessel. It has already been shown that the α rays, which produce the greater proportion of ionization in the gas, consist of positively charged particles projected with great velocity. Suppose for simplicity that each atom of active matter, in the course of its change, gives rise to one projected α particle. Each α particle will produce a certain average number of ions in its path before it strikes the boundaries or is absorbed in the gas. Since the number of projected particles per second is equal to the number of atoms changing per second, the number of atoms n_{t} which change per second at the time t is given by

n_{t}/n_{0} = e^{-λt},

where n_{0} is the initial number which change per second. On this view, then, the law of decay expresses the result that the number of atoms changing in unit time, diminishes according to an exponential law with the time. The number of atoms N_{t} which remain unchanged after an interval t is given by

N_{t} = [integral]_{t}^[infinity] n_{t} dt
= (n_{0}/λ)e^{-λt}.

If N_{0} is the number of atoms at the beginning,

N_{0} = n_{0}/λ.
Thus N_{t}/N_{0} = e^{-λt} (1),

or the law of decay expresses the fact that the activity of a product at any time is proportional to the number of atoms which remain unchanged at that time.

This is the same as the law of mono-molecular change in chemistry, and expresses the fact that there is only one changing system. If the change depended on the mutual action of two systems, the law of decay would be different, since the rate of decay in that case would depend on the relative concentration of the two reacting substances. This is not so, for not a single case has yet been observed in which the law of decay was affected by the amount of active matter present.

From the above equation (1)

dN_{t}/dt = -λN_{t},

or the number of systems changing in unit time is proportional to the number unchanged at that time.

In the case of recovery of activity, after an active product has been removed, the number of systems changing in unit time, when radio-active equilibrium is produced, is equal to λN_{0}. This must be equal to the number q_{0} of new systems applied in unit time, or

q_{0} = λN_{0},
and λ = q_{0}/N_{0};

λ has thus a distinct physical meaning, and may be defined as the proportion of the total number of systems present which change per second. It has different values for different types of active matter, but is invariable for any particular type of matter. For this reason, λ will be termed the "radio-active constant" of the product.

We are now in a position to discuss with more physical definiteness the gradual growth of Th X in thorium, after the Th X has been completely removed from it. Let q_{0} particles of Th X be produced per second by the thorium, and let N be the number of particles of Th X present at any time t after the original Th X was removed. The number of particles of Th X which change every second is λN, where λ is the radio-active constant Th X. Now, at any time during the process of recovery, the rate of increase of the number of particles of Th X = the rate of production - the rate of change; that is

dN/dt = q_{0} - λN.

The solution of this equation is of the form N = ae^{-λt} + b, where a and b are constants.

Now when t is very great, the number of particles of Th X present reach a maximum value N_{0}.

Thus, since N = N_{0} when t = [infinity],
                               b = N_{0};

since N = 0 when t = 0,

          a + b = 0;
hence b = -a = N_{0},

and the equation becomes

N/N_{0} = 1 - e^{-λt}.

This is equivalent to the equation already obtained in section 130, since the intensity of the radiation is always proportional to the number of particles present.

134. Influence of conditions on the rate of decay. Since the activity of any product, at any time, may be taken as a measure of the rate at which chemical change takes place, it may be used as a means of determining the effect of conditions on the changes occurring in radio-active matter. If the rate of change should be accelerated or retarded, it is to be expected that the value of the radio-active constant λ will be increased or decreased, i.e. that the decay curve will be different under different conditions.

No such effect, however, has yet been observed in any case of radio-active change, where none of the active products produced are allowed to escape from the system. The rate of decay is unaltered by any chemical or physical agency, and in this respect the changes in radio-active matter are sharply distinguished from ordinary chemical changes. For example, the rate of decay of activity from any product takes place at the same rate when the substance is exposed to light as when it is kept in the dark, and at the same rate in a vacuum as in air or any other gas at atmospheric pressure. Its rate of decay is unaltered by surrounding the active matter by a thick layer of lead under conditions where no ordinary radiation from outside can affect it. The activity of the matter is unaffected by ignition or chemical treatment. The material giving rise to the activity can be dissolved in acid and re-obtained by evaporation of the solution without altering the activity. The rate of decay is the same whether the active matter is retained in the solid state or kept in solution. When a product has lost its activity, resolution or heat does not regenerate it, and as we shall see later, the rate of decay of the active products, so far examined, is the same at a red heat as at the temperature of liquid air. In fact, no variation of physical or chemical conditions has led to any observable difference in the decay of activity of any of the numerous types of active matter which have been examined.

135. Effect of conditions on the rate of recovery of activity. The recovery of the activity of a radio-element with time, when an active product is separated from it, is governed by the rate of production of fresh active matter and by the decay of activity of that already produced. Since the rate of decay of the activity of the separated product is independent of conditions, the rate of recovery of activity can be modified only by a change of the rate of production of fresh active matter. As far as experiments have gone, the rate of production, like the rate of decay, is independent of chemical or physical conditions. There are indeed certain cases which are apparent exceptions to this rule. For example, the escape of the radio-active emanations from thorium and radium is readily affected by heat, moisture and solution. A more thorough investigation, however, shows that the exception is only apparent and not real. These cases will be discussed more in detail in chapter VII, but it may be stated here that the differences observed are due to differences in the rate of escape of the emanations into the surrounding gas, and not to differences in the rate of production. For this reason it is difficult to test the question at issue in the case of the thorium compounds, which in most cases readily allow the emanation produced by them to escape into the air.

In order to show that the rate of production is independent of molecular state, temperature, etc., it is necessary in such a case to undertake a long series of measurements extending over the whole time of recovery. It is impossible to make accurate relative comparisons to see if the activity is altered by the conversion of one compound into another. The relative activity in such a case, when measured by spreading a definite weight of material uniformly on a metal plate, varies greatly with the physical conditions of the precipitate, although the total activity of two compounds may be the same.

The following method^[6] offers an accurate and simple means of studying whether the rate of production of active matter is influenced by molecular state. The substance is chemically converted into any compound required, care being taken that active products are recovered during the process. The new compound is then spread on a metal plate and compared with a standard sample of uranium for several days or weeks as required. If the rate of production of active matter is altered by the conversion, there should be an increase or decrease of activity to a new steady value, where the production of active matter is again balanced by the rate of decay. This method has the great advantage of being independent of the physical condition of the precipitate. It can be applied satisfactorily to a compound of thorium like the nitrate and the oxide which has been heated to a white heat, after which treatment only a slight amount of emanation escapes. The nitrate was converted into the oxide in a platinum crucible by treatment with sulphuric acid and ignition to a white heat. The oxide so obtained was spread on a plate, but no change of its activity was observed with time, showing that in this case the rate of production was independent of molecular state. This method, which is limited in the case of thorium, may be applied generally to the uranium compounds where the results are not complicated by the presence of an emanation.

No differences have yet been observed in the recovery curves of different thorium compounds after the removal of Th X. For example, the rate of recovery is the same whether the precipitated hydroxide is converted into the oxide or into the sulphate.

136. Disintegration hypothesis. In the discussion of the changes in radio-active bodies, only the active products Ur X and Th X have been considered. It will, however, be shown later that these two products are only examples of many other types of active matter which are produced by the radio-elements, and that each of these types of active matter has definite chemical as well as radio-active properties, which distinguish it, not only from the other active products, but also from the substance from which it is produced.

The full investigation of these changes will be shown to verify in every particular the hypothesis that radio-activity is the accompaniment of chemical changes of a special kind occurring in matter, and that the constant activity of the radio-elements is due to an equilibrium process, in which the rate of production of fresh active matter balances the rate of change of that already formed.

The nature of the process taking place in the radio-elements, in order to give rise to the production at a constant rate of new kinds of active matter, will now be considered. Since in thorium or uranium compounds there is a continuous production of radio-active matter, which differs in chemical properties from the parent substance, some kind of change must be taking place in the radio-element. This change, by which new matter is produced, is very different in character from the molecular changes dealt with in chemistry, for no chemical change is known which proceeds at the same rate at the temperatures corresponding to a red heat and to liquid air, and is independent of all physical and chemical actions. If, however, the production of active matter is supposed to be the result of changes, not in the molecule, but in the atom itself, it is not to be expected that the temperature would exert much influence. The general experience of chemistry in failing to transform the elements by the action of temperature is itself strong evidence that wide ranges of temperature have not much effect in altering the stability of the chemical atom.

The view that the atoms of the radio-elements are undergoing spontaneous disintegration was put forward by Rutherford and Soddy as a result of evidence of this character. The discovery of the material nature of the [Greek: alpha] rays added strong confirmation to the hypothesis; for it has been pointed out (section 95) that the expulsion of [Greek: alpha] particles must be the result of a disintegration of the atoms of the radio-element. Taking the case of thorium as an example, the processes occurring in the atom may be pictured in the following way. It must be supposed that the thorium atoms are not permanently stable systems, but, on an average, a constant small proportion of them—about one atom in every 10^{16} will suffice—break up per second. The disintegration consists in the expulsion from the atom of one or more [Greek: alpha] particles with great velocity. For simplicity, it will be supposed that each atom expels one [Greek: alpha] particle. It has been shown that the [Greek: alpha] particle of radium has a mass about twice that of the hydrogen atom. From the similarity of the [Greek: alpha] rays from thorium and radium, it is probable that the [Greek: alpha] particle of thorium does not differ much in mass from that of radium, and may be equal to it. The [Greek: alpha] particles expelled from the thorium atoms as they break up constitute what is known as the "non-separable activity" of thorium. This activity, measured by the [Greek: alpha] rays, is about 25 per cent. of the maximum. After the escape of an [Greek: alpha] particle, the part of the atom left behind, which has a mass slightly less than that of the thorium atom, tends to rearrange its components to form a temporarily stable system. It is to be expected that it will differ in chemical properties from the thorium atom from which it was derived. The atom of the substance Th X is, on this view, the thorium atom minus one [Greek: alpha] particle. The atoms of Th X are far more unstable than the atoms of thorium, and one after the other they break up, each atom expelling one [Greek: alpha] particle as before. These projected [Greek: alpha] particles give rise to the radiation from the Th X. Since the activity of Th X falls to half its original value in about four days, on an average half of the atoms of Th X break up in four days, the number breaking up per second being always proportional to the number present. After an atom of Th X has expelled an [Greek: alpha] particle, the mass of the system is again reduced, and its chemical properties are changed. It will be shown (section 154) that the Th X produces the thorium emanation, which exists as a radio-active gas, and that this in turn is transformed into matter which is deposited on solid bodies and gives rise to the phenomena of excited activity. The first few successive changes occurring in thorium are shown diagrammatically below (Fig. 50).

Fig. 50.

Thus as a result of the disintegration of the thorium atom, a series of chemical substances is produced, each of which has distinctive chemical properties. Each of these products is radio-active, and loses its activity according to a definite law. Since thorium has an atomic weight of 237, and the weight of the [Greek: alpha] particle is about 2, it is evident that, if only one [Greek: alpha] particle is expelled at each change, the process of disintegration could pass through a number of successive stages and yet leave behind, at the end of the process, a mass comparable with that of the parent atom.

It will be shown later that a process of disintegration, very similar to that already described for thorium, must be supposed to take place also in uranium, actinium and radium. The full discussion of this subject cannot be given with advantage until two of the most important products of the three substances thorium, radium and actinium, viz. the radio-active emanations and the matter which causes excited activity, have been considered in detail.

137. Magnitude of the changes. It can be calculated by several independent methods (see section 246) that, in order to account for the radio-activity observed in thorium, about 3 × 10^4 atoms in each gram of thorium suffer disintegration per second. It is well known (section 39) that 1 cubic centimetre of hydrogen at atmospheric pressure and temperature contains about 3·6 × 10^{19} molecules. From this it follows that one gram of thorium contains 3·6 × 10^{21} atoms. The fraction which breaks up per second is thus about 10^{-17}. This is an extremely small ratio, and it is evident that the process could continue for long intervals of time, before the amount of matter changed would be capable of detection by the spectroscope or by the balance. With the electroscope it is possible to detect the radiation from 10^{-5} gram of thorium, i.e. the electroscope is capable of detecting the ionization which accompanies the disintegration of a single thorium atom per second. The electroscope is thus an extraordinarily delicate means for detection of minute changes in matter, which are accompanied, as in the case of the radio-elements, by the expulsion of charged particles with great velocity. It is possible to detect by its radiation the amount of Th X produced in a second from 1 gram of thorium, although the process would probably need to continue thousands of years before it could be detected by the balance or the spectroscope. It is thus evident that the changes occurring in thorium are of an order of magnitude quite different from that of ordinary chemical changes, and it is not surprising that they have never been observed by direct chemical methods.

1. Crookes, Proc. Roy. Soc. 66, p. 409, 1900.
2. Becquerel, C. R. 131, p. 137, 1900; 133, p. 977, 1901.
3. Rutherford and Soddy, Phil. Mag. Sept. and Nov. 1902. Trans. Chem. Soc. 81, pp. 321 and 837, 1902.
4. Rutherford and Soddy, Phil. Mag. Sept. 1902.
5. The general method of regarding the subject would be unchanged, even if it were proved that the radio-activity of thorium is not due to thorium at all but to a small constant amount of a radio-active impurity mixed with it.
6. Rutherford and Soddy, Phil. Mag. Sept. 1902.