XXII. On the Relative Activity of the Emanation and the Active Deposit from Thorium and from Actinium. By Howard L. Bronson, Ph.D. [Communicated by Professor E. Rutherford, F.R.S.] Many important calculations in radioactivity are based on the assumption that each atom of the various radioactive bodies, at each transformation, gives off either one or no [alpha] particle. This is the most natural assumption, and, in the case of radium and its transformation products, has the support of many theoretical considerations. It is also supported by the fact that radium, radium emanation, radium A, and radium C, each produces about the same amount of ionization, when the four substances are in equilibrium [citation redacted]. This latter point was verified by the writer in the case of radium emanation, radium A, and radium C. Dr. Boltwood suggested that the writer should investigate this question in the case of thorium and its active deposit, as he had evidence that the results obtained in the case of radium did not hold true in the case of thorium. The results obtained in this investigation entirely confirm Dr. Boltwood's view, and led the writer to investigate the same question in the case of actinium. Method and Apparatus. The method adopted was to compare directly the ionization produced by the emanation and the active deposit when the two substances were in equilibrium. This could be easily done in the case of both thorium and actinium, because the transformation periods of their emanations are very short compared with those of their active deposits. The thorium, or actinium, was placed between two layers of filter-paper in a small vessel, and a current of air was drawn through this filter-paper and into the testing vessel by means of a filter-pump. The air current was kept fairly steady by placing between the pump and the testing vessel a large air-chamber, which was connected to the outside air by a mercury trap. After the emanation had been drawn through the testing vessel for a known length of time, or until the active deposit had reached a maximum value, the ionization current due to the sum of the emanation and active deposit was measured. The air current was then cut off and the emanation allowed to decay to a negligible amount, and the ionization current due to the active deposit alone was measured. The difference between these two currents, except 292 [header] for a very small correction, was a measure of the ionization due to the emanation. I£ the active deposit had not reached a maximum, it was easily calculated, since the rates of transformation of all the products involved were known. All the measurements were made by means of an electrometer and " constant deflexion method," and some of them were verified by means of the " rate " method to insure that no serious mistakes were being made. Two cylindrical testing vessels were used. Both were 40 cms. long, and had central electrodes about 0'2 cm. in diameter. The diameters of the testing vessels were 18 cms. and 3' 6 cms. respectively. The largest one allowed the [alpha] particles of the active deposit on the central electrode to be entirely absorbed in the air, except at the ends. The smaller one made the mean free path of the [alpha] particles from the different active substances more nearly equal. All the experiments were repeated several times under as different conditions as possible. The quantity of thorium and actinium used, the velocity of the air current, and the saturation potential on the testing vessel, were all changed several times. The saturation potential was changed from <30 to 600 volts, and both positive and negative values were used. Thus the active deposit was sometimes deposited on the central electrode and sometimes on the cylindrical surface. In order to make sure of this point for the sake of calculation, the central electrode was removed and tested, and found to have on its surface more than 90 per cent, of the active deposit, when the outside of the testing vessel was connected to the positive pole of the battery. A second testing vessel was put in series with the first one, to make sure that none of the active deposit was drawn through by the air current. No evidence of this was ever detected. Measurements and Calculations. Table I. gives a sample set of observations obtained with the thorium emanation in the larger testing vessel. The time was reckoned from the starting of the emanation through the testing vessel. The maximum value of the active deposit was calculated on the assumption that thorium A decays to half value in 10*6 hours. There are small irregular variations in the ionization currents, which are due to slight changes in the velocity of the air current, and to the humidity, which affects considerably the emanating power of the thorium. At the end of 88 hours the air current and potential of the testing vessel were changed and another set of measurements taken. Several such changes were made, and the [header] 293 Table I. Time. Th. Emanation + Active Deposit Active deposit Calc. Max of active deposit Emanation Ratio Activities. [table redacted] measurements were continued for 10 consecutive days. A similar set of observations was taken with the smaller testing-vessel. The experiments with actinium required much less time, because the transformation period of its active deposit is much shorter than that of thorium. In the case of actinium also the experimental conditions were varied as much as possible. As would be expected, the ratio of the ionization due to the active deposit to that due to the emanation was affected by the size of the vessel and the location of the active deposit. None of the other changes in the experimental condition affected this ratio more than a few per cent. Most of these variations could probably be removed by sufficient care, but this was unnecessary for the purpose of the present paper, as the accuracy of the measurements is quite as great as that of the assumptions on which the calculations are based. Now Hahn [citation redacted] has shown that the active deposit from thorium contains two [alpha] ray products, thorium B and thorium C, and that the maximum ranges of their [alpha] particles in air are 5'0 and 8'6 cms. respectively. He has also shown that the [alpha] rays from these two products produce about the same number of ions per cm. of path. Now if the [alpha] particle from thorium emanation, which has a range in air of 5"5 cms., is similar to those from thorium B and thorium C, and if it produces the same number of ions per cm. (when it has the same velocity), then we should expect that the ionization produced by the active deposit would certainly be greater than that produced by the emanation. The exact value of the ratio would, of course, depend upon the mean free paths and the velocities of the different [alpha] particles in the given vessel. The results given in Table I. show that the ionization due to the active 294 [header] deposit, instead of being greater, was less than one half that due to the emanation. Before attempting any explanation of the above results, we will calculate with some degree of accuracy the relative activities that would be expected from the emanation and its active deposit under the conditions of the above experiment. In order to calculate the mean free paths of the [alpha] particles from the different substances, certain assumptions were made: first, that the cylinder was infinite in length, that is, the effect of the ends was neglected ; second, that the emanation was distributed uniformly throughout the cylinder ; third, that the [alpha] particles were shot off equally in every direction, and therefore that only one half of those from the active deposit produced any ions in the air. For the purpose of this calculation, the first assumption should be approximately true, for the emanation was drawn diagonally across the testing vessel, entering about 5 cms. from one end and leaving the vessel on the opposite side and about the same distance from the other end. There should, therefore, be very little emanation or active deposit near the ends. In .any case, any errors introduced into the calculations by this assumption would affect the mean free paths of the a particles from the emanation and active deposit by nearly the same amount, and would therefore have very little effect on their ratio. It is difficult to estimate the accuracy of the second assumption, but the air current was so rapid at the entrance to the vessel that the emanation should have been well distributed. The third assumption is commonly made in all radioactive calculations. On these assumptions the mean free path, in the larger testing vessel, of the [alpha] particles from the active deposit from thorium would be half their maximum range, since one half the [alpha] particles were absorbed by the metal and the other half have their maximum range in air. This gives [citation redacted] cms. as the sum of the mean free paths of the [alpha] particles from thorium B and thorium C. In calculating the mean free path of the [alpha] particles from the emanation, it was found easier to use a geometrical than an analytical method. On a large sheet of clear mica were drawn concentric circles from 1 to 9 cms. in radius and radial lines cutting these circles every 10°. In order to find the mean free path of [alpha] particles shot out from a given point, two sections of the cylinder were taken through this point, one containing the axis and the other at right angles to it. The centre of the concentric circles on the mica was then placed at the given point in each section in turn, and [header] 295 the length of the free path along each radial line was read off directly. The average of the measurements in the two planes, in general, gives a good value for the mean free path of an [alpha] particle starting from the given point. Table II. gives the results of the measurements for the mean free path of the [alpha] particle of thorium emanation in the larger testing vessel. [table redacted] Column I. gives the distance of the point from the axis of the cylinder. Column II. is the mean of II a. and II b. which give respectively the mean free paths of [alpha] particles in the two perpendicular planes. Since the emanation is uniformly distributed in the vessel, the number of atoms of the emanation at a given distance from the axis of the cylinder is proportional to their distance from the axis. Therefore it is necessary to take the sum of the product of columns I. and II. and divide by the sum of column I. in order to get the mean free path of the [alpha] particles for the entire cylinder. The products of columns I. and II. is given in column III. We have seen that the sum of the mean free paths of the [alpha] particles from thorium B and thorium C is 6'8 cms. This value divided by 45 cms. gives 1*51, which should be the ratio of the ionization due to the active deposit to that due to the emanation, if an atom of each substance in breaking up throws out one [alpha] particle, and if the number of ions produced per cm. by an [alpha] particle is the same for each cm. of its path. That the latter assumption is not strictly true has been shown by Bragg, McClung, Hahn, as well as by the writer. This introduces only a small correction, however, and will be discussed later. Now the measured ratio of the activities of the active deposit and the emanation was about 296 [header] .38 (see Table I.), or almost exactly one quarter of the calculated ratio. A complete summary o£ the measurements and calculations for thorium and actinium is given in Table III. Table III. Products. Vessel. Max. range in air. Mean free path in vessel Ratio of paths, active deposit. Emanation Cal. Ratio of ionizations. Active deposit Emanation Measured ratio of ionizations [table redacted] [header] 297 In calculating the mean free paths of the [alpha] particles from the emanation in the smaller vessel, the same method was employed, but it was found to be more accurate to take the two perpendicular planes through the point in such a way that one contained the axis and the other was parallel to it. Also in calculating the mean free path of the [alpha] particle from the active deposit on the inner surface of the smaller cylinder, there was such a great difference in the mean free paths in the maximum and minimum planes, that it was found necessary to take six planes at angles of 30° with one another. The ratios given in column VI. were calculated by taking into account the fact that an [alpha] particle does not produce the same number of ions over each cm. of its path. A curve showing the relation between the ionization due to an [alpha] particle of radium C and the distance from the end of its ionizing path, was given by the writer in Phil. Mag. for June 1906, and is reproduced here. [figure redacted] This curve gives the ionization for the last 4*5 cms. of the ionizing path of an [alpha] particle, and can be extrapolated with little error to cover the last 6 cms. of its path. The extrapolation of the curve to S'6 cms., which is the maximum ionizing path of the [alpha] particle from thorium C, introduces so much uncertainty that the results given in column VI. [footer] 298 [header] are of very little value in the case of thorium. However, in this case the differences between columns V. and VI. are comparatively small. The following calculation of the ratio at the bottom of column VI. shows the method used. The values of the ionizations are taken from an extrapolation of the above figure. Activity of the actinium emanation. Total ionization over the last 5*8 cms. of path for average range of 273 cms. = 325 Activity of the active deposit. Total ionization over the last 5*5 cms. of path = 838 for average range of Ionization due to active deposit 265 Ionization due to emanation ~ 325 [table redacted] Discussion of Results. The agreement between the figures in columns VI. and VIII. is surprising considering the nature of the assumptions and calculations. These results could be explained by assuming that there were present other active substances of very short transformation periods. It would require three such products in the case of thorium and one in the case of actinium. Hahn's measurements on the ionization ranges of the products of thorium and actinium give no evidence of this, and it seems very unlikely. A simpler and more satisfactory explanation would seem to be — that an atom of thorium 0, in breaking up, gives of the same number of [alpha] particles as an atom of thorium B and that an atom of thorium emanation gives off four times this number also that an atom of actinium emanation, in breaking up, gives off twice as many [alpha] particles as an atom of its active deposit. These results raise the very interesting question as to the number of [alpha] particles given off when an atom of any radioactive substance breaks up. If the above conclusions are correct, it has been shown that this number is not the same for every active substance. As the measurements in this [header] 299 paper are only relative, no light has been thrown on the question as to the actual number of [alpha] particles thrown out by an atom of any one of the substances. The writer hopes to continue the present investigation and extend it to other members of the radioactive family. It is possible that light may, in this way, be thrown on the question of the actual number of [alpha] particles thrown off by atoms of the different active substances. Former Results with Radium. The results in the case of radium are added here for the sake of comparison. They were obtained from curves of the rise and decay of the active deposit from radium. These curves were obtained by the writer in 1905, while investigating the transformation periods of the different products of which the active deposit from radium is composed. Table IV. is similar to Table III., but gives only a single set of results. These results indicate, as was said at the beginning of the paper, that the same number of [alpha] particles are thrown off by the disintegration of an atom of either radium emanation, radium A, or radium C. Table IV. Products Max range in air. Mean free path in vessel Ratio of paths Emanation. Active deposit Cal. Ratio of ionization Emanation Active deposit Measured ratio of ionizations [table redacted] The vessel used had a diameter of 4*8 cms. and a length of 20 cms. The active deposit was on the central electrode, and in making the calculations the actual length of the vessel was taken into account. Macdonald Physics Building, McGill University, Montreal, April 27th, 1908. [footer]