LII. Homogeneous Röntgen Radiation from Vapours. By J. Crosby Chapman, B.Sc, Layton Research Scholar of the University of London (King's College) ; Research Student of Gonville and Caius College, Cambridge [Communicated by Prof. C. G. Barkla, M.A., D.Sc.]. ALL bodies when exposed to Röntgen radiation emit secondary X-rays. It has been shown [citation redacted] that these secondary rays consist of two types — a scattered radiation having the same penetrating power as the primary beam and resembling it in that it is heterogeneous and an X radiation characteristic only of the element used as radiator and independent of the penetrating power of the exciting primary beam. The elements belonging to the group with atomic weights from hydrogen to sulphur have been shown to give out, when excited, a great preponderance of the first type of radiation termed scattered radiation [citation redacted], while those in the group from [header] 447 chromium onwards emit almost wholly the characteristic radiation which, on account o£ its homogeneity, suffers equal percentage absorptions when transmitted through equal thicknesses of aluminium. By determining these percentage absorptions the values of [lambda]/p (where [formula redacted] and p = density of aluminium) have been obtained for the different elements giving characteristic radiations. Previously, in experiments performed for determining the coefficient - , the elements used have been, for purposes o£ convenience, in the solid state, either pure or in the form of compounds. The following [figure redacted] experiments were undertaken at the suggestion of Prof. Barkla with a view to showing that the same type of homogeneous radiation is emitted by the elements, whether they are in tho solid state or in the form of vapour. 448 [header] The apparatus used to determine the nature of the Secondary Radiation emitted by the vapours when exposed to the rays, consisted o£ an iron box which contained the gas : this was fitted with aluminium windows and was placed as shown in the diagram (p. 44-7), so that the radiations reaching the electroscopes could come only from the vapour inside the chamber. That such is the case with this arrangement is indicated by the dotted lines in the figure which mark the path taken by the extreme scattered rays. The process of the experiment was as follows. The radiation from the vapour inside the chamber was allowed to pass into both electroscopes, which were of the ordinary gold-leaf type described by Prof. Barkla. The deflexion in the electroscope M was observed, while there was a certain deflexion in the standardiser S. An aluminium sheet of required thickness was then placed in front of the electroscope M, and the deflexion again read while the standardising electroscope suffered the same deflexion as before. In this way the percentage of the radiation which had been absorbed could be determined, and thence a value for [formula redacted] . Owing to the difficulties in obtaining and working with gases containing elements of atomic weight greater than 52, in the experiment it has been impossible to determine the coefficients [formula redacted] for more than two of the elements. The vapours of ethyl bromide and ethyl iodide were employed on account of their, comparatively speaking, high vapour pressures at the temperature of the experiment. Secondary Radiation from the Vapour of Ethyl Bromide and from Solid Bromine Compounds. Air which had previously been bubbled through ethyl bromide and afterwards passed through a glass spiral immersed in water at 4° C, so that it was saturated at a temperature lower than that of the room, was drawn through the chamber by means of a filter-pump and a steady condition was obtained. The X-ray bulb, which was permanently connected to a pump, was kept at a suitable degree of hardness in order to obtain a maximum intensity of radiation for measuring purposes. The following typical results were obtained with the bromide, proceeding as previously described. [header] 449 Radiation from Vapour of Ethyl Bromide. (Time of observation 2 to 3 minutes.) Percentage absorption by Aluminium previous to absorption in other column. Percentage absorption by Aluminium (-00626 cm.) after absorption in column 1. [table redacted] As the value of [formula redacted] for solid bromine had not previously been determined, the gas-chamber was replaced by a plate of sodium bromide obtained by sticking the powder to a thin aluminium sheet The same observations were repeated with Radiation from Solid Bromine Compounds. Percentage absorption by Aluminium previous to absorption in other columns. Percentage absorption by Aluminum (.0026) after absorption in column I) Radiation from Sodium Bromide. Radiation from Bromyl Hydrate. [table redacted] [formula redacted] 450 [header] the radiation from the plate as with that from the bromide vapour. A plate of bromyl hydrate was also used in this way. It will be observed that the value of [formula redacted] obtained from the vapour has, within the limits of error, the same value as that from the solid, thus showing that the two radiations are identical in character. In order to find the nature of the curve connecting the atomic weights with [formula redacted], in the neighbourhood of bromine the values of [formula redacted] for the radiations from selenium strontium, molybdenum, were determined. Element used as radiator. Value of [formula redacted] for radiation. [table redacted] Using these values combined with others known before to plot [formula redacted] against atomic weights, a smooth curve results, on which the value of [formula redacted] for bromine lies, showing that the latter both in the solid aud vapour state gives out a characteristic radiation the absorption coefficient of which follows the law determined for solid elements. Radiations from Vapour of Methyl Iodide and from Solid Iodine. The apparatus was similar to that used with the ethyl bromide with the exception that, in this case, a quantity of the iodide almost sufficient to saturate the space inside the chamber was poured into an aluminium dish in the box ; the filter pump and saturation bottle were dispensed with. In addition the X-ray bulb was hardened by abstracting a little air with the pump, in order that the rays given off might excite the hard iodine radiation. The intensity of the rays from the ethyl iodide was very much less than that from ethyl brqmide owing to only a [header] 451 small part of the incident primary being sufficiently penetrating to excite the characteristic iodine radiation. This combined with the fact that hard rays do not ionize to any large extent made the times of observation much longer than in the other case. Radiation from Vapour of Methyl Iodide. (Time of observation 15 to 20 minutes.) Percentage absorption by Al previous to absorption in column II. Percentage absorption by Al ('0377 cm.) after absorption in column I. [table redacted] A plate of solid iodine was constructed and the radiation from it examined in the same manner as the bromide plate, with the following results : — Radiation from Solid Iodine. Percentage absorption by Al previous to absorption in column II. Percentage absorption by Al ("0377 cm.) after absorption in column I. [table redacted] Again, the value of [formula redacted] for the solid and vapour was the 452 [header] same. By continuing the curve before mentioned, it will be seen that this value lies approximately on it. Although it has only been proved in these two cases that elements in the solid and vapour state emit the same type of radiation, yet it is safe to conclude that what applies here holds generally ; especially considering that the atomic weights of bromine and iodine are well separated. It is evident that this similarity of character in the radiations is what would follow from the fact that the phenomena of secondary X-rays are atomic in their nature. Bombardment of Atoms by Ejected Corpuscles. In a previous paper [citation redacted] facts have been brought forward indicating that the characteristic secondary radiation does not result from the subsequent bombardment of atoms by ejected corpuscles. A slight adaptation of the above experiment shows this. For if carbon dioxide and hydrogen are used separately under the same conditions as the gas in which the vapour of ethyl bromide is passed into the chamber, in the former case it is the carbon dioxide gas which is chiefly bombarded by ejected electrons, while with hydrogen it is the ethyl bromide itself which has for the most part to stop the expelled corpuscles. Therefore, if subsequent bombardment causes the characteristic radiation, we should expect greater intensity with the hydrogen than with the carbon dioxide as the gas. This point can be investigated experimentally. The apparatus used was practically identical with that previously described. The tin box was moved farther back and the primary rays were cut down by a lead tunnel in place of the slits then used. At the same time, the electroscope S was moved into a position to receive radiations from a thin stick of selenium so placed in this tunnel that a small part of the exciting radiation from the X-ray bulb fell on it. The electroscope M was brought nearer to the box, this was possible owing to the alteration in breadth of the primary beam. Since the atomic weight of selenium is 79 while that of bromine is 80, the intensity of secondary radiation from the stick of selenium in the tunnel standardizes the power of the incident rays in exciting the homogeneous bromine radiation. In the first part of the experiment, the carbon dioxide gas obtained from a cylinder was saturated with ethyl bromide at 3°*5 C, and was passed through the chamber till a steady state was reached. The deflexion in the electroscope receiving radiations from the chamber, while the other [header] 453 electroscope suffered a certain deflexion, was noticed. The .carbon dioxide cylinder was replaced by a hydrogen "kip," and hydrogen saturated at the same temperature was passed under similar conditions through the box, and the deflexion of the chamber electroscope, while the standardize]' underwent the same deflexion, was noticed. The results are shown : — Intensities of Radiations in the two cases. Temperature of saturation = 3°*5 C. Deflexion of standardizing electrosrope. Deflexion of chamber electroscope with carbon dioxide as saturated gas. Deflexion of chamber electroscope with hydrogen as saturated gas. [table redacted] Amount of secondary radiation with H 2 as gas saturated with C 2 II 5 Br Amount of secondary radiation with C0 2 as gas saturated with C 2 H 5 Br This slight difference in the intensities is of the order of magnitude which would result from the excess of absorption in the heavier carbon dioxide gas. Knowing the vapour pressure at 3°*5 C, some relative idea of the magnitude of the difference can be deduced. For, assuming that the absorption of the [beta] particles varies directly as the absorbing mass, we get : - Case I. Amount of corpuscular radiation absorbed by ethyl bromide Amount of corpuscular radiation absorbed by C0 o Case II. Amount of corpuscular radiation absorbed by ethyl bromide [formula redacted] Amount of corpuscular radiation absorbed by H 2 Thus if the expelled electrons do by bombarding the bromine atom make it emit its characteristic radiation, the above calculations show that there must be a most noticeable difference in the intensities of radiation in the two cases. 454 [header] The experimental results obtained show, however, that there is practically no difference in the intensities for the two gases, which proves that the bombardment theory is quite untenable. In conclusion my best thanks are due to Professor Barkla for his interest and encouragement during the carrying out of these experiments. Wheatstone Laboratory, Kino-'c$ College.