Coefficient of recombination of ions, produced by x-rays, as function of x-ray exposure time, initial ion concentration and time of recombination.---The present work constitutes a study of the recombination of ions and of ions and electrons in gases, by means of a new direct method, using constant radiation from a Coolidge x-ray tube as the ionizing agent, together with a rotating commutator shutter. This method permits wide limits of independent variation of: (1) x-ray exposure (${t}^{\ensuremath{'}}$); (2) initial ion concentration (${n}_{0}$); (3) time of recombination ($t$). Results in air show that $\ensuremath{\alpha}$, previously assumed constant and equal to 1.6\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}6}$ actually varies as a function of ${t}^{\ensuremath{'}}$ and of $t$. It is high, 4\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}6}$ or more, for small values of the above quantities and drops rapidly to nearly a constant value. $\ensuremath{\alpha}$ is also found to be a function of ${n}_{0}$. These facts can only be explained on the assumption that the ions are not initially uniformly distributed, but are in pairs along the paths of the x-rays. This leads to fictitiously high calculated values of $\ensuremath{\alpha}$ corresponding to low apparent values of ${n}_{0}$ and $n$, erroneously computed on the assumption of random distribution. As ${t}^{\ensuremath{'}}$, and with it ${n}_{0}$ and also $t$, increase, $\ensuremath{\alpha}$ approaches a constant value between 0.8 and 0.9\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}6}$, which is probably nearly the true value. This result lends strong support to the theory of Thomson that the process of diffusion plays the major role in the initial stages of the mechanism of recombination; i.e. in bringing ions together.Coefficient of recombination in argon: mechanism of recombination, formation of negative ions by electron attachment to impurities.---In argon, where electrons remain free for appreciable intervals, $\ensuremath{\alpha}$ is nearly constant. For the same time intervals it has only about half the value found in argon-oxygen mixtures, where the electrons attach rapidly, or the value found in dry air. On the basis of the kinetic theory $\ensuremath{\alpha}$ would be expected to be the same in argon as in air. The constancy of $\ensuremath{\alpha}$ with $t$ and ${t}^{\ensuremath{'}}$ is due to the rapid achievement of random distribution when free electrons are present. The value of $\ensuremath{\alpha}$ indicates that the coefficient of recombination for electrons and positive ions is less than for positive and negative ions. This conclusion is in agreement with the results of Kenty on recombination between electrons and positive ions in argon. Consequently $\ensuremath{\alpha}$, as observed in pure argon, is not a true measure of the recombination of positive and negative ions, but serves as an indication of the rate of negative ion formation through the attachment of electrons to impurities present.
Read full abstract