A method is described for investigating the ionization produced in helium by collisions between quasi-stationary atoms-that is, atoms possessing only thermal velocities-and a beam of neutral helium atoms with kinetic energy less than 100 eV. Ionization is shown to begin when the kinetic energy of the bombarding atoms is twice as great as the minimum kinetic energy which electrons must possess in order to ionize helium.
I n tr o d u ctio nThe excitation and ionization produced when electrons make collisions with the atoms of a gas have been extensively investigated over a number of years, and a certain amount of work has also been done with positive ions as bombarding par ticles; but there is little information concerning the effects produced when the bombarding particles are neutral atoms. The possibility of producing reasonably intense beams of atoms with kinetic energy greater than th a t which could be obtained from thermal sources was first indicated by the work of Kallmann & Rosen (1929, 1930. They showed th a t when positive ions passed through the gas from which they had been formed, an ion might capture an electron from a ' stationary ' atom or molecule, and go on its way as a neutral particle, with its speed and direction of motion unchanged. If, therefore, a beam of ions produced from a gas traverses a chamber containing th a t gas, and if, on emerging from this chamber, the remaining ions are deflected out of their original path, we obtain a beam of neutral particles, whose interaction with ' stationary * atoms may be investigated. The kinetic energy of the particles in the neutral beam can be controlled by adjusting the potential difference accelerating the original positive ions, and the homogeneity in velocity of the neutral particles is simply th at of the ion beam.Several attem pts have been made to measure the minimum kinetic energy which such neutral atoms must possess in order to excite or to ionize gases through which they pass; but the earlier results were vitiated by ignorance of the nature and relative magnitudes of secondary effects, such as the release of electrons from metal surfaces by atomic impact, which masked or simulated the onset of gaseous ioniza tion. These secondary effects have now been investigated, chiefly by Rostagni (1934a, 6), and the precautions necessary to avoid confusing them with the ionization resulting from atomic collisions are fairly well established. Rostagni (1934c) also showed th at the efficiency of neutral atoms in producing ionization in a gas is very small. Thus in order to determine the minimum kinetic energy which such an atom requires in order to ionize by collision, the experimental arrangements must ensure th a t the largest possible number of atoms enters the ionization chamber, where the pressure must be such th at as great a fraction as possible makes collisions with