Application of high-resolution gamma spectroscopy to the study of atomic collision cascades in solids

In the Crystal-GRID (Gamma Ray induced Doppler Broadening) technique one uses the fact that atomic nuclei can be excited by a
thermal neutron capture reaction. The de-excitation process takes place via the emission of gamma-quanta. The emission of the first
gamma-quantum after the neutron capture leads to a recoil of the excited nucleus. Typical recoil energies are
in the range of several hundreds of eV and therefore much higher than the displacement threshold of the atom. The recoiling atom will start to
collide with its neighbours losing successively kinetic energy until
thermal energies are reached.

During the collision process a second gamma-ray might be emitted by the atomic nucleus and, as the emitter was moving, the energy of the
gamma-quantum
will be Doppler shifted. The Doppler-shift is a very direct indicator to study the motion of the emitting particle and yields information on the
recoil
process of the atom. However, due to the isotropic orientation of the recoil direction one observes a
Doppler-broadening of the the Intensity distribution I(Eg2). The sensitivity of the technique to the study of atomic collision processes is
enhanced if single crystalline targets are used. In this case the microscopic anisotropy of the slowing down process induces a characteristic
fine structure of the Doppler-broadened line shapes. Measurements with differently aligned targets with respect to the axis of observation give
different line shapes I(Eg2) and allow therefore to accumulate information on the microscopic collision processes. A detailed description of the
experimental technique will be given together with an overview on already performed experiments. In these experiments the comparison of
experimental data to predictions deduced from Molecular Dynamics simulations has allowed to obtain new repulsive interatomic potentials.

Currently, the measurement of I(Eg2) is carried out using a double flat crystal geometry, which allows to obtain a relative energy resolution of
(Delta E)/E = 10-6. The very small solid angle of 10-11 of this geometry requires the use of massive targets. This limits significantly the variety
of compounds, which can be studied by this technique. The use of a double Dumond geometry would allow to overcome this problem. Recent
technical improvements of the GAMS5 spectrometer towards a double bent crystal geometry will be demonstrated. The current status of the
spectrometer and the expected capabilities will be presented.