WITCH, a Penning trap for weak interaction studies

This work is the completion of the installation of the WITCH set-up and the first tests and commissioning of it. The first goal of the WITCH experiment is to improve the present limit on a scalar interaction in nuclear $\beta$-decay by determining the $\beta$-neutrino angular correlation parameter $a$ via a precise measurement of the shape of the energy spectrum of the recoil ions. The development of the WITCH set-up and its installation at ISOLDE (CERN) were recently completed. The principle of WITCH is based on a combination of a Penning trap to confine the radioactive ions and a retardation spectrometer to probe the energy of the recoil ions resulting from $\beta$-decays in the trap. Extensive computer simulations show that for a reasonable measurement time a precision on the $a$-parameter of 0.5% can be achieved. This corresponds to an upper limit for the scalar interaction constant Cs/Cv < 9% at 95% C.L. Designing and constructing a set-up as large and complex as the WITCH set-up takes time, several years in this case. The installation period was finished in autumn 2004 while intensive commissioning of the set-up was performed during the whole year 2004. A necessary requirement to perform the first tests was that some control system to operate the electronics (e.g. the HV power supplies) had to be available. The main aim of these tests was to check the operation of the beam transport, the pulsing down of the ion beam and the injection of ions into the high magnetic field, to test Penning trap basics and check the spectrometer operation, and finally to optimize as many settings as possible. These tests showed that the full set-up up to the spectrometer is now operational, although several efficiencies still have to be improved. The results of the commissioning stage were carefully analysed and possible improvements were suggested. This includes both technical modifications as well as the necessary tests in order to optimize the set-up and achieve the required effciency. In addition, GEANT4 simulations were performed to study the behaviour of the $\beta$-particles in the set-up. These simulations yielded the best place for the $\beta$-detector and also provided an estimate of the $\beta$-background in the experiment. This background is somewhat higher than expected. Tests to clarify the situation are suggested and some ideas to modify the set-up in view of this background are given. The WITCH experiment is not restricted to the study of the possible admixture of a scalar current in weak interactions. Due to its unique feature of being able to measure the recoil ion spectrum it also opens possibilities for other observables. A few ideas for other experiment that might be performed with this set-up are briefly discussed.