Comparison of the performance of a high voltage generator insulated by gas or liquid dielectric.

A module of a wireless high voltage generator was tested immersed in both gaseous and liquid environments providing electrical insulation. The overall performance of the module as well as a detailed performance of the key components are reported, and a comparison between the results in gas and liquid is given. The tests performed on the liquid dielectric show that it is a valid alternative to high pressure gas electrical insulation.

Realization of a high voltage generator by series connection of floating modules.

A high voltage generator built by a series connection of 100 kV modules was produced. The series connection feasibility is ensured by the inherent floating character of each module which is wireless powered by high efficiency photovoltaic cells illuminated by a laser system. Each module is equipped with a control and monitoring board allowing excellent stabilization of the high voltage output. The performance of the system in terms of reliability, stability, and efficiency was evaluated. In particular using a three module setup, we achieved a maximum voltage of 234 kV with stability better than 0.1%.

A modular optically powered floating high voltage generator.

The feasibility of fully floating high voltage (HV) generation was demonstrated producing a prototype of a modular HV system. The primary power source is provided by a high efficiency semiconductor power cell illuminated by a laser system ensuring the floating nature of each module. The HV is then generated by dc-dc conversion and a HV multiplier. The possibility of series connection among modules was verified.

Neutron spectrometry with a monolithic silicon telescope.

A neutron spectrometer was set-up by coupling a polyethylene converter with a monolithic silicon telescope, consisting of a DeltaE and an E stage-detector (about 2 and 500 microm thick, respectively). The detection system was irradiated with monoenergetic neutrons at INFN-Laboratori Nazionali di Legnaro (Legnaro, Italy). The maximum detectable energy, imposed by the thickness of the E stage, is about 8 MeV for the present detector. The scatter plots of the energy deposited in the two stages were acquired using two independent electronic chains. The distributions of the recoil-protons are well-discriminated from those due to secondary electrons for energies above 0.350 MeV. The experimental spectra of the recoil-protons were compared with the results of Monte Carlo simulations using the FLUKA code. An analytical model that takes into account the geometrical structure of the silicon telescope was developed, validated and implemented in an unfolding code. The capability of reproducing continuous neutron spectra was investigated by irradiating the detector with neutrons from a thick beryllium target bombarded with protons. The measured spectra were compared with data taken from the literature. Satisfactory agreement was found.

A feasibility study of a solid-state microdosimeter.

A solid-state silicon detector is a challenging device for microdosimetry, mainly because it can provide sensitive zones of the order of a micrometer. Moreover, these detectors are characterized by a high spatial and a good energy resolution. However, they may present some limitations, such as: (i) the minimum detectable energy which is limited by the electronic noise; (ii) radiation hardness; (iii) the geometry of the sensitive volume; (iv) the field-funnelling effect; (v) the non-tissue-equivalence of silicon. This work discusses a feasibility study of a microdosimeter based on a monolithic silicon telescope, consisting of a DeltaE and an E stage-detector, about 1 and 500 microm thick, respectively. Charges are collected separately in the two stage-detectors. The use of the DeltaE stage coupled with a tissue-equivalent converter was investigated as a solid-state microdosimeter. Irradiations with monoenergetic neutrons were performed at the INFN-Laboratori Nazionali di Legnaro (Italy). The field-funnelling effect appears to be negligible from the comparison of the experimental data with the results of Monte Carlo simulations, performed with the FLUKA code. The preliminary results of an analytical approach for the correction for geometrical effects and tissue-equivalence are also presented.

Performance of a neutron spectrometer based on a PIN diode.

The neutron spectrometer discussed in this work consists of a PIN diode coupled with a polyethylene converter. Neutrons are detected through the energy deposited by recoil-protons in silicon. The maximum detectable energy is -6 MeV and is imposed by the thickness of the fully depleted layer (300 microm for the present device). The minimum detectable energy which can be assessed with pulse-shape discrimination (PSD) is -0.9 MeV. PSD is performed with a crossover method and setting the diode in the 'reverse-injection' configuration (i.e. with the N+ layer adjacent to the converter). This configuration provides longer collection times for the electron-hole pairs generated by the recoil-protons. The limited interval of detectable energies restricts the application of this spectrometer to low-energy neutron fields, such as the ones which can be produced at facilities hosting low-energy ion accelerators. The capacity to reproduce continuous neutron spectra was investigated by optimising the electronic chain for pulse-shape discrimination. In particular, the spectrometer was irradiated with neutrons that were generated by striking a thick beryllium target with protons of several energies and the measured spectra were compared with data taken from the literature.

A recoil-proton spectrometer based on a p-i-n diode implementing pulse-shape discrimination.

A recoil-proton spectrometer was created by coupling a p-i-n diode with a polyethylene converter. The maximum detectable energy, imposed by the thickness of the totally depleted layer, is approximately 6 MeV. The minimum detectable energy is limited by the contribution of secondary electrons generated by photons in the detector assembly. This limit is approximately 1.5 MeV at full-depletion voltage and was decreased using pulse-shape discrimination. The diode was set up in the 'reverse-injection' configuration (i.e. with the N+ layer adjacent to the converter). This configuration provides longer collection times for the electron-hole pairs generated by the recoil-protons. The pulse-shape discrimination was based on the zero-crossing time of bipolar signals from a (CR)2-(RC)2 filter. The detector was characterised using monoenergetic neutrons generated in the Van De Graaff CN accelerator at the INFN-Laboratori Nazionali di Legnaro. The energy limit for discrimination proved to be approximately 900 keV.

A feasibility study of a single event spectrometer based on semiconductor devices.

The electronics employed around particle accelerators can be disturbed or damaged because of single event effects (SEE). The most likely effect is the single event upset (SEU) which may affect all memory devices. In the case of high energy accelerators, SEUs are mostly produced by secondary charged particles generated by neutron interactions. The measurement of the energy and the lineal energy distribution of these neutron-induced charged particles was proposed. As a first approach, a commercial p-i-n photodiode was employed. This device was irradiated with thermal and monoenergetic fast neutrons. Some effects limiting the use of such a detector as a SEE spectrometer were observed, giving guidelines for the design of an application specific integrated circuit (ASIC). The possibility of creating a solid state microdosemeter by coupling the ASIC with a tissue-equivalent radiator is discussed. Moreover, the p-i-n photodiode covered with a hydrogenated plastic radiator may be employed as a proton-recoil spectrometer.