S. Lagomarsino

Minimum ionizing and alpha particles detectors based on epitaxial semiconductor silicon carbide

The relatively high value of the energy required to produce an electron-hole pair in silicon carbide, SiC, by a minimum ionizing particle (MIP) against the value for Si, imposes severe constrains in the crystallographic quality, the thickness and the doping concentration of the SiC epitaxial layer used as the detection medium. In this work, a 40 /spl mu/m thick 4 H-SiC epitaxial layer with a low doping concentration of /spl sim/5/spl times/10/sup 13/ cm/sup -3/ was used in order to have a relatively high number (/spl sim/2200) of e-h pairs generated by a MIP and to deplete the total active layer at relatively low reverse bias (60 V). The detectors are realized by the formation of a nickel silicide (Ni/sub 2/Si) on the silicon surface of the epitaxial layer (Schottky contact) and of the ohmic contact on the backside of a 4 H-SiC heavily doped substrate. We present experimental data on the charge collection properties with /spl alpha/-particles from /sup 241/Am and /spl beta/-particles from /sup 90/Sr. In both cases, a 100% charge collection efficiency, CCE, is demonstrated and the diffusion contribution of the minority charge carriers to CCE is pointed out. The charge spectrum for MIPs from /sup 90/Sr shows a full detection efficiency with the pedestal (noise) clearly separated by the signal (Landau distribution) at reverse bias values comparable and higher than the one needed to totally deplete the layer. Moreover, no degradation was observed at 94/spl deg/C in the CCE and in the energy resolution of the /sup 241/Am alpha-signal from the SiC detector.

Radiation hardness after very high neutron irradiation of minimum ionizing particle detectors based on 4H-SiC p/sup +/n junctions

In this work we analyzed the radiation hardness of SiC p/sup +/ n diodes used as minimum ionizing particle (MIP) detectors after very high 1 MeV neutron fluences. The diode structure is based on ion implanted p/sup +/ emitter in an n-type epilayer with thickness equal to 55 /spl mu/m and donor doping N/sub D/=2/spl times/ 10/sup 14/cm/sup -3/. The diode breakdown voltages were above 1000 V. At 1000 V the leakage currents are of the order of 1 nA for all the measured diodes. The full depletion voltage is near 220-250 V. The charge collection efficiency to minimum ionizing particle has been investigated by a /sup 90/Sr /spl beta/ source. At 250 V the collected charge of the unirradiated diodes saturates near 3000 e/sup -/. At bias voltages over 100 V the energy spectrum of the collected charge was found to consist of a signal peak well separated from the noise. At around 250 V the signal saturates, in agreement with CV results. These devices have been irradiated at 6 different fluences, logarithmically distributed in the range 10/sup 14/-10/sup 16/ (1 MeV) neutrons/cm/sup 2/. The leakage current after irradiation decreases. The collected charges decrease for increasing fluences, remaining very high only until some 10/sup 14/ n/cm/sup 2/.

Three-dimensional diamond detectors: Charge collection efficiency of graphitic electrodes

Implementation of 3D-architectures in diamond detectors promises to achieve unreached performances in the radiation-harsh environment of future high-energy physics experiments. This work reports on the collection efficiency under β-irradiation of graphitic 3D-electrodes, created by laser pulses in the domains of nanoseconds (ns-made-sensors) and femtoseconds (fs-made-sensors). Full collection is achieved with the fs-made-sensors, while a loss of 25%–30% is found for the ns-made-sensors. The peculiar behaviour of ns-made sensors has been explained by the presence of a nano-structured sp3-carbon layer around the graphitic electrodes, evidenced by micro-Raman imaging, by means of a numerical model of the charge transport near the electrodes.

Characterisation of epitaxial SiC schottky barriers as particle detectors

Abstract Epitaxial SiC devices have been tested as radiation detectors for minimum ionising particles. The devices used are based on a commercial 4H–SiC epitaxial n-type layer deposited onto a 4H–SiC n + type substrate wafer doped with nitrogen. Single-pad Schottky contacts have been produced by deposition of a 1000-A gold film on the epitaxial layer using a lift-off technology and ohmic contacts have been deposited on the rear substrate side. The capacitance–voltage characteristics have been measured to determine the net effective doping in the space charge layer and the maximum active thickness of the devices. The measurements showed possible non-uniformity in the net doping of the epitaxial layer. The charge collection efficiency (CCE) has been tested by means of a 0.1 mCi 90 Sr β-source. A 100% CCE is measured at the maximum active thickness, which is achieved above approximately 400 V. The charge signal of the SiC devices is stable and reproducible, with no evidence of priming or polarisation effects, due to the high crystalline quality of the epitaxial layer.

Controlled variation of the refractive index in ion-damaged diamond

Abstract A fine control of the variation of the refractive index as a function of structural damage is essential in the fabrication of diamond-based optical and photonic devices. We report here about the variation of the real part of the refractive index at λ xa0=xa0632.8xa0nm in high-quality single-crystal diamond damaged with 2 and 3xa0MeV protons at low-medium fluences (10 13 –10 17 ionsxa0cm −xa02 ). After implanting the samples in 125xa0×xa0125xa0μm 2 areas with a raster scanning ion microbeam, the variation of optical thickness of the implanted regions was measured with laser interferometric microscopy. The results were analyzed with a model based on the specific damage profile. The technique allows the direct fabrication of optical structures in bulk diamond based on the localized variation of the refractive index, which will be explored in future works.

Silicon-on-diamond material by pulsed laser technique

We present a method to bond directly silicon and diamond plates to obtain a single silicon-on-diamond material, with a carbon–silicon interface of unprecedented quality. The bonding is performed at room temperature, via picosecond 355 nm pulsed laser irradiation of the silicon-diamond interface, through the transparent diamond. The obtained material exhibits excellent mechanical strength and uniformity of the bonding, as shown by mechanical tests and analysis of the cross section based on scanning electron microscopy. The bonding is ascribed to silicon carbide nanolayers at the interface which, along with amorphous silicon nanolayers, have been quantitatively detected and evaluated by means of optical spectroscopy measurements. A physical insight into the processes occurring at the diamond-silicon interface during the pulsed irradiation and cooling has been provided by a finite element numerical model. A rationale is then given for the observed SiC bond in terms of silicon and diamond melting and inter-di...

Polycrystalline diamond synthesis by means of high power pulsed plasma glow discharge CVD

Abstract A pulsed glow discharge reactor for chemical vapour deposition of high quality diamond films is presented. The Raman quality and the morphology of the diamond films exhibit a strong dependence on the discharge pulse shape. This result is explained with a simple model involving the average current density j0 and the average squared amplitude of the pulse 〈Δj2〉 as relevant parameters. This CVD method does not require any substrate pretreatment, and the nucleation rate is seen to increase with current density, methane concentration and pressure. The quality of the deposits is independent of the inter-electrode distance in the 25–35 mm range. The influence of the substrate temperature on the diamond morphology and on diamond etching from the substrate is discussed.

Silicon Carbide for High Signal to Noise Ratio MIPs Detection From Room Temperature to 80

The relatively low value of the number of electron-hole (e-h) pairs per micron for Minimum Ionizing Particles (MIPs) in SiC against the value for Si, imposes severe constrains on the crystallographic quality, the thickness and the doping concentration of the SiC epitaxial layer used as detection medium. In this work, a 85 mu m thick 4H-SiC epitaxial layer with a low doping concentration, Neff les 1 times 1014 cm-3 , was used in order to have a high number ( ap 4700) of e-h pairs generated by a MIP in the active region. We present experimental data on the charge spectrum for beta MIPs from a 90 Sr source, collected in a temperature range from room temperature up to 81degC.

^{\circ}

Deep levels in undoped chemical vapor deposited (CVD) diamond films have been characterized by thermally stimulated current spectroscopy (TSC) in the range of 300-650 K. The TSC results have been tentatively correlated tothe performance of the samples as on-line dosimeters and particle detectors. The TSC signal is dominated by a set of deep levels with an activation energy in the range of 1.0-1.4 eV. The trapping activity of these levels, which can be related to grain boundaries, strongly influences the detector performance at room temperature. After neutron irradiation up to the fluence of 2 x 10 1 5 n/cm 2 the amplitude of the TSC signal decreases of about one order of magnitude, the pumping effect becomes significantly less pronounced and the charge collection efficiency decreases of about 30%. Thus, the radiation-induced removal of these deep levels must be accompanied by the creation of other traps, probably vacancy-related and not visible by TSC in this temperature range, which have little effect on the dynamic response of the device but can affect the charge collection efficiency.

C

Schottky diodes based on a 4H-SiC epitaxial n-type layer deposited onto a 4H-SiC n/sup +/ type substrate have been tested as particle detectors. The charge collection efficiency (CCE) has been tested by means of a 0.1mCi /sup 90/Sr /spl beta/-source and with 5.48 MeV /spl alpha/-particles from /sup 241/Am. The response of the SiC devices, investigated over a range of thickness up to /spl sim/20/spl mu/m, is characterized by a 100%CCE, the charge signal is stable and reproducible, with no evidence of priming or polarization effects.