The germanium crystals used for the fabrication of nuclear radiation detectors are required to have a purity and crystalline perfection which is unsurpassed by any other solid material. These crystals should not have a net electrically active impurity concentration greater than 10l0 cm−3 and be essentially free of charge trapping defects.

Such perfect crystals of germanium can be grown only because of the highly favorable chemical and physical properties of this element. However, ten years of laboratory scale and commercial experience has still not made the production of such crystals routine. The origin and control of many impurities and electrically active defect complexes is now fairly well understood but regular production is often interrupted for long periods due to the difficulty of achieving the required high purity or to charge trapping in detectors made from crystals seemingly grown under the required conditions.

The compromises involved in the selection of zone refining and crystal grower parts and ambients is discussed and the difficulty in controlling the purity of key elements in the process is emphasized. The consequences of growing in a hydrogen ambient are discussed in detail and it is shown how complexes of neutral defects produce electrically active centers.

As a base material for semiconductor devices, silicon is more widely used than any other semiconductor. The physical properties, in particular the bandgap which is significantly larger than that of germanium, makes the material extremely important for electronic devices. The world's total annual production of silicon is at present some 2000 t [1]. Compared with this, the 10 kg/year of silicon that is used for detectors is rather modest. However, since work on semiconductor radiation detectors started 25 years ago, silicon in addition to germanium forms the centre of interest as the basis for production of nuclear radiation spectrometers, mainly as high energy particle detectors, but also as X-ray detectors.

High resistivity silicon single crystals for radiation detectors were grown by the floating-zone (FZ) method with and without one-pass zone refining of ultra pure starting material obtained by the decomposition of monosilane gas. In this method, boron is removed from monosilane during gas generation. The main residual impurity is phosphorus from phosphine which is removed with zeolite A after distillation of monosilane. Photoluminescence analysis and an improved 4-point probe resistivity measuring method were used to evaluate the material.

Semiconductor nuclear radiation detectors require deep depletion depths (0.03–3.0 cm) and effective charge collection distances which are several times longer than these depletion depths. These requirements place stringent limitations on the net electrically active impurity concentration, and on the concentration of deep centers which can trap carriers generated by the incident nuclear radiation. This need for extremely pure material distinguishes the interests and efforts of the semiconductor detector community from the rest of the semiconductor community. This paper reviews the characterization of shallow-level, deep-level, neutral, and extended defects in germanium and silicon for nuclear radiation detectors. Photothermal ionization spectroscopy has been used extensively to identify the residual hydrogenic impurities in high-purity (∣NA–ND∣ ≈ 1010–1011 cm−3 ) germanium and silicon. Deep level transient spectroscopy has been effectively used to detect and identify deeper levels in high-purity germanium. Residual neutral defects are not necessarily passive: they may complex to form deep or shallow levels, they may precipitate, or they may act as nucleation sites for precipitation. The properties of extended defects (dislocations, lineage, inclusions, precipitates) and their effects on device performance are fundamentally less well understood, as the origin of the electrical activity of these defects is uncertain. It has been found in numerous instances that chemical interactions among defects are important even in these high-purity semiconductors.

Selected aspects of NaI(Tl), Bi4Ge3O12 and CdWO4 growth are addressed. Purity and growth control, dopant level effects, specific imperfections and plastic deformation in large scale commercial production are considered.

The current possibilities as well as the limitations today of cadmium telluride photon counters and spectrometers are reviewed. The various types of detectors are considered together with their main characteristics. These detectors are compared with mercuric iodide counters and possible applications of these devices are explained.

Scintillation performance of recent (2X) grown BGO is studied. Results indicate BGO to perform better than 8% 137Cs FWHM NaI(Tl) at energies 2.6 Mev and higher. Even the low energy performance of BGO is suitable for considering it in lieu of NaI(Tl) for many spectroscopic applications.

Silicon avalanche photodiodes have recently been shown to be a potential replacement for vacuum tube photomultipliers in many nuclear scintillation detector applications. The large active area, low noise, and ease of use of these solid-state photomultipliers makes them ideally suited to scintillation detector applications where overall size and ruggedness are a major concern. Historically, avalanche photodiodes have been limited for use in this capacity by small active areas, low internal gains, and poor optical sensitivity at the wavelengths at which most solid scintillator materials emit. Recent advances as the result of research aimed directly at the solution to these problems however, have successfully demonstrated one inch active area silicon avalanche photodiodes which produce a FWHM resolution of 9.5% for Cs137 at room temperature when coupled to a 1″ × 1″ NaI(Tl) scintillation crystal. Improvements to both material quality and device structure have advanced the state-of-the-art to make silicon avalanche photodiodes a viable alternative in scintillation gamma spectroscopy as well as for large area optical, beta, and low energy x-ray detectors.

Silicon nuclear radiation detectors are available today in a large variety of sizes and types. This profusion has been made possible by the ever increasing quality and diameter silicon single crystals, new processing technologies and techniques, and innovative detector design. The salient characteristics of the four basic detector groups, diffused junction, ion implanted, surface barrier, and lithium drift are reviewed along with the silicon crystal requirements. Results of crystal imperfections detected by lithium ion compensation are presented. Processing technologies and techniques are described. Two recent novel position-sensitive detector designs are discussed—one in high-energy particle track reconstruction and the other in x-ray angiography. The unique experimental results obtained with these devices are presented.

An overview of the present status and future requirements of material for germanium radiation detectors is presented. Fabrication and storage problems for both lithium-drifted, doped germanium and high-purity germanium are compared to demonstrate the reasons for the recent complete dominance of the latter in commercially available radiation detectors. The effect of electrically active point and line defects on the resolution and operating characteristic of high-purity germanium radiation detectors is discussed. Emphasis is placed on deep impurities and dislocations. Present and future applications of high-purity germanium radiation detectors are reviewed.

The detector material used in this experimments is Ultra-High Purity p-type Si crystal. The material was single-crystalized through floating zone process from poly-crystal grown by thermal decomposition of highly refined mono-silane gas which had been purified by molecular sieves of specially prepared Zeolite powder.

The resisitivity at room temperature is normally above 30 kΩ.cm, and the value of the highest grade ones exceeds 100kΩ.cm which corresponds to the Boron concentration of 1.5×l011 B/cm3.

The potentiality of this material for detector use was investigated through Surface Barrier Detector fabrication. Detectors of above 4 mm thick and of excellent characteristics both at room temperature and at liq. N2 temperature could readily be fabricated. Owing to tie extremely high resisitivity, detectors can be made either partially depleted or totally depleted simply by properly selecting the wafer resisitivity and the thickness.

In detector fabrication, the proper surface chemical treatment is very important, and seriously affects the leakage current and breakdown characteristics.

Ultra-High Purity Si is promising as new detector material and has good potentiality to replace Si(Li) as followings:

1) simple and easy detector fabrication,

2) potentiality of thick detector fabrication (∼1cm),

3) no precipitation problem of Li ions, and

4) feasibility of thick ⊿E detector with thin entrance window on both faces.

A remarkable characteristic of the surface barrier detector with a side groove cooled down to the nitrogen temperature has been observed. That is an abrupt decrease of the output noise at a definite voltage in increasing the bias voltage slowly. This is caused by a carrier injection from the side surface of the groove near the neck narrowed by the side groove. This excess noise disappears after the depletion layer goes through the neck of the side groove. This is confirmed by the fact that the capacitance of the detector decreases abruptly with the decrease of the noise at the same bias voltage. This detector is capable to be operated at a voltage higher than this voltage with a low noise. The maximum bias voltage applied to the detector is 3000 V.

Single crystal sections of HgI2 (about 1 cm thick) have been evaluated for charge carrier transport properties. Using these thick detector structures, surface effects produced during fabrication are reduced, enhancing the bulk property characteristics. The standard time-of-flight method was used to determine electron and hole mobilities. Lifetime measurements for electrons and holes were made by direct observation of the carrier decay where crystal transit times were long compared to lifetimes. Nonlinear charge carrier velocities have been observed during mobility measurements. These nonlinearities impair a partial charge collection technique [1,2,3] for spectra generation taken with thick detectors while having a lesser effect on the standard full charge collection approach. Partial charge collection methods have produced greater peak efficiencies than full charge collection, and investigations of crystal properties are being used to enhance this method.

Recent developments on HgI2 detectors fabricated from platelets grown by the Polymer Controller Growth (PCG) technique have resulted in a better understanding of this remarkable process and provide increased hope for the future of room temperature operable, high resolution x-ray detectors. The benefits of PCG are higher purity using reagent grade materials, better control of stoichiometry, rapid growth, and simplified fabrication. This latter benefit is even more significant when the extreme fragility of HgI2 crystals is considered. The problems of device resolution crystal stability and uniformity, and control ot platelet growth were investigated under a NASA funded program. HgI2 devices fabricated from PCG platelets gave room temperature energy resolutions under 400 eV (FWHM) for the 5.9 eV Mn x-ray. Models for the polymer assisted transport are discussed along with an analysis of spectra obtained from PCG grown devices. Reproducibility has been demonstrated both for platelet growth and device performance.

This paper describes the fabrication of mercuric iodide nuclear radiation detectors suitable for X and gamma ray spectrometry at room temperature. The active area of the detectors studied are between 0.2 and 1.5cm sq and they are up to 0.5mm thick. The method of producing a stable electrical contact to the crystal using sputtered germanium has been studied. The X-ray resolution of a 1.5cm sq. area detector at 32 keV is 2.3 keV FWHM when operated at room temperature in conjunction with a time variant filter amplifier. A factor which is important in the fabrication of the detector is the surface passivation necessary to achieve a useful detector life.

This type of detector has been used on a wavelength dispersive X-ray spectrometer for energy measurements between 10 and 100 keV. The advantages over the scintillation counter, more commonly used, is the improved resolution of the HgI2 detector and its smaller size. The analyser is primarily used for the detection of low levels of heavy metals on particulate filters. The detectors have also been used on an experimental basis for gamma ray backscatter measurements in the medical field.

Mercuric iodide is, potentially, an interesting material for room temperature high efficiency gamma or X-ray spectroscopy. However, still several parameters limit the performance of these devices and even the crystal growth is not fully under control today, poor reproductibility being reached.An optimization of growth conditions seems difficult since no parameter of the material could be correlated with the spectrometer's performance. Therefore, the effective progress in recent years was rather limited.

The goal of this paper is to investigate systematically if a correlation can be found between bulk stoichiometry for crystal grown under certain conditions, thermally stimulated current (TSC) measurements and nuclear performance. In this study, two families of crystals could be identified. Best results are obtained for nearly stoichiometric materials, exhibiting a specific TSC spectrum. In particular, the peak appearing at 173 K constitutes a good indication of the detection capabilities of the material, both for spectroscopic resolution and polarization.

The growth conditions have been adjusted to produce in a fully reproducible way high performance devices.

Ohmic and rectifying contacts on high resistivity etched P-type cadmium telluride have been studied in order to produce diode structures.For this,we have first investigated the properties of gold contacts obtained by chemical reactions of CdTe dippedin gold chloride.Both electrical characterization and structure have been analyzed as a function of the experimental conditions of the contact deposition.The results can be interpreted in terms of a current flow enhanced by tunnelling through the Au-CdTe junction and related to the structure of the interface a few tens of nanometer below the gold contact. In addition,several rectifying contacts have been investigated , in order to achieve a structure having low leakage current.

An about 108 ohm-cm AgGaSe2 crystal of 0.5 mm × 4 mm × 4 mm was polished and contacts were made by evaporating 130 μg/cm2 gold of 3 mm diameter on the two faces of the crystal. The detector was tested using 5.5 MeV alpha-particles at room temperature. Noise increased above an applied bias voltage of 80 V. For + 50 V applied on the electrode opposite to the particle incident electrode, i.e., for electron traversal mode, the preamplifier output pulses showed a risetime of 20 μs and an amplitude of about one-tenth of that obtained from a silicon surface-barrier detector whereas the silicon detector showed a risetime of 0.07 μs. For - 50 V applied on the same electrode, i.e., for hole traversal mode, no pulses were observed.