Rex C. Trammell

THE EFFECTS OF CARRIER TRAPPING IN SEMICONDUCTOR GAMMA-RAY SPECTROMETERS.

Abstract Currently accepted theories predicting the effects of carrier trapping on the spectrum from semiconductor gamma-ray spectrometers are reviewed. The physical models used in these theories are re-examined and a new theory is presented which predicts quite different effects.

Radiation Damage Resistance of Reverse Electrode GE Coaxial Detectors

Two high-purity germanium coaxial detectors, having opposite electrode configurations from one another, but fabricated from the same germanium crystal, were irradiated simultaneously with fast neutrons from an unmoderated 252Cf source. Both detectors were 42 mm diam. The detector having the conventional electrode configuration was about 28 times more sensitive to radiation damage than was the detector having the p+ contact on the coaxial periphery. These results prove that germanium coaxial detectors having the conventional electrode configuration should not be used in any situation subject to significant radiation damage. This conclusion was anticipated because the defects produced by neutron and proton irradiation of germanium act predominantly as hole traps.

Mechanism for fast neutron damage of Ge(HP) detectors

Abstract The effect on high purity germanium detectors of the disordered regions (r ∼ 100 A ) created by fast neutrons is theoretically and experimentally addressed. The hole trapping cross section of these defects is a function of their net negative charge and the applied electric field. We estimate σ ≈ 10−9 – 10−10 cm2 immediately after biasing an n-type detector and σ ≈ 10−10 – 10−12 cm2 as hole trapping and detrapping reach a steady state in the depleted detector. Resolution transients observed immediately after biasing n-type and p-type Ge(HP) coaxial detectors are reported and are shown to be consistent with the neutralization (n-type) or the charging (p-type) of the thermal equilibrium state. However on the basis of these transients we cannot exclude the possibility that point defects play a decisive role in the steady state resolution degradation. The ionization or activation of traps after cycling the p-type detector off/on can be consistently interpreted as due to an acceptor level near Ev + 0.27 eV. The duration of the transient observed in n-type germanium was reduced to less than 1 h by placing a 6.5 μCi 60Co source on the end cap. Thus this transient does not significantly diminish the advantage of the contracting polarity employed on n-type coaxial detectors in reducing the effect of the hole trapping on resolution.

Deterioration of large Ge(Li) diodes caused by fast neutrons

Abstract The large Ge(Li) gamma-ray detector has become a powerful tool for the investigation of nuclear reactions. Unfortunately these detectors are quite susceptible to fast neutron damage and this makes it difficult to decide whether or not to risk using a detector to study reactions at accelerators where fast neutrons are inevitably present. After ruining several detectors we decided to study the problem in a controlled way. A 30 cm 3 true coaxial diode was systematically irradiated by neutrons from a plutonium-beryllium source. An increase in the width of the 2.614 MeV gamma-ray from 208 Tl was first detected after an irradiation of 5 × 10 7 n/cm 2 . When the total irradiation had reached 6 × 10 8 n/cm 2 , the peak width had increased by more than 50 percent. The irradiated detector was then reprocessed to remove the damage. The diode was again subjected to neutron irradiation. The second curve of resolution deterioration as a function of neutron flux was quite similar to the first one. The procedure was repeated a third time with similar results. Thus, reprocessing a detector effectively removes the neutron damage at a cost of only about 10 to 15 percent of the original price. A method was also developed for using the gamma-ray spectrum to evaluate the amount of neutron flux incident on the detector. The number of counts in the 693 keV peak [from (n,n′) reaction with the 72 Ge in the detector] can be multiplied by 20 to get a rough measure of the neutron flux in n/cm 2 .

Performance of Reverse Electrode HPGE Coaxial Detectors after Light Damage by Fast Neutrons

Several reverse electrode HPGe coaxial detectors fast neutron-damaged to various fluences were annealed at room temperature for varying periods of time. For comparison, in one case a dry ice temperature anneal was made. The result of these annealings on the energy resolution at 1.33 MeV was measured. The effect of source intensity and energy on the energy resolution and line shape are discussed.

Neutron Damage in Ge(Hp) Coaxial Detectors

We report energy resolution, capacitance, and depletion voltage transients in neutron damaged coaxial Ge(HP) detectors fabricated from n-type and p-type germanium. The disordered region model for fast neutron degradation of germanium detectors was supported by 1) the large trapping cross sections observed (~ 10-11 cm2), 2) changes in the steady state charge on the traps over four decades of induced current density, and 3) the inability of the isolated defect hypothesis to self-consistently account for the ratio of thermal emission of holes in liquid nitrogen to liquid argon cooled detectors. The results of in-cryostat annealing of these detectors are reported.

An ultra-high-throughput, high-resolution gamma-ray spectroscopy system

A high-throughput gamma-ray spectroscopy system is described. It consists of the following: a high-purity Ge (HPGe) detector; a transistor-reset preamplifier (TRP); a gated integrator, shaping amplifier; an analog-to-digital converter (ADC); and digital control and storage circuits. The gated integrator uses a quasi-rectangular prefilter. With the detector used, this prefilter provided approximately 16% less noise than a gated integrator using a conventional Gaussian prefilter. The ADC has 14-b accuracy, differential nonlinearity >

High-rate and high-energy gamma-ray spectroscopy using charge trapping and ballistic deficit correction circuits

A study of the resolution of large, coaxial, reverse electrode, HPGe detectors was performed over the energy range from 100 keV to 10 MeV and triangular amplifier shaping times from 0.5 mu s to 6 mu s. Resolutions were calculated using an approach based on the Trammell-Walter equation. The effect of ballistic deficit was included in the calculations by the introduction of a term, the ballistic efficiency, to the Trammell-Water equation. Experimental data were collected over an energy range from 122 keV to 2.6 MeV on three detectors with relative efficiencies of 76%, 56%, and 29%. For these three detectors, the data indicated that the triangular shaping amplifier with charge trapping and ballistic deficit correction offered better resolution than a gated integrator for shaping times >or=2 mu s, while the gated integrator produced better resolution for shaping times >

Comparative Timing Performance of Large Volume Hpge Germanium Detectors

The performance characteristics of a timing spectrometer suitable for use with large volume germanium detectors was studied. Timing resolution was measured as a function of constant fraction shaping delay, timing filter amplifier shaping time constants, discriminator threshold setting, and detector bias. Timing resolution data was collected for 14 detector-preamplifier systems including 9, coaxial HPGe p-type detectors, and 5 coaxial HPGe n-type detectors. Timing resolutions at FWHM and FWTM as a function of energy are presented for the energy range 150 + 5OkeV to 1330 + 50keV.

Characteristics of High-Rate Energy Spectroscopy Systems Using HPGe Coaxial Detectors and Time-Variant Filters

A high-rate, high-resolution gamma spectrometer system is described. The system consists of a reverse electrode HPGe coaxial detector, a transistor reset preamplifier, an active, semi-Gaussian prefilter, a gated integrator, and a unique data acquisition system consisting of a 10 ¿s, 13 bit ADC, fast FIFO memory, 8k by 23 bit data memory, and computer interface circuitry under the control of a Z-80A ¿P. The effects of the various components on the throughput are described and throughput data is presented. The resolution and peak shift for various shaping times are presented for count rates up to 1 Mcps input rate using a mixed 22Na and 60Co source. The low rate resolutions of 57Co and 60Co for various shaping times using either the semi-Gaussian or gated integrator output are discussed as well as the low energy resolution and peak shifts in the presence of high energy events.