WO1999003155A1 - Detecteur a ionisation haute resolution et reseau de tels detecteurs - Google Patents

Detecteur a ionisation haute resolution et reseau de tels detecteurs Download PDF

Info

Publication number
WO1999003155A1
WO1999003155A1 PCT/US1998/014055 US9814055W WO9903155A1 WO 1999003155 A1 WO1999003155 A1 WO 1999003155A1 US 9814055 W US9814055 W US 9814055W WO 9903155 A1 WO9903155 A1 WO 9903155A1
Authority
WO
WIPO (PCT)
Prior art keywords
detector
region
grid
detectors
frisch
Prior art date
Application number
PCT/US1998/014055
Other languages
English (en)
Inventor
Douglas S. Ph. D. Mcgregor
Ronald A. Rojeski
Original Assignee
The Regents Of The University Of Michigan
Sandia Corporation
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US09/075,351 external-priority patent/US6175120B1/en
Application filed by The Regents Of The University Of Michigan, Sandia Corporation filed Critical The Regents Of The University Of Michigan
Priority to AU82928/98A priority Critical patent/AU8292898A/en
Publication of WO1999003155A1 publication Critical patent/WO1999003155A1/fr

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01T—MEASUREMENT OF NUCLEAR OR X-RADIATION
    • G01T1/00—Measuring X-radiation, gamma radiation, corpuscular radiation, or cosmic radiation
    • G01T1/16—Measuring radiation intensity
    • G01T1/24—Measuring radiation intensity with semiconductor detectors
    • G01T1/241—Electrode arrangements, e.g. continuous or parallel strips or the like
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L31/00—Semiconductor devices sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation; Processes or apparatus specially adapted for the manufacture or treatment thereof or of parts thereof; Details thereof
    • H01L31/08—Semiconductor devices sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation; Processes or apparatus specially adapted for the manufacture or treatment thereof or of parts thereof; Details thereof in which radiation controls flow of current through the device, e.g. photoresistors
    • H01L31/10—Semiconductor devices sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation; Processes or apparatus specially adapted for the manufacture or treatment thereof or of parts thereof; Details thereof in which radiation controls flow of current through the device, e.g. photoresistors characterised by potential barriers, e.g. phototransistors
    • H01L31/115—Devices sensitive to very short wavelength, e.g. X-rays, gamma-rays or corpuscular radiation

Definitions

  • This invention relates to ionization detectors and, in particular, to high-resolution ionization detectors and an array of such detectors.
  • Ionization detectors such as semiconductor radiation detectors have been studied since 1945, in which tiny AgCl crystals cooled to low temperatures were used to observe radiation induced electrical pulses. Such devices were impractical due to the stringent cooling requirements and their diminutive size.
  • Si(Li) detectors must be cooled in order to reduce the concentration of thermally excited charge carriers that add noise to the measurement, thereby reducing the energy resolution.
  • Ge(Li) devices are much more severe, in which the crystal must remain cooled to 77K to prevent redistribution of the Li in the device (through diffusion), as well as reduce thermal noise.
  • Ge(Li) detectors are ruined if they are allowed to warm up to room temperature, and the devices can only be re-drifted once or twice before severe gamma ray energy resolution degradation appears.
  • Zone refinement of Ge eventually eliminated the need for Li drifting in Ge material to produce high-energy resolution gamma ray detectors.
  • the high purity Ge (HPGe) detectors do not need to be constantly cooled, but must be cooled to 77K during operation to reduce thermal noise and to prevent harmfully high leakage currents from passing though the detector. Hence, the cryogenic requirement remains.
  • Float zone refinement of Si also has worked to produce high quality material for charged particle detectors, but Si based X-ray detectors are still manufactured from Li drifted material.
  • One method to increase intrinsic gamma ray interaction efficiency in a semiconductor radiation detector is to utilize semiconductors composed of relatively high atomic number elements. Such a condition becomes practical only if the material of choice is available in reasonably large volume.
  • a method to reduce the cryogenic cooling requirement is to use a semiconductor with a relatively large band gap (1.4 eV - 2.5 eV) such that the concentration of thermally produced charge carriers is minimal at room temperature.
  • a reasonably competitive semiconductor material for a room temperature operated gamma ray spectrometer would be composed of materials with high Z values and have a wide band gap.
  • the only single element semiconductors are carbon (diamond - C), silicon (Si), germanium (Ge) and tin (Sn).
  • CdZnTe material only recently has become commercially available, and the material showed promise as a possible room temperature operated gamma radiation spectrometer. With a band gap near 1.6 eV and material atomic numbers of 48, 30 and 52, CdZnTe has the fundamental properties required for a room temperature operated semiconductor detector. Permutations of the single polarity electron sensing idea have been realized with CdZnTe material with interesting results. For instance, the co-planar grid concept allowed for the realization of fairly high-resolution CdZnTe detectors at room temperature with energy resolution of 3.7% FWHM at 662 keV for a 5mm x 5mm x 5mm device (125 mm 3 ) as described in the U.S. patent to Luke 5,530,249.
  • An object of the present invention is to provide an improved ionization detector and an array of such detectors having improved energy resolution (i.e. measurement accuracy) without the need for electronic pulse rejection compensation and correction techniques.
  • Another object of the present invention is to provide an improved ionization detector and an array of such detectors which are relatively inexpensive to make and are operable at room temperature.
  • Still another object of the present invention is to provide a Frisch grid semiconductor radiation detector and an array of such detectors wherein only a single standard charge-sensitive pre-amplifier need be provided without the need for a custom pre-amplifier or coupling circuit in a measuring circuit.
  • a high-resolution, ionization detector includes an ionization substrate having a first surface and a second surface opposing the first surface, a first electrode disposed at the first surface and a second electrode disposed at the second surface.
  • the detector also includes a reference pattern of conductive or semi-conductive material which divides the ionization substrate into an interaction region, a measurement region and a pervious region which separates the interaction and measurement regions. The pervious region defines a reference plane to measure charge carrier movement in the measurement region and directs charge carrier flow within the substrate.
  • Voltage potential circuits apply voltage potentials to the reference pattern of material and to the first and second electrodes so that a bias voltage is applied across substantially the entire substrate and so the voltage potential applied to the reference pattern is between the voltage potentials applied to the first and second electrodes.
  • Charge carriers of a first polarity type move from the interaction region, through the pervious region and into the measurement region.
  • the detector further includes a signal measurement circuit for measuring at the second electrode or signal generated by the charge carriers of the first polarity type moving from the interaction region, through the pervious region and within the measurement region between the pervious region and the second surface.
  • the reference pattern is a grid such as a Frisch grid having at least two substantially parallel strips of conductive or semi-conductive material.
  • the substrate also has a third surface and a fourth surface opposing the third surface between the first and second surfaces and wherein the Frisch grid is a Frisch grid ring disposed at the third and fourth surfaces.
  • the reference pattern includes a first strip disposed at the third surface and a second strip disposed at the fourth surface.
  • the ionization detector is a semiconductor, single-charge carrier, radiation detector wherein the semiconductor substrate is a semiconductor substrate which operates as a room temperature, gamma ray detector.
  • the signal measurement circuit may include a single preamplifier circuit.
  • the first electrode can be substantially larger than the second electrode to take advantage of the "small pixel" effect.
  • the substrate may be shaped like a trapezoid, a pyramid frustum, a conical frustum, or any other shape where the substrate is geometrically weighted.
  • the first surface may be curved to allow for more uniformity with charge carrier transport within the interaction region.
  • the second surface has a plurality of trenches formed therein separated by towers wherein the reference pattern is formed in the trenches and the second electrode is formed on the towers so that the reference pattern is a Frisch grid.
  • an array of detectors is provided.
  • the substrates of the detectors are formed from a single crystal or amorphous material.
  • an array of the detectors has the substrates of the detectors formed from separate crystals or amorphous materials.
  • the detectors may be arranged so that the array forms a substantially coaxial or spherical device.
  • the semiconductor Frisch grid radiation detector has three distinct regions which are labeled (1) the interaction region, (2) the pervious region, and (3) the measurement region.
  • the interaction region of the semiconductor Frisch grid detector is typically the largest of the three regions. It is larger than the measurement region in volume and it is typically larger than the pervious region in volume. The interaction region may be larger than the combined volume of both the measurement region and the pervious region.
  • the main function of the interaction region is to provide volume or mass for forms of ionizing radiation to interact within. Ionizing radiation include (1) electromagnetic waves and photons, such as gamma rays and X-rays, (2) charged particles, such as alpha particles, heavy ions, beta particles, positrons, and electrons, and (3) neutrons.
  • Ionizing radiation that interacts within the interaction region excites electron-hole pairs in the materials such that both types of charge carriers become mobile and freely conducting.
  • Applied voltage or electric field allows for the electrons to be driven toward one direction and the holes in another.
  • the voltage may be applied such that the electrons or holes are driven toward the pervious region and measurement region.
  • the pervious region acts as a reference plain for charge carriers
  • the pervious region is the location of the Frisch grid, which acts as the reference plane by which the induced charge is measured as charge carriers transit between the pervious region center through the measurement region to the measurement region contact.
  • the pervious region Frisch grid may be applied through contacting or non-contacting outside contacts, through an embedded or buried structure, through a thin porous structure, or through a trenched structure.
  • the pervious region will typically be smaller than the interaction region in volume.
  • the pervious region may be smaller than the measurement region volume, the same volume as the measurement region volume, or a greater volume than the measurement region volume.
  • the pervious region will most likely be formed of the same material as the interaction and measurement region, but may also be formed from doped semi-conducting material or different materials than that of the interaction region and pervious region.
  • the interaction region may be composed of low doped p-type GaAs
  • the pervious region may be composed (at least partially) of AlGaAs
  • the measurement region may be n-type GaAs.
  • the Frisch grid connected to the pervious region may be constructed by a number of different means.
  • the Frisch grid acts as a reference electrode to measure induced charge as free charge carriers move between it and the outer boundary of the measurement region. Examples of implementing a Frisch grid with the pervious region are as follows:
  • the Frisch grid is formed through outside contacts .
  • an outer electrode is fabricated onto a material or semiconductor such that an external electrode is fastened to the device.
  • the external electrode(s) may be a set of parallel conductive strips connected to the pervious region.
  • the external electrode may be a continuous contact that forms concentrically around the pervious region.
  • the external electrode(s) may be a broken line of conductive strips that are in contact with the pervious region ( Figure 35).
  • the external electrode may be composed of a metal substance or a semiconductor substance.
  • the external electrode(s) may be an ohmic contact or a rectifying contact composed of n-type or p-type material.
  • the Frisch grid is formed by a pattern of trenches or depressed regions with conducting material or semi-conducting material in the trenches or depressed regions ( Figures 34, 37a-c, 38, 43).
  • the pervious region is the region adjacent the Frisch grid. Charge carriers are driven from the interaction region, through the trenches or depressed region pervious grid and into the measurement region(s).
  • the grid may be formed with metallic electrodes or from semiconducting material.
  • the semiconducting material may be p-type or n-type.
  • the grid may form a rectifying contact to the bulk material, thereby reducing leakage current and charge carrier losses into the grid.
  • the Frisch grid is formed through outside conductors that are not in intimate contact with the detecting material ( Figures 39, 40, 41 ,
  • the outside non-contacting Frisch grid may be composed of conducting plates adjacent to the pervious region ( Figures 39, 40) or a concentric conducting electrode ( Figures 41, 42) surrounding the pervious region.
  • the scheme uses a conductive ring or concentric electrode that is not in direct contact with the detector material as the Frisch grid around the pervious region.
  • the space between the Frisch grid ring or concentric contact may be filled with a vacuum, air, or an insulating material ( Figures 40, 42).
  • the insulating material may be in intimate contact with the detecting material.
  • the Frisch grid is formed through a buried conductive grid region (i.e. Figure 3).
  • the metallic or semiconducting grid region is buried in the device and acts as the Frisch grid.
  • the pervious region is the non-grid region plane in which the grid lies.
  • the buried grid may be formed by a number of different means including diffusion, ion drifting, ion implantation, chemical vapor deposition growth and any of the various means of epitaxial growth.
  • the Frisch grid is formed through a very thin film of conducting or semiconducting material that is buried in the detector ( Figure 44).
  • the device has a membrane or plane of conducting or semiconducting material that is very thin, on the order of only a few atomic spacings.
  • the plane acts as both the pervious region and the Frisch grid reference plane. Charge carriers are accelerated toward the plane from within the interaction region. Their high velocity allows them to punch through or tunnel through the very thin plane and into the measurement region.
  • the measurement region acts as the region in which free charge carriers (electrons or holes) excited by ionizing radiation are swept into by an electric field.
  • the measurement region is the region in which the main portion of induced charge is measured as charged carriers move through its volume to the measurement region boundaries.
  • the measurement region volume is designed to be smaller than the interaction region volume.
  • the main portion of induced charge signal is induced on the signal electrode as free charge carriers leave the pervious region and enter into the measurement region.
  • the ratio of the combined volumes of the pervious region and interaction region should be greater than the measurement region volume.
  • V j + V p should be greater than or equal to V m , or
  • the geometrically weighted semiconductor Frisch grid detector uses three fundamental effects to increase the radiation energy resolution.
  • the design uses the:
  • A. Geometrical Weighting Effect The shape of the device is tapered such that one end is larger than the other. An electrode or an array of electrodes are placed on the small end. An electrode or an array of electrodes are placed at the large end. The shape of the device increases the overall volume while also increasing the ratio of
  • the number of charge carriers originating in the interaction region and moving toward the measurement region are much greater than the number of opposite polarity charge carriers originating in the measurement region and moving toward the interaction region.
  • a Frisch grid is implemented using one of the previously described techniques nearest the small end of the device.
  • the Frisch grid acts to screen charge carrier motion in the interaction region.
  • the angle of the device can range from 0 degrees, which describes a basic planar type semiconductor Frisch grid radiation detector, to a very wide angle up to 180 degrees (Figure 43).
  • prototype devices have a small end angle ⁇ of 45° that perform very well.
  • the devices can have small end angles that increase up to 90° and still perform well with the outside contact style Frisch grid device structure described above. Increasing the angle above 90° to a range between 90° and 180° will still produce a working three region device, yet the effective screening of the Frisch grid diminishes as the angle increases for a wide angle device as described above. Rounding the bottom of the geometrically- weighted device improves performance by reducing deviation in the transit length of charge carriers moving from the far cathode region ( Figure 33).
  • FIGURE 1 is a schematic diagram of a semiconductor single carrier spectrometer (SSCS) constructed in accordance with the present invention
  • FIGURE 2 is a diagram similar to Figure 1 for use with an electron dominant device
  • FIGURE 3 is a diagram similar to Figures 1 and 2 for an electron dominant device
  • FIGURE 4 is a sectional view of the detector of Figure 3;
  • FIGURE 5 is an enlarged sectional view of the part of the biased detector of Figure 4 within dotted lines and showing contour bias lines;
  • FIGURE 6 is a view similar to Figure 5 except showing electric field lines which frame the paths that a charge carrier will follow;
  • FIGURE 7 is an enlarged sectional view of another part of the biased detector of Figure 4 with other dotted lines and showing contour bias lines;
  • FIGURE 8 is a view similar to Figure 7 except electric field lines are shown;
  • FIGURE 9 is a simple depiction of a semiconductor radiation detector; a bias is placed across a block of semiconducting material; gamma rays excite electron-hole pairs, and the electrons and holes are swept apart by the electric field; the displacement of the charge carries induces a current to flow through the external circuit, which is integrated to form a voltage pulse relative in size to the total integrated current that flows as a result of the charge carrier displacement;
  • FIGURE 10 is a charge induction map for a planar detector in which the hole extraction factor is 0.5 and the electron extraction factor is 5; shown is the induced charge as a function of the gamma ray interaction location; gamma ray absorption attenuation and Compton scattering are not considered; horizontally flat regions in the total induced charge on the charge induction map indicate energy peak formation in the pulse height spectrum, whereas sloped regions indicate long tail regions or continua in the pulse height spectrum; a
  • FIGURE 11 is an expected pulse height spectrum from a planar detector with the characteristic charge induction map of Figure 10; Compton scattering and gamma ray attenuation are neglected in the simulation; the "full energy peak" appears at a channel lower than expected for full charge collection due to electron and hole trapping; the tail appears as a consequence of position dependent charge collection;
  • FIGURE 12a shows the basic configuration for a Frisch grid gas detector; the gas detector volume is separated into an interaction region and a measurement region; electron charge carriers are drifted toward the anode by an applied bias;
  • FIGURE 12b shows the induced charge that appears at the anode is zero until the electrons pass through the grid; after passing through the grid, a charge is induced proportional to the distance traveled by the charge carriers between the grid and the anode;
  • FIGURE 13 shows the basic configuration for a parallel strip semiconductor Frisch grid radiation detector
  • FIGURES 14a and 14b show the interaction region is much larger than the measurement region; as a result, most of the measured induced charge appears from the electrons drifted from the interaction region into the measurement region and not from electron-hole pairs excited in the measurement region, as illustrated in Figure 14b;
  • FIGURES 15a, 15b, and 15c show the weighting potential distributions of three parallel strip devices with dimensions of (a) 5mm x 2mm x 5mm, (b) 5mm x 3mm x 5mm, and (c) 5mm x 5mm x 5mm, respectively; the Frisch grid strips are placed 1mm back from the anode in all cases, and the strip is 0.5mm wide. It is clear that the Frisch grid screening is best for the thinnest detector, and poor for the cube-shaped device;
  • FIGURES 16a, 16b, and 16c show the weighting potential distributions of three parallel strip devices with dimensions of (a) 10mm x 2mm x 10mm, (b) 10mm x 4mm x 10mm, and (c) 10mm x 6mm x 10mm; respectively; the Frisch grid strips are placed 2mm back from the anode in all cases, and the strip is 1.0 mm wide.
  • the Frisch grid screening is best for the thinnest device, and is even better than the 5mm x 2mm x 5mm device as shown in Figure 15a; the weighting potential distribution of the 10mm x 4mm x 10mm device is substantially identical to the 5mm x 2mm x 5mm device shown in Figure 15a;
  • FIGURE 17 shows room temperature spectra of 662 keV gamma rays from the 5mm x 2mm x 5mm CdZnTe semiconductor parallel strip Frisch grid detector with the grid “off” (floating) and "on” (-80 volts);
  • FIGURE 18 shows room temperature spectra of 662 keV gamma rays from the 5mm x 2mm x 5mm CdZnTe semiconductor parallel strip Frisch grid detector with the grid “off” (floating) and "on” (-80 volts); the detector has been shielded with a lead attenuator such that gamma rays interacted only in the interaction region;
  • FIGURE 19 shows room temperature spectra of 662 keV gamma rays from the 10mm x 2mm x 10mm CdZnTe semiconductor parallel strip Frisch grid detector with the grid “off” (floating) and "on” (-70 volts);
  • FIGURES 20a, 20b, and 20c are three possible designs for a geometrically weighted semiconductor Frisch grid radiation spectrometer; shown are (a) a trapezoidal design, (b) a pyramid frustum, and (c) a conical frustum, respectively;
  • FIGURE 21 is a diagram of the general features for a geometrically weighted semiconductor Frisch grid radiation detector
  • FIGURE 22 shows weighting potential for a CdZnTe trapezoid prism detector without a Frisch grid; a distance of 10mm separates the small anode from the much larger cathode; charge induction is greatest in the region nearest the small anode, an effect known as the "small pixel effect";
  • FIGURES 23a and 23b show a comparison of charge induction maps for a (a) CdZnTe planar radiation detector and a (b) CdZnTe trapezoid prism detector, respectively; a distance of 10mm separates the anode from the cathode in both cases;
  • FIGURE 24 shows modeled results of the expected pulse height spectrum from a CdZnTe planar detector and a CdZnTe trapezoid prism geometrically weighted detector; the distance between the anode and the cathode in both cases is 10mm; Compton scattering and gamma ray attenuation were excluded from the model;
  • FIGURE 25 is a modeled voltage dependent pulse height spectrum of a mono-energetic gamma ray source from a CdZnTe trapezoid prism geometrically weighted detector; the distance between the anode and the cathode is 10mm; there is a unique formation of dual peaks at low voltages and a pseudo- peak at high voltages; Compton scattering and gamma ray attenuation were excluded from the model;
  • FIGURE 26 is the observed voltage dependent pulse height spectrum of 122 keV gamma rays from a CdZnTe trapezoid prism geometrically weighted detector with the Frisch grid turned off; the distance between the anode and the cathode is 10mm; the dual peaking phenomenon is clearly shown; it is suspected that the voltage distribution is very low across most of the detector due to bad electrical contacts;
  • FIGURE 27 is the observed voltage dependent pulse height spectrum of 662 keV gamma rays from a CdZnTe trapezoid prism geometrically weighted detector with the Frisch grid turned off; the distance between the anode and the cathode is 10mm; dual peaks form at low voltages, and a pseudo-peak forms at high voltages; the anode was held at 200 volts and the cathode voltage was changed for the experiment;
  • FIGURE 28 is a weighting potential for a CdZnTe trapezoid prism detector with the Frisch grid turned on; a distance of 10mm separates the small anode from the much larger cathode; the Frisch grid works to effectively screen induced charge from charge carrier motion in the interaction region; hence, it is primarily the charges moving in the measurement region between the Frisch grid and the anode that will induce charge on the external pre-amplifier circuit;
  • FIGURE 29 shows comparison spectra of 511 keV gamma rays from 2 Na as measured with a CdZnTe trapezoid prism detector with the Frisch grid turned off and on; shown is the distinct advantage to having the Frisch grid on such a device; the resolution improvement is significant and results from the composite effects of the geometrical weighting, the small pixel effect, and the Frisch grid effect;
  • FIGURE 30 is a schematic view of a trapezoid Frisch grid detector set up for gamma ray spectroscopy; the simple connections are rugged and easy to implement;
  • FIGURE 31 is a schematic view of electronic components and set up used to test trapezoid Frisch grid detectors
  • FIGURE 32 is a possible pseudo-coaxial array of geometrically weighted parallel strip Frisch grid detectors; a tungsten shield can be inserted into the coaxial central hole to provide gamma ray shielding from oppositely placed detectors, thereby allowing the device to accumulate position information of external neutron sources; an outer lead collar can be used to help reduce external gamma ray background;
  • FIGURE 33 is a sectional view of a spherical array of detectors each of which has a rounded based;
  • FIGURE 34 is a schematic view of a geometrically- weighted, etched-trench, Frisch grid single carrier radiation spectrometer
  • FIGURE 35 is a perspective view of a detector of the present invention having concentric and segmented Frisch grid strips;
  • FIGURE 36 is a perspective view of a detector of the present invention having relatively wide parallel Frisch grid strips
  • FIGURES 37a-37c are side elevational perspective views of three different trenched type semiconductor Frisch grid radiation detectors;
  • FIGURE 38 is a perspective view of a segmented depression or trench type semiconductor Frisch grid detector;
  • FIGURE 39 is a perspective view of a detector including a semiconductor substrate which does not contact external Frisch grid strips;
  • FIGURE 40 is a perspective view similar to Figure 39 including insulating material between the pervious region of the substrate and the external strips;
  • FIGURE 41 is a perspective view of a cylindrical detector including a non-contacting Frisch grid ring
  • FIGURE 42 is a view similar to Figure 41 including insulating filling or material between the pervious region and the Frisch grid ring;
  • FIGURE 43 is a side elevational view of a wide angle (i.e. 180°) detector with an etched-style Frisch grid;
  • FIGURE 44 is a side view of a detector of the present invention wherein the reference pattern is a membrane buried in the semiconductor substrate .
  • the detector 20 includes a pervious or reference pattern 22 of conductive or semiconductive material based on a Frisch grid onto or into a semiconductor ionization substrate 24, thereby separating the electron and hole induced charge signals.
  • the reference pattern 22 divides the substrate into an interaction region 26, a pervious region 28 (i.e. under the reference pattern 22) and a measurement region 30.
  • the detector 20 also includes a first contact or electrode 32 disposed at a first surface 34 of the interaction region 26 and a second contact or electrode 36 disposed at a second surface 38 of the measurement region 30.
  • radiation quanta interact in the interaction region 26 of the detector 20. Since the interaction region 26 is much larger than the measurement region 30, more interactions occur in the interaction region 26.
  • carrier type 1 i.e. electrons
  • carrier type 2 i.e. holes
  • Interactions can occur anywhere in the detector 20, but since the interaction region 26 is larger than the measurement region 30, most of the interactions occur in the interaction region 26.
  • the ratio of the interaction region volume to that of the measurement region volume gives the ratio of the number of events that occur in the interaction region 26 to that in the measurement region 30.
  • a detector designed with the interaction region 26 being 1 cm thick and the measurement region 30 being 20 microns thick would have a ratio of 500: 1 , which means that for every interaction that occurs in the measurement region 30, it is expected that 500 will occur in the interaction region 26.
  • Electron-hole pairs excited in the interaction region 26 are drifted in opposite directions, in which case one of the carrier types drift toward the reference pattern 22 and the measurement region 30 and the other carrier type drifts in the opposite direction.
  • a voltage is applied across the interaction region 26 and a voltage is also applied across the measurement region 30.
  • VI, V2 and V3 represent different voltages applied at the region 26, the region 28, and the region 30, respectively.
  • Carriers that enter the measurement region 30 induce a charge through a potential held across the measurement region 30, hence becoming the measured charge. Since only one type of carrier drifts into the measurement region 30, the other no longer contributes to the measured signal.
  • the detector 20 is therefore a semiconductor single charge carrier device (SSCCD) or a semiconductor single carrier spectrometer (SSCS) since it needs only to measure the charge from one type of carrier transiting in the detector 20.
  • the detector 20 can be tailored such that the one carrier has a high CEF (i.e. "Carrier Extraction Factor”) value and renders high resolution by simply altering the interaction region width.
  • CEF Carrier Extraction Factor
  • FIG 2 one such method of realizing the detector of the present invention would be to construct a two region diode as follows and as shown in Figure 2, wherein items having the same or similar structure or function as in Figure 1 have the same reference numeral, but a single prime designation.
  • One side of a device 20' would be highly doped p-type material.
  • the interaction region 26' would be lightly doped n-type material 32'.
  • the reference pattern 22' would be highly doped p-type material and the measurement region 30' would be n-type material. Highly doped n-type material would make up the back contact 36' to the measurement region 30'.
  • the device 20' could also be made a number of different ways, such as replacing the above-mentioned p regions with n regions and the above- mentioned n regions with p regions.
  • the reference pattern 22' may in fact be a metal instead of a semiconductor. It is also possible to replace region 32' and region 36' with a metal.
  • the SSCS Device 20' as a Radiation Spectrometer
  • the SSCS device 20' can be operated as a radiation spectrometer in the following fashion.
  • a voltage is placed across the interaction region 26' and a voltage is placed across the measurement region 30'. Due to the larger volume of the interaction region 26', radiation interactions have a higher probability of occurring in the interaction region 26' than the measurement region 30'. Radiation quanta interacting in the interaction region 26', as indicated at 40', excite electron-hole pairs.
  • the polarity of the bias voltage is placed such that the carriers with the highest CEF drift toward the measurement region 30' .
  • electrons are the carriers with the highest CEF although holes may be the carrier with the highest CEF as well. If the electrons are the carriers with the highest CEF, the voltage at the end contact 32' of the interaction region 26' is negative with respect to the pervious division region 28' . As a result, the electrons drift toward the pervious region 28'. The movement of charge carriers in the interaction region 26' does not need to be monitored.
  • the electrons pass through the pervious region 28' they enter the voltage region between the pervious region 28' and the back contact 36' to the measurement region 30' .
  • the potential drop is monitored as the electrons transit through the capacitance of the measurement region 30' and the induced charge is recorded. Since holes do not move into the interaction region, they do not contribute to the signal of measured charge, as indicated at 42'.
  • VI is a negative voltage with respect to the pervious region 28'.
  • the junction at region 32' and region 26' is reversed biased.
  • V2 is a positive voltage with respect to the grid 22'.
  • the junction at the pervious region 28' and the region 36' is reversed biased.
  • the measured charge is due only to electron motion and hole motion is no longer a factor, which makes the device 20' a semiconductor single carrier spectrometer.
  • This type of device can be made such that holes are the carriers measured instead of electrons.
  • FIGs 3 and 4 show one design for a semiconductor single carrier spectrometer (SSCS) 20" wherein items having the same or similar structure and/or function have the same reference numeral as in Figures 1 and 2, but have a double prime designation.
  • SSCS semiconductor single carrier spectrometer
  • Figure 3 shows the general construction of one such device.
  • region 22" is a highly doped p-type contact or region. Radiation interactions occur preferentially in the interaction region 26' ' and excite electron- hole pairs.
  • the interaction region 26" is low doped n-type, v-type, or intrinsic material.
  • the pervious region 28" is composed in this example of a p-type grid 22 of lines 23" buried in the material.
  • the measurement region 20' ' consists of n-type or intrinsic material.
  • the back contact 36" is made from highly doped n- type material.
  • One method of operating the SSCS 20" is to hold the grid 22" at ground and apply negative voltage to region 32" . Positive voltage is applied to region 36" with respect to the grid 22" . Other methods can be used as well.
  • region 32" can be grounded, region 22' ' biased with positive voltage, and region 36' ' biased with a higher positive voltage than region 22" .
  • Another method of accomplishing the bias may be to ground region
  • the device 20' ' might be designed as follows.
  • the region 32' ' would be a highly doped n-type region.
  • the interaction region 26" is low doped p-type, ⁇ -type, or intrinsic material.
  • the pervious region 22" composed in this example of a n-type grid 22" of lines 23" buried in the material.
  • the measurement region 30" consists of p-type or intrinsic material.
  • the back contact 36" is made from highly doped p-type material.
  • FIG. 4 shows two areas enclosed with dotted lines. These areas are expanded in Figures 5 through 8.
  • Figures 5 and 7 show numerically calculated equipotential voltage lines in the 5 regions across the SSCS 20' ' shown in Figures 3 and 4.
  • the grid 22" is held at ground.
  • the region 22" is held at -90 volts and the region 26" is held at 40 volts.
  • Other voltage bias schemes may be used.
  • Figures 5 and 7 show that voltage drop across the SSCS 20" is fairly constant. Only very slight deviations occur near the grid lines 22' ' shown in Figures 5 and 7.
  • Figures 5 and 7 show that the voltage is well divided and that carriers will transit from the interaction region 26' ' into the measurement region 30" .
  • Figures 5 and 7 indicate that the potential drop as the carriers move through the measurement region 30' ' can be measured and the induced charge measured will be due to the drift of only one type of carrier (in this example the electrons).
  • Figures 6 and 8 show numerically calculated electric field lines for the SSCS device 20" .
  • the electric field lines go around the grid 22' ' and through the openings in both Figures 6 and 8. This clearly shows that the carriers (in this case electrons) will be directed around the grid lines 23 ' ' and into the measurement region 30' ' . Hence, the carriers will be guided into the measurement region 30' ' from the interaction region 26" .
  • the induced charge they create due to their motion through the potential in the measurement region 30" is recorded.
  • the magnitude of the induced charge measured gives a relative measurement of the initial energy absorbed in the interaction region 26' ' from the original quantum or quanta of radiation.
  • the rectifying grid 22' ' divides the two regions 26" and 30' ' such that a reference plane exists to measure charge movement in the measurement region 30" in addition to directing electron flow.
  • the SSCS Device 20" Can Provide Information of the Interaction Depth
  • the change in pulse height can be monitored as the carriers (both electrons and holes) transit across the interaction region 26".
  • a measurement device can be placed across the interaction region 26' ' and the rate of the signal rise or the rise time of the voltage pulse can be monitored.
  • the pulse rises as the carriers transit across the interaction region 26" .
  • the rate at which the pulse rises will change as soon as one carrier enters into the measurement region 30" or as soon as one carrier type exits the device at the opposite end ( Figure 3, region 32").
  • the rise times and pulse shapes are functions of the location where the radiation quantum interacted and can be used to yield information where the radiation interaction occurred along the length of the interaction region 26" .
  • the SSCS Device 20" Can Provide Information of the Interaction Location in the X-Y Coordinate Plane
  • the pervious region 28" can be designed so that information is yielded as to where in the plane of region 28" the carriers transited.
  • One method of implementation would be to fabricate a set of doped semiconductor lines 23" going in one direction and another set (above or below) going in a direction perpendicular to the first set. The pattern would appear to be a cross hatch of lines going in the X and Y directions. Carriers transiting through the grid 22' ' would go through the X-Y pattern and the location of transit can be determined with a charge-splitting resistor network.
  • the SSCS Device 20" As A Radiation Activated Carrier Emitter
  • the semiconductor single carrier spectrometer 20' ' can be used as a radiation activated electron emitter.
  • the basic device 20" can be altered such that the measurement region is replaced with field emission tips.
  • the voltage bias is held such that the tips are depleted of charge carriers.
  • a bias voltage held external to the tips causes electrons to be emitted from the device through field emission.
  • carriers guided into the measurement region 30" will be extracted external to the SSCS 20" by field emission.
  • the electrons extracted can be measured in a number of different ways. For instance, the electrons might be measured with a bank of dynodes (such as in a photomultiplier tube), a microchannel plate, or a luminous or phosphor screen.
  • the SSCS Device 20" As A Carrier Emission Transistor
  • the bias scheme to the SSCS 20' ' can be changed such that it can be operated as a transistor.
  • the grid region 28" serves as a gate
  • region 32" serves as the carrier source
  • the measurement region 30" serves as the drain. If the measurement region 30" is replaced with field emitter tips, the transistor becomes a carrier emission transistor.
  • the emission rate of the carriers is controlled by adjusting the voltage on the grid 22' ' and by controlling the voltage potential at the field emission tips.
  • the potential on the field emission tips is kept such that tip material is depleted of carriers, hence the charge carriers available in the emission tips are diminished.
  • the SSCS device 20" operated as a field emission transistor allows for its use in vacuum microelectronics.
  • Each of the detectors 20, 20' and 20" can be viewed as a stacked detector or array formed from a single crystal or amorphous material.
  • the stack is a stack of individual detectors fabricated such that they are intimately grown, evaporated, deposited or any other means by which the detectors can be made to be in intimate contact through processing and fabrication.
  • the detectors in the stack can be used together or separately.
  • Section I The Effects of Char e Carrier Trapping
  • a simple planar semiconductor detector is operated by applying a bias voltage across the bulk of the material. Ionizing radiation excites electron-hole pairs that are drifted apart by the device electric field. Electrons are drifted toward the anode and holes are drifted toward the cathode. An induced charge is produced at the terminals of the device by the moving free charge carriers, and the induced charge can be measured by an externally connected circuit.
  • the Shockley-Ramo theorem shows that the induced charge appearing at the terminals of a planar device from moving point charges is proportional to the distance displaced by the moving point charges, regardless of the presence of space charge.
  • the charge in induced charge Q* can be represented by:
  • X represents the interaction location in the detector as measured from the cathode and p is the carrier extraction factor represented by:
  • Equation 2 can be used to calculate the charge induction map that results from position dependent charge collection across a radiation detector. For the following example, it is assumed that all interactions are from photoelectric events. As a result, full gamma ray energy deposition is given to the photoelectron and Compton scattering is not considered. It is also assumed that the detector is uniformly irradiated and significant gamma ray attenuation does not occur.
  • gamma ray events that occur at the cathode cause electron-hole pairs to appear at the cathode.
  • the electrons are swept toward the anode, but the holes are already at the cathode, hence there is virtually no hole motion across the detector.
  • the measured signal is due to electron motion across the device.
  • the effect of charge carrier losses from trapping prevents the measured charge from being 100% of the initially excited charge.
  • the measured induced charge is 92% of the initially excited charge.
  • the holes contribute to the induced charge as they move across the detector toward the cathode.
  • the low hole extraction factor of 0.5 indicates severe trapping, in which case the total measured induced charge is only 43% of the initial excited charge.
  • the signal measured for gamma ray events near the cathode will be over twice the magnitude of signals measured from gamma ray events that occur near the anode.
  • the measured induced charge is position dependent upon where the gamma ray event occurs.
  • Regions that appear horizontally flat on the total induced charge curve indicate a region in which the position-dependent charge collection deviates only slightly. As a result, a "peak" will appear in the multi-channel analyzer (MCA) pulse height spectrum (see Figure 11). Regions that are not horizontally flat will appear as "tail regions" in the MCA pulse height spectrum. From Figure 10, it becomes clear that both electrons and holes must be collected with high efficiency in order to reduce or eliminate position-dependent charge collection effects that cause resolution degradation in planar type semiconductor radiation spectrometers.
  • One method that has been shown to reduce position dependent charge collection problems with gas based ion chamber radiation detectors is the implementation of the Frisch grid.
  • a similar situation to that explained for the semiconductor detector is experienced with gas filled ion chambers, in which electron-ion pairs are produced by gamma ray interactions in the gas.
  • the electron mobilities are much higher than the positive ion mobilities, hence the extraction times of the electrons are considerably less than the extraction times of the ions.
  • the measured pulse amplitude becomes dependent on the initial gamma ray interaction location in the ion chamber.
  • a Frisch grid gas ion chamber is designed to measure the induced charge primarily from electrons, and the general configuration and operation of a Frisch grid ion chamber is shown in Figure 12a.
  • the technique involves separating the radiation detector into two regions by a screen or grid. One region is large and is where the majority of gamma ray interactions occur in the device.
  • a gamma ray interaction occurring in the main volume of the detector excites electron-ion pairs.
  • An externally applied electric field drifts the excited charge carriers in opposite directions, in which the electrons drift through the grid and into the measurement region of the device.
  • the positive ions drift the opposite direction and do not enter into the smaller measurement region of the device.
  • the induced charge produced at the anode results from charge carriers moving between the conductive grid and the anode and not from charge motion between the cathode and the grid.
  • the detector is primarily sensitive to only the electron charge carriers.
  • the measurement region is well shielded such that relatively few gamma ray interactions occur in the measurement region, hence almost all of the measured induced charge is from electrons that have been swept into the measurement region from the interaction region.
  • the semiconductor Frisch grid would be used to compensate for the effects of trapped hole (or electron) charge carriers, not just for slow rise times from poor charge carrier mobilities.
  • the semiconductor Frisch grid detector actually has three regions (not two), those being (a) the interaction region, (b) the pervious region, and (c) the measurement region.
  • the semiconductor Frisch grid detector will still suffer from electron (or hole) charge loss due to trapping in the interaction, pervious and measurement regions.
  • the semiconductor Frisch grid detector will most likely be smaller than the Frisch grid ion chamber, hence shielding the measurement region is difficult. As a result, gamma ray interactions will occur in all three regions, and some of the induced charge in the measurement region will be from hole motion originating in the measurement region.
  • the interaction region is typically the largest volume of the device, and most of the gamma ray interactions occur in the interaction region simply because it composes the largest percentage of the detector volume.
  • the pervious region is the location where the Frisch grid is located. It should be composed of the same material as the detector, or at least a material that allows for electrons to freely move through it into the measurement region. The most straightforward method of producing such a region is to have it be composed of the very same material as the interaction and measurement regions of the device.
  • the grid structure can be implemented by many different ways, including but not limited to outside contacts, outside and buried p or n doped regions, and grooved metallized regions.
  • Section 3. The Planar Parallel Strip Frisch Grid Radiation Detector
  • the device as previously discussed, has three major sections: those being the interaction region, the pervious region and the measurement region.
  • the parallel grid of Figure 13 is in direct contact with the pervious region, and the pervious region is composed of the same material as the interaction region and the measurement region.
  • the "weighting potential” concept is often used to determine the dependence of the observed (or measured) induced charge on the geometry and electrode configuration of a detector.
  • N the number of electrodes in a space are connected to ground the general expression derived by
  • Ej(r) is the electric field at point r due to a unit potential on conductor 1 with all others (2 ... N) grounded (set at 0)
  • v is the carrier velocity
  • V is the potential at conductor 1.
  • the change in the weighting potential over ⁇ x is equivalent to the change in normalized induced charge ( ⁇ Q*/Q 0 ) for charge carriers moving through the same distance ⁇ x.
  • the weighting potential of the device is greatly influenced by the length-to-width ratio, and not just the device width.
  • Figure 15a shows the calculated weighting potential of 5mm x 2mm x 5mm device
  • Figure 15b shows the calculated weighting potential of a 5mm x 3mm x 5mm device
  • Figure 15c shows the weighting potential of a 5mm x 5mm x 5mm device.
  • Figure 16a shows the weighting potential across a 10mm x 2mm x 10mm device
  • Figure 16b shows the calculated weighting potential of a 10mm x 4mm x 10mm device
  • Figure 16c shows the weighting potential of a 10mm x 6mm x 10mm device. Notice that deviations in the weighting potential for the 10mm x 2mm x 10mm device are much less than that of the 5mm x 2mm x 5mm device, which was observed experimentally. Also, the weighting potential of the 5mm x 2mm x 5mm device is identical to the 10mm x 4mm x 10mm device, indicating that under conditions of similar electric field strengths, the two devices should perform similarly (excluding differences in electron charge carrier trapping effects).
  • the larger 10mm x 6mm x 10mm device clearly has a less distorted weighting potential than the smaller 5mm x 5mm x 5mm device.
  • the length to width ratio determines the expected spectroscopic performance of the device (at least in part).
  • Detector 1 Two CdZnTe parallel strip Frisch grid detectors generally of the type shown in Figure 13 were designed and tested. Detector 1 had dimensions of
  • 5mm x 2mm x 5mm (weighting potential of Figure 15a).
  • the parallel strips were 0.5 mm wide and were fabricated onto the 5mm x 5mm faces at a distance of 1mm back from the anode.
  • Detector 2 had dimensions of 10mm x 2mm x 10mm (weighting potential of Figure 16a). The parallel strips were 1mm wide and were fabricated onto the 10mm x 10mm faces at a distance of 2mm back from the anode.
  • Detector 1 (5mm x 2mm x 5mm) was operated with the anode grounded and the cathode held at -400 volts. Operating voltages above 400 volts caused unacceptably high surface leakage currents.
  • the detector was tested with both the parallel Frisch grid “off” (floating) and biased at -80 volts.
  • the device was uniformly irradiated from the side by 662 keV gamma rays from a calibrated 137Cs source.
  • Figure 17 shows room temperature spectra from the device with the grid "off” and "on". With the grid "off", no discernable full energy peak is apparent. However, a gamma ray peak becomes clearly obvious when the parallel grid is "on”.
  • the measured energy resolution of the 662 keV peak is 6.2% at FWHM.
  • Fig. 18 Shown in Fig. 18 are spectra taken with the parallel strip Frisch grid “off” (floating) and turned “on” (-80 volts). No full energy peak is apparent when the parallel grid was “off”, however a full energy peak with 4.5% energy resolution at FWHM becomes obvious when the grid was "on” . It is clear that the device was operating as a true single carrier sensitive semiconductor Frisch grid detector.
  • Detector 2 (10mm x 2mm x 10mm) was operated with the anode grounded, the cathode biased at -400 volts, and the parallel grid either "off” (floating) or "on” (-70 volts). Again, severe surface leakage currents prevented the application of operating voltages above 400 volts across the detector bulk.
  • Figure 19 shows room temperature spectra taken with the 10mm x 2mm x 10mm CdZnTe parallel strip Frisch grid semiconductor spectrometer with the grid “off” and “on” .
  • the grid “off” With the grid “off” , a barely discernable full energy peak is apparent.
  • the 662 keV peak becomes obvious, with a measured energy resolution of 5.91 % at FWHM.
  • the induced charge deviation across the interaction region of the 10mm x 2mm x 10mm device should be less than that of the 5mm x 2mm x 5mm device, which may explain why the resolution is better for the larger detector.
  • the weighting potentials for the devices are different primarily because the width-to- length ratio is smaller for the 10mm x 2mm x 10mm detector from the 5mm x 2mm x 5mm detector.
  • the parallel grids help to flatten the weighting potential in the interaction region more so for the larger detector than the smaller detector. The result indicates that a thicker device (10mm x 4mm x 10mm) would have similar performance as the 5mm x 2mm x 5mm device.
  • Equation 2 describes the expected induced charge as a function of the gamma ray interaction depth. Since the carrier extraction factors are much higher for electrons than holes, the largest induced charge pulse will appear from electrons that are excited near the cathode. The low bias voltage of -400 volts produces a relatively low electric field across the device, hence the electron velocity is relatively low, and the electrons will undergo some appreciable trapping before reaching the anode.
  • the resulting maximum pulse height measured by the external circuit is relatively low.
  • the largest induced charge pulse is generated by electrons excited in the middle of the pervious region. These electrons have undergone no appreciable trapping before entering the measurement region, and the resulting pulse height corresponding to the induced charge will be much higher (as observed).
  • the observed spectra from the CdZnTe detector with the grid "off' and "on” further demonstrates that the detector is operating as a true semiconductor Frisch grid device.
  • the parallel strip Frisch grid detector is doomed to be relatively thin, with the best prototype devices tested to date having a length to width ratio of 5.
  • a stacked series of devices can also be fabricated, yet the case is non-ideal in that the Compton scatter losses can be significant, and numerous devices must be used to fabricate a useful system. Coupling noise will ultimately reduce the number of detectors that can be linked together while still maintaining reasonably high gamma ray energy resolution.
  • a system of stacked (or coupled) planar parallel strip Frisch grid radiation detectors can also be limited in size.
  • a new design can overcome many of these problems, that being the "Geometrically Weighted Semiconductor Frisch Grid Radiation Spectrometer".
  • planar parallel strip semiconductor Frisch grid detectors show promise as room temperature operated gamma ray spectrometers.
  • fundamental problems include:
  • the interaction region to measurement region ratio (W W m ) is small, which results in significant spectrum contamination from gamma ray interactions in the measurement region.
  • the device must remain thin in order to maintain a good length to width ratio, thereby preserving good screening of induced charge from hole transport.
  • the device is too small to efficiently recapture Compton scattered gamma rays. As a result, it is only good for measuring low energy gamma rays. If the device is fabricated wider to increase the gamma ray interaction efficiency, then the weighting potential distribution deviates largely across the device, which causes resolution degradation. Stacking the devices offers a solution to increase the efficiency, but coupling noise between the devices ultimately limits gamma ray energy resolution.
  • the solution is to fabricate a detector with the Frisch grid end very small and the interaction region very large. Hence, induced charge from hole motion remains well screened while increasing the size of a single device.
  • the scheme allows for large detectors to be manufactured with increased gamma ray interaction efficiency while maintaining good Frisch grid screening. Additionally, the Wi/W m ratio will increase as a natural consequence of the design, thereby significantly reducing spectrum contamination from gamma ray interactions in the measurement region.
  • Figures 20a-c shows three possible designs of the geometrically weighted semiconductor Frisch grid radiation detector.
  • the designs incorporate three major effects to improve gamma ray energy resolution, those being (a) the geometric weighting effect, (b) the small pixel effect, and (c) the Frisch grid effect.
  • the overall volume of the device is described by:
  • the volume of the parallelepiped design is 200mm 3
  • the volume of the trapezoid design would be 600 mm 3 .
  • the overall result is a 3.2 times increase in detector gamma ray sensitivity in the interaction region for the trapezoidal design, while retaining a higher rejection ratio for gamma ray interactions occurring in the measurement region.
  • the geometric shape of the trapezoidal detector design allows for a higher ratio of gamma ray interactions to occur in the interaction region over the measurement region than that observed for a parallelepiped parallel strip Frisch grid detector.
  • the general trend for F is to approach unity for both designs until the Frisch grid ultimately contacts the anode, yet the volume of the trapezoid design remains much larger than the parallelepiped design, thereby being more efficient as a gamma ray sensor.
  • the gamma ray probability distribution function is highest near the cathode and lowest near the anode for a trapezoidal shaped semiconductor Frisch grid detector.
  • the normalized total gamma ray probability distribution function for a trapezoidal device is:
  • the "small pixel" effect is the unique weighting potential and induced charge dependence observed with devices having non-symmetrically sized electrodes.
  • the signal formation from a basic planar type of semiconductor detector was discussed earlier, in which it was found that the induced charge has a linear relationship between the distance a carrier travels and the observed increase in induced charge (see equation 1). Such a relationship is not true when the contacts of a device are not the same size.
  • the weighting potential changes much more abruptly near the anode than the region near the cathode.
  • more charge is induced as charge carriers move in the vicinity of the small anode than charge carriers moving in vicinity near the cathode.
  • more charge is induced as charge carriers move in the vicinity of the small anode than charge carriers moving in vicinity near the cathode. From the natural effect of geometrical weighting, more charge carrier pairs are produced near the cathode over that of the anode.
  • Figure 22 shows the weighting potential of the trapezoid prism device described in Part A. As can be seen, the change in induced charge is greatest in the region near the anode. Hence, electron dominated pulses result without even having a Frisch grid on the device.
  • Figures 23 a and 23b show a comparison between the charge induction map of a 10 mm wide planar CdZnTe detector and that of a 10 mm wide (height) CdZnTe trapezoid prism device (dimensions described in Part A). Typically quoted properties of CdZnTe were used for the calculations. As can be seen, the trapezoid prism has a much larger "flat" region than the planar device. Coupling the charge induction response with the interaction probability distribution function in a trapezoid detector yields a much different spectral response than that of a planar device.
  • Figure 24 shows a comparison of a simulated gamma ray spectrum from a planar device. As can be seen, the "tail” is significantly suppressed in the trapezoid device, and a “pseudo-peak" forms as a natural consequence of the coupling between geometrical weighting and the small pixel effect.
  • Pseudo-peak formation from such a device is modeled and shown in Figure 25.
  • the CdZnTe trapezoidal detector was modeled with typically quoted values for CdZnTe. Shown is the expected spectral dependence as a function of increasing bias voltage across the trapezoid device. Compton scattering and gamma ray attenuation were not considered in the calculation.
  • the model predicts the formation of "dual peaking" or "satellite" peaks at low voltages. This can be recognized as a direct consequence of the unusual charge induction dependence shown in Figure 23b, in which there are two flat regions that appear in the charge induction map for the trapezoid detector. As the voltage is increased, the resolution at first gets much better. At higher voltages, the resolution degrades and a full "pseudo-peak" forms near the location expected for the actual gamma ray full energy peak.
  • Figure 26 shows the spectral response from 122 keV 57 Co gamma rays of a geometrically weighted CdZnTe Frisch grid detector (manufactured by eV Products) with the grid "off" (floating).
  • the device has the same dimensions as described in Part A. As can be seen, two peaks appear for one gamma ray energy, a predicted response for low voltages. Yet the applied voltage is 2000 volts, indicating that the voltage is not being dropped across the device uniformly, a general indication of bad electrical contacts. Satellite peaks were observed for all gamma ray energies tested when the grid was "off” . The satellite peaks in all cases disappear when the Frisch grid is turned on.
  • Figure 27 shows full pseudo-peak formation from a trapezoid device, in which 662 keV gamma rays are measured from a 137 Cs source.
  • the device has the same dimensions as described in Part A except that it is 17mm long instead of only 10 mm long.
  • the trapezoidal device exhibited the entire response as predicted from Figure 25, indicating that the contacts on the device are much better than the eV Products manufactured device.
  • Figure 28 shows the weighting potential of a trapezoid prism Frisch grid radiation spectrometer with the Frisch grid turned “on”. Clearly, the outstanding screening for such a device is evident. With the Frisch grid turned “on”, the device maintains highly effective interaction region screening while maintaining a much larger interaction region than the planar parallel Frisch grid design.
  • Figure 29 shows two room temperature (29 °C) spectra of 511 keV 22 Na annihilation gamma rays, one with the Frisch Grid turned “off” and the other with the Frisch grid turned “on”.
  • the CdZnTe geometrically weighted trapezoid prism device is the 17 mm long trapezoid prism detector. As can be seen, the resolution improvement is significant and results from the composite effects of the geometrical weighting, the small pixel effect, and the Frisch grid effect.
  • the device was fabricated from "counter grade" material.
  • the device had dimensions of:
  • Anode width 2 mm
  • the bulk CdZnTe used was "counter grade” material acquired from eV Products. Generally, "counter grade” material is regarded as unfit for fabrication into radiation spectrometers.
  • the device was tested using only conventional, commercially-available, NIM electronics, and was hooked directly to a single standard Ortec 142A preamplifier, as illustrated in Figure 31. No electronic pulse correction, rejection or compensation was used.
  • the spectra came directly from pulses shaped by an Ortec 142A preamplifier and a Canberra 2022 amplifier. The detector was operated at room temperature and no detector cooling in any fashion was performed.
  • the device performed very well at room temperature and did not require cooling in order to observe outstanding results. Additionally, high resolution was achieved without electronic pulse rejection, compensation and correction techniques.
  • the device operated with standard NIM electronics and with only a standard Ortec 142A charge sensitive preamplifier attached directly to the detector anode. There is no need for a custom preamplifier or coupling circuit to achieve very good results, although such a circuit may improve the device performance.
  • room temperature gamma ray energy resolution observed from the device include:
  • the shape of the proposed geometrically weighted semiconductor Frisch grid detectors allows for a compact arrangement of the devices into a pseudo-coaxial design as illustrated in Figure 32. Multiple detectors can be linked together to produce a large volume device. Connecting many devices together and correcting for slight differences in performance has been demonstrated with a very simple resistor network, hence still only one preamplifier is necessary to operate the pseudo-coaxial device array. Overall the device is a compact, room- temperature-operated, high energy resolution, high gamma ray efficiency detector that needs only conventional electronics to operate.
  • the geometric weighting effect can be coupled to the present co-planar grid technology to produce a geometrically weighted co-planar grid radiation detector.
  • the devices operate in the same manner as described for the Frisch grid devices, except that the single polarity sensing is performed with the co-planar grid cubic or planar styled detectors.
  • the co-planar grid radiation detector is the device type that has demonstrated the highest resolution for room temperature operated compound semiconductor radiation detectors. Yet, the device is difficult to fabricate and reproduce, and it often requires more than one output preamplifier to differentiate the signal (thereby producing the single polarity sensing effect).
  • the new device the geometrically weighted semiconductor Frisch grid radiation detector, offers a very easy to fabricate and reproducible method of single polarity sensing. Additionally, the device requires only one output to a conventional commercially available preamplifier.
  • the new device can also be arranged in a pseudo-coaxial or pseudo-spherical design to help facilitate compactness for a detector array.
  • the new invention allows for room-temperature-operated semiconductor radiation detectors to supply high-resolution spectroscopic information about gamma ray sources, charged particles sources, or other sources of ionizing radiation.
  • Such devices can be used for medical probes, remote survey units, remotely located radiation monitors, personal dosimetry meters, or any other application requiring a room temperature operated high resolution semiconductor gamma ray or radiation spectrometer.
  • Figures 32 and 33 there are illustrated two arrays of geometrically weighted Frisch Grid single-carrier radiation spectrometers. Trapezoidal shape allows for good packing of detectors in the array in order to form a large device or system.
  • the rounded bases of the interaction regions allow for more uniformity with carrier transport within the interaction region.
  • Shown in Figure 33 is one possible packing array design that can be used with trapezoidal shaped detectors. As can be seen, the trapezoidal shapes can be arranged to produce a pseudo-spherical device. Shown is one possible configuration in which eight detectors compose the completed spherical array detection device. The rounded bases of the interaction regions demonstrate the possible configuration of a pseudo-spherical detector array design.
  • each base is rounded which creates a situation in which the electrons or holes will have less fluctuation in travel distance in the near base region. Additionally, the electric field will be more umform than in the trapezoidal, pyramidal, or frustum (conical) devices.
  • the device takes advantage of:
  • the shape insures that one carrier is dominant by mere statistics, i.e. more events occur in the base region than the upper grid ring region, therefore more charge carriers are transported from the base to the grid ring than vice versa.
  • the grid takes advantage of the Frisch grid effect, in which only the charge carriers excited in the base region that reach the grid ring can produce a pulse for the external circuit to measure.
  • the region above the grid ring, the measurement region can have a higher electric field than the base region, therefore insuring that the electric field and weighting potential are intensified in the measurement region.
  • the rounded bottom is designed to reduce transport length fluctuations that may appear in the trapezoidal, pyramidal, or frustum (conical) designs.
  • FIG. 34 there is illustrated an etch trench Frisch grid device.
  • the geometry used shows a larger cathode than the anode. However, the case could also be made in which the anode is larger than the cathode. Any shape with one end larger than the other qualifies as a geometrically weighted device.
  • the etch trench device of Figure 34 is an advanced form of the parallel strip Frisch grid device of Figures 1 - 4. Instead of stacking the devices in intimate contact, the individual Frisch components are fabricated by a series of etched trenches into the substrate. The contacts behaving as the Frisch grid are applied to the bottom of the device etched trenches or wells, and the collecting contacts are applied to the top of the device "towers".
  • the shape insures that one carrier is dominant by mere statistics, i.e., more events occur near the base region than the measurement region, therefore more charge carriers are transported from the base to the Frisch grid than vice versa.
  • the grid takes advantage of the Frisch grid effect, in which only the charge carriers excited in the base region that pass through the Frisch grid will produce a pulse for the external circuit to measure.
  • the measurement region can have a higher operating electric field than the base region, therefore insuring that the electric field and weighting potential are intensified in the measurement region.
  • the etched trench pattern with the electrodes "toggled” in and out of the trenches produces a real Frisch grid with a tighter grid formation than the parallel strip or concentric Frisch grid.
  • the new design helps to reduce the overall leakage current between the grids, thereby allowing for higher differential voltages to be applied between the grids. Additionally, the Frisch grid screening effect is improved over the single module parallel or concentric designs.
  • Figure 35 discloses a version of a Frisch grid detector, generally indicated at 50, formed through outside contacts.
  • the detector 50 includes concentric and segmented Frisch grid strips 52, a pervious region 54, a measurement region 55, and an interaction region 56.
  • the device 50 also includes a contact 57 to the measurement region 55 and a contact or electrode 58 to the interaction region 56.
  • FIG 36 there is illustrated another version of a Frisch grid detector 60 formed through outside contacts.
  • the detector 60 has a very wide or large pervious region.
  • the detector 60 includes grid contacts 62 to the wide pervious region.
  • the detector 60 also includes an interaction region 64, a measurement region 65, and a contact 66 to the measurement region 65.
  • Device 60 also includes a contact 67 to the interaction region 64.
  • FIGs 37a-37c there are illustrated sample profiles of trench-type semiconductor Frisch grid radiation detectors, generally indicated at 70, 80 and 90, respectively.
  • the detectors 70, 80 and 90 include top contacts 72, 82 and 92, respectively.
  • the devices 70, 80 and 90 also include measurement regions 74, 84 and 94, respectively. Furthermore, the devices 70,
  • the 80 and 90 include Frisch grids 75, 85 and 95, respectively.
  • the Frisch grids 75, 85 and 95 define pervious regions 76, 86 and 96, respectively.
  • the devices 70, 80 and 90 further include interaction regions 77, 87 and 97, respectively. Still further, the devices 70, 80 and 90 include bottom contacts 78, 88 and 98, respectively.
  • a detector 100 which is of a segmented, depression type or trench-type semiconductor Frisch grid detector wherein the trenches run in different directions, thereby isolating measurement region towers 102.
  • a top contact On top of each of the towers 102 is a top contact
  • the device 100 also includes a Frisch grid contact 106, a pervious region under the Frisch grid contact 106, an interaction region 107, and a bottom contact 108.
  • the device 110 includes a pair of external, non-contacting Frisch grid strips 112 which define a pervious region 113.
  • the device 110 also includes a measurement region 114 and an interaction region 115.
  • the device 110 includes a contact 116 to the measurement region 114 and a contact 117 to the interaction region 115.
  • the device 120 is similar to the device 110 of Figure 39, except that the device 120 includes a layer of insulating material 122 on a pervious region 123.
  • the device 120 further includes a measurement region 124 and an interaction region 125.
  • the device 120 also includes external, non-contacting Frisch grid strips 126 which are spaced from the pervious region 123 by the insulating material 122.
  • the device 120 further includes a contact 127 to the measurement region 124 and a contact 128 to the interaction region 125.
  • the device 130 includes a pervious region 131 defined by an external and non-contacting Frisch grid ring 132.
  • the pervious region 131 separates a measurement region 133 from an interaction region 134.
  • the device 130 also includes a contact 135 to the measurement region 133 and a contact 136 to the interaction region 134.
  • the device 140 also includes an external, non-contacting Frisch grid ring 141 which defines a pervious region 142. However, the device 140 further includes an insulating filling 143. The device 140 also includes a measurement region 144 and an interaction region 145.
  • the device 140 also includes a contact 146 to the measurement region 144 and a contact 147 to the interaction region 145.
  • the Frisch grid detector 150 which may be hemispherical or cylindrical in shape.
  • the Frisch grid detector 150 may be implemented with a buried structure, entrenched or depressed structure or a membrane structure.
  • the device 150 is a wide angle (180°) semiconductor Frisch grid detector with an etched style Frisch grid 152 and small measurement region contacts 154.
  • the detector 150 includes a measurement region 155, a pervious region 156, and an interaction region 157.
  • the device 150 also includes a contact 158 to the interaction region 157.
  • a device 160 which is a semiconductor Frisch grid radiation detector with a buried membrane-type Frisch grid region 162.
  • the region or grid 162 is defined by a membrane which is very thin and acts like a Frisch grid as well as a pervious region.
  • the grid 162 may be made from a very thin metal or a very thin semiconductor.
  • the grid may form a rectifying contact to the main material of the detector's interaction and/or measurement regions as illustrated in Figure 44. Charge carriers are accelerated to the membrane 162 and punch through or tunnel through the membrane 162 and into the measurement region.
  • the device 160 also includes a contact 164 to the interaction region and a contact 166 to the measurement region. While the best mode for carrying out the invention has been described in detail, those familiar with the art to which this invention relates will recognize various alternative designs and embodiments for practicing the invention as defined by the following claims.

Landscapes

  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Molecular Biology (AREA)
  • Health & Medical Sciences (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • Electromagnetism (AREA)
  • Engineering & Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Power Engineering (AREA)
  • Measurement Of Radiation (AREA)

Abstract

L'invention concerne un détecteur à ionisation haute résolution et un réseau de tels détecteurs utilisant un modèle de référence se rapportant à une matière conductrice ou semi-conductrice pour définir des régions d'interaction et de mesure et une région perméable dans un substrat d'ionisation, fait notamment en un matériau de type CdZnTe. Le détecteur à ionisation est un détecteur de rayonnement semi-conducteur à température ambiante. Diverses géométries d'un tel détecteur et un réseau de tels détecteurs permettent de réaliser des spectromètres gamma à résolution relativement haute et fonctionnant à température ambiante. Par exemple, un détecteur de 1 cm3 est capable de mesurer un rayonnement gamma du Cs137 à 662 keV d'une résolution d'énergie à température ambiante approchant 2 % à LMII. Deux principaux types de tels détecteurs comprennent une grille de détection Frisch bandes parallèles semi-conducteur et la grille de détection Frisch prisme trapézoïdal semi-conducteur à pondération géométrique. Le détecteur à pondération géométrique est sensible à des résolutions d'énergie à température ambiante (24 °C) de 2,68 % de la LMII dans le cas des rayons gamma du Cs137 à 662 keV et de 2,45 % de la LMII dans le cas des rayons gamma du Co60 à 1,332 MeV. Les détecteurs fonctionnent bien sans techniques de correction, de compensation ou de réjection d'impulsions électroniques. Ces dispositifs qui fonctionnent à température ambiante avec une simple électronique de type NIM bin du commerce ne nécessitent pas, pour de bons résultats spectroscopiques, d'étages de préamplification ou de refroidissement spéciaux.
PCT/US1998/014055 1997-07-10 1998-07-07 Detecteur a ionisation haute resolution et reseau de tels detecteurs WO1999003155A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU82928/98A AU8292898A (en) 1997-07-10 1998-07-07 High-resolution ionization detector and array of such detectors

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US5209297P 1997-07-10 1997-07-10
US60/052,092 1997-07-10
US09/075,351 US6175120B1 (en) 1998-05-08 1998-05-08 High-resolution ionization detector and array of such detectors
US09/075,351 1998-05-08

Publications (1)

Publication Number Publication Date
WO1999003155A1 true WO1999003155A1 (fr) 1999-01-21

Family

ID=26730166

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1998/014055 WO1999003155A1 (fr) 1997-07-10 1998-07-07 Detecteur a ionisation haute resolution et reseau de tels detecteurs

Country Status (2)

Country Link
AU (1) AU8292898A (fr)
WO (1) WO1999003155A1 (fr)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1351309A2 (fr) * 2002-04-02 2003-10-08 Hitachi, Ltd. Détecteur de rayonnement et appareil de radiation
RU2640320C1 (ru) * 2017-02-09 2017-12-27 Акционерное общество "Государственный научный центр Российской Федерации Троицкий институт инновационных и термоядерных исследований" (АО "ГНЦ РФ ТРИНИТИ") Способ регистрации импульсного ионизирующего излучения
CN108709905A (zh) * 2018-06-25 2018-10-26 复旦大学 基于非均匀光照的光增强型气敏元件
CN111211197A (zh) * 2013-12-27 2020-05-29 同方威视技术股份有限公司 高纯锗探测器

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5391882A (en) * 1993-06-11 1995-02-21 Santa Barbara Research Center Semiconductor gamma ray detector including compositionally graded, leakage current blocking potential barrier layers and method of fabricating the detector
US5530249A (en) * 1994-06-09 1996-06-25 The Regents Of The University Of California Electrode configuration and signal subtraction technique for single polarity charge carrier sensing in ionization detectors
EP0777277A1 (fr) * 1995-09-07 1997-06-04 Santa Barbara Research Center Détecteur de radiations ionisantes
EP0851512A1 (fr) * 1996-12-24 1998-07-01 Commissariat A L'energie Atomique Dispositif de détection de rayonnements ionisants a semi-conducteur de haute résistivité

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5391882A (en) * 1993-06-11 1995-02-21 Santa Barbara Research Center Semiconductor gamma ray detector including compositionally graded, leakage current blocking potential barrier layers and method of fabricating the detector
US5530249A (en) * 1994-06-09 1996-06-25 The Regents Of The University Of California Electrode configuration and signal subtraction technique for single polarity charge carrier sensing in ionization detectors
EP0777277A1 (fr) * 1995-09-07 1997-06-04 Santa Barbara Research Center Détecteur de radiations ionisantes
EP0851512A1 (fr) * 1996-12-24 1998-07-01 Commissariat A L'energie Atomique Dispositif de détection de rayonnements ionisants a semi-conducteur de haute résistivité

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
MCGREGOR D S ET AL: "CdZnTe semiconductor parallel strip Frisch grid radiation detectors", NUCLEAR SCIENCE SYMPOSIUM AND MEDICAL IMAGING CONFERENCE (NSS/MIC), ALBUQUERQUE, NM, USA, 9-15 NOV. 1997, vol. 45, no. 3, pt.1, ISSN 0018-9499, IEEE Transactions on Nuclear Science, June 1998, IEEE, USA, pages 443 - 449, XP002080053 *

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1351309A2 (fr) * 2002-04-02 2003-10-08 Hitachi, Ltd. Détecteur de rayonnement et appareil de radiation
EP1351309A3 (fr) * 2002-04-02 2006-06-07 Hitachi, Ltd. Détecteur de rayonnement et appareil de radiation
US7141797B2 (en) 2002-04-02 2006-11-28 Hitachi, Ltd. Radiation detector and radiation apparatus
EP1788632A2 (fr) * 2002-04-02 2007-05-23 Hitachi, Ltd. Détecteur de rayonnement et appareil de radiation
EP1788632A3 (fr) * 2002-04-02 2007-05-30 Hitachi, Ltd. Détecteur de rayonnement et appareil de radiation
US7315025B2 (en) 2002-04-02 2008-01-01 Hitachi, Ltd. Radiation detector and radiation apparatus
CN111211197A (zh) * 2013-12-27 2020-05-29 同方威视技术股份有限公司 高纯锗探测器
RU2640320C1 (ru) * 2017-02-09 2017-12-27 Акционерное общество "Государственный научный центр Российской Федерации Троицкий институт инновационных и термоядерных исследований" (АО "ГНЦ РФ ТРИНИТИ") Способ регистрации импульсного ионизирующего излучения
CN108709905A (zh) * 2018-06-25 2018-10-26 复旦大学 基于非均匀光照的光增强型气敏元件
CN108709905B (zh) * 2018-06-25 2020-10-30 复旦大学 基于非均匀光照的光增强型气敏元件

Also Published As

Publication number Publication date
AU8292898A (en) 1999-02-08

Similar Documents

Publication Publication Date Title
US6175120B1 (en) High-resolution ionization detector and array of such detectors
US6333504B1 (en) Semiconductor radiation detector with enhanced charge collection
US5677539A (en) Semiconductor radiation detector with enhanced charge collection
US6781132B2 (en) Collimated radiation detector assembly, array of collimated radiation detectors and collimated radiation detector module
KR100264312B1 (ko) 개선된 전하 수집능력을 갖는 반도체 방사선 검출기
US8461541B2 (en) Radiation detector, method of manufacturing a radiation detector and use of the detector for measuring radiation
Takahashi et al. Recent progress in CdTe and CdZnTe detectors
US6069360A (en) Method and apparatus for electron-only radiation detectors from semiconductor materials
US5117114A (en) High resolution amorphous silicon radiation detectors
US9121953B2 (en) Array of virtual Frisch-grid detectors with common cathode and reduced length of shielding electrodes
US9134439B2 (en) Array of virtual Frisch-grid detectors with common cathode and reduced length of shielding electrodes
Akimov Silicon radiation detectors
Watanabe et al. CdTe stacked detectors for Gamma-ray detection
US7060523B2 (en) Lithium-drifted silicon detector with segmented contacts
McGregor et al. Performance of CdZnTe geometrically weighted semiconductor Frisch grid radiation detectors
WO1999003155A1 (fr) Detecteur a ionisation haute resolution et reseau de tels detecteurs
He et al. Effects of charge sharing in 3-D position sensitive CdZnTe gamma-ray spectrometers
Cirignano et al. Evaluation of CZT detectors with capacitive Frisch grid structure
Castoldi et al. A new position sensing X-ray detector: Working principle and experimental results
Roos Semiconductor Detectors
Goldan et al. Unipolar charge sensing using Frisch grid technique for amorphous selenium radiation detectors
Niraula et al. Study of multi-electrodes structure in CdTe nuclear radiation detectors
Bolotnikov et al. Array of virtual Frisch-grid detectors with common cathode and reduced length of shielding electrodes
US7884332B1 (en) Radiation detector
Donmez Characterization of single-sided cadmium zinc telluride strip detectors for high energy astrophysics applications

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AL AM AT AU AZ BA BB BG BR BY CA CH CN CU CZ DE DK EE ES FI GB GE GH GM GW HR HU ID IL IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MD MG MK MN MW MX NO NZ PL PT RO RU SD SE SG SI SK SL TJ TM TR TT UA UG UZ VN YU ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): GH GM KE LS MW SD SZ UG ZW AM AZ BY KG KZ MD RU TJ TM AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE BF BJ CF CG CI CM GA GN ML MR NE SN TD TG

DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application
NENP Non-entry into the national phase

Ref country code: KR

REG Reference to national code

Ref country code: DE

Ref legal event code: 8642

122 Ep: pct application non-entry in european phase
NENP Non-entry into the national phase

Ref country code: CA