A spectrum correction method for semiconductor gamma detectors

Abstract

The present invention discloses a method for correcting the energy spectrum of a semiconductor gamma detector, comprising the following steps: using a fast-front, low-noise charge-sensitive amplifier to perform analog integration on the current pulse signal output by the detector, then digitizing the waveform after the integration and amplification, and extracting the rise time and amplitude information of the waveform on this basis; eliminating cases with too short rise times, and the remaining cases are valid cases; for the valid cases, they are divided into two categories according to the rise time characteristics of the waveform; using a radioactive source to calibrate the detector, fitting the relationship between the waveform amplitude and rise time of the above two categories of valid cases obtained by the test, and compensating the waveform amplitude value accordingly, and finally obtaining the corrected energy spectrum. The method proposed by the present invention can significantly improve the low-energy tailing effect of the full-energy peak of a hemispherical or quasi-hemispherical compound semiconductor gamma detector, thereby improving energy resolution, and has broad application prospects in the fields of nuclear industry, nuclear medicine, environmental radiation detection, etc.

Description

Technical Field

[0001] The present invention belongs to the technical field of radiation detection, and in particular relates to an energy spectrum correction method applied to a semiconductor gamma detector. Background Art

[0002] Semiconductor detectors, with their advantages of high energy resolution, fast temporal response, and simple structure, have been widely used in a variety of fields, including the nuclear industry, nuclear medicine, and nuclear radiation detection. Compound semiconductor materials (such as cadmium zinc telluride, cadmium telluride, gallium arsenide, and mercury iodide) have wide band gaps and low leakage currents, allowing detectors made from them to operate at room temperature. They are widely used for X-ray and gamma-ray energy spectrum measurements. However, common compound semiconductor materials have low hole mobility, making holes easily trapped. This results in significant signal amplitude (charge) loss, limiting the detector's energy resolution. An effective approach currently is to create a hemispherical or quasi-hemispherical electric field within the detector. This approach creates a high electric field intensity near the anode and a rapid potential drop away from the anode, resulting in a low and relatively stable potential across most of the detector. Therefore, for most incident particles, the current pulse generated by a hemispherical or quasi-hemispherical semiconductor gamma detector is primarily contributed by electron drift, with the contribution from holes being negligible, significantly improving the detector's energy resolution.

[0003] Despite this, for a considerable number of events, the signal charge cannot be fully collected because the impact location is too close to the anode, the electron drift path is too short, and the weighted potential difference between the end and starting points of the drift path is not large. At the same time, for events where the impact location is close to the cathode, some electrons are captured due to the excessively long electron drift distance, resulting in incomplete charge collection. Both of these factors lead to charge loss, causing the full-energy peak of the gamma energy spectrum to no longer conform to the normal distribution. The left half of the full-energy peak is broadened, and the energy resolution deteriorates. This is the low-energy tailing effect of the gamma full-energy peak. The low-energy tailing effect not only leads to a decrease in energy resolution, but in scenarios with complex source terms, it can easily cause the lower-energy and lower-activity gamma full-energy peak to be annihilated by the higher-energy full-energy peak, seriously affecting the effectiveness of nuclide identification. If further corrections can be made, its performance can be greatly improved.

[0004] Currently, several research results on energy spectrum correction for compound semiconductor detectors have been published both domestically and internationally. Typical examples include "Research on Depth Sensitivity and Energy Correction Techniques Based on Pixel-Type CZT Detectors" by Wu Jun et al. from Chengdu University of Technology and "Development of CZT Surface-Based Pixel Detectors and Implementation of Amplitude Correction Techniques" by Li Miao et al. from Chongqing University. However, these studies all focus on pixel-type semiconductor detectors. Research on gamma spectrum correction for quasi-hemispherical semiconductor detectors is primarily focused on the paper "Performance of a New CdZnTe Portable Spectrometric System for High Energy Applications" (IEEE Transactions on Nuclear Science, Vol. 52, No. 5, October 2005) by L. Verger et al., but this work only eliminates some instances of short rise times and does not compensate for amplitude.

[0005] In order to further improve the gamma energy spectrum resolution of the quasi-hemispherical semiconductor detector, the present invention proposes to combine the structural characteristics and electric field characteristics of the quasi-hemispherical semiconductor detector itself, classify the gamma events according to the waveform rise time, and on this basis, compensate and correct the signal amplitude (i.e., charge) measurement value, thereby improving the low-energy tailing effect of the full-energy peak of the gamma energy spectrum. Since the low-energy tailing effect is mainly manifested at the bottom left of the full-energy peak, one-tenth the height width of the full-energy peak (FWTM) can be introduced to characterize the low-energy tailing effect. This method can improve the energy resolution of the detector, and in particular can significantly reduce the FWTM resolution of the full-energy peak, so that the full-energy peak corresponding to the lower activity nuclide close to the left side of the full-energy peak is more prominent, thereby improving the identification ability of gamma nuclides. Summary of the Invention

[0006] The purpose of the present invention is to propose a gamma spectrum correction method to address the problem in the prior art that the mobility of electrons and holes in the electric field of quasi-hemispherical compound semiconductor gamma detectors is quite different, resulting in signal charge loss and thus low-energy tailing. The method can compensate for the charge collection loss caused by the short electron drift path and electron capture, thereby improving the energy resolution of the full-energy peak.

[0007] To achieve the above object, the technical solution provided by the present invention is as follows:

[0008] A method for correcting an energy spectrum of a semiconductor gamma-ray detector comprises the following steps:

[0009] Step 1: Use a fast-front, low-noise charge-sensitive amplifier to perform analog integration on the current pulse signal output by the detector, then sample and digitize the waveform after integration and amplification, and extract the rise time and amplitude information of the waveform on this basis;

[0010] Step 2: Eliminate cases with too short rise times, and the remaining cases are valid cases;

[0011] Step 3: For valid cases, classify them into two categories based on the rise time characteristics of the waveform: the first category is cases where the electron drift path is too short, and the second category is cases where the electron drift path is too long;

[0012] Step 4: Use a radioactive source to calibrate the detector and fit the relationship between the waveform amplitude and rise time of the two types of events obtained from the test. Through fitting, the amplitude compensation formulas for the two types of events are obtained. The amplitude values ​​of the two types of events are compensated accordingly. The amplitude values ​​of all compensated events are then statistically analyzed to obtain the corrected energy spectrum.

[0013] Furthermore, in step 2, the case where the rise time is too short is eliminated, that is, a parameter t0 is set. If the waveform rise time t r If it is less than t0, the case is discarded.

[0014] Furthermore, in step 4, the amplitude value of the first type of event is compensated, that is, for a given time t1, if the waveform rise time t r If it is less than t1, it compensates for the charge collection loss caused by the electron drift path being too short and the weighted potential difference between the end point and the starting point of the drift path being too small, where t1>t0.

[0015] Furthermore, in step 4, the amplitude value of the second type of event is compensated, that is, for a given time t1, if the waveform rise time t r If it is greater than t1, the charge collection loss caused by electron capture is compensated, where t1>t0.

[0016] Furthermore, the parameters of the amplitude compensation formula for compensating the amplitude value of the first type of event or the amplitude value of the second type of event are obtained by measuring a standard radioactive source of known energy and fitting the rise time and amplitude of its waveform; the amplitude compensation formula can be adopted in various ways, including analytical functions, or constructed as a lookup table in the form of numerical values.

[0017] The principle of the present invention is to first collect the original waveform. The original waveform is generated by the following process: after gamma rays enter the semiconductor detector, the generated electrons drift in the quasi-hemispherical electric field. The generated current pulse passes through the charge-sensitive amplifier, and the output is an integrated and amplified voltage waveform. Its amplitude A is proportional to the collected charge Q, and its rise time is determined by the inherent rise time of the amplifier and the time the electrons drift in the detector electric field.

[0018] According to Ramo's theorem, the collected charge Q satisfies:

[0019] Q=qU w

[0020] Among them, U w It is the weighted potential difference between the end point and the starting point of electron drift. In a quasi-hemispherical detector, it is proportional to the potential difference between the two points inside the detector.

[0021] For gamma rays striking very close to the anode, the resulting electron drift path is short, and the weighted potential difference across the drift path is small, resulting in incomplete charge collection. This short drift path also shortens the drift time, resulting in a signal with a short rise time. Different rise times correspond to different degrees of charge loss. The signal amplitude of these events can be corrected based on their rise time, or they can be discarded if the rise time is too short.

[0022] For gamma rays striking a great distance from the anode, the long drift distance of electrons causes some electrons to be captured during the drift process, resulting in incomplete charge collection. This signal, characterized by a long drift time due to the long electron drift time, is characterized by a long rise time, with different rise times corresponding to varying degrees of charge loss. For this signal, amplitude compensation is also applied based on the rise time.

[0023] The technical solution proposed by the present invention can be seen as further enhancing the advantages of the quasi-hemispherical compound semiconductor gamma detector, enabling it to measure gamma energy spectra with higher energy resolution. In complex radioactive environments, especially where multiple nuclides are present with significantly varying activities, this solution can reduce the low-energy tailing effect of the gamma nuclide full-energy peak, significantly improving the spectrometer's ability to discriminate between radionuclides, thereby better meeting the needs of accurately measuring the gamma-ray energy spectrum and identifying nuclides. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flowchart of the energy spectrum measurement correction of the present invention;

[0025] Figure 2 A waveform diagram provided by an embodiment of the present invention;

[0026] Figure 3 A diagram of a detector structure model provided by an embodiment of the present invention;

[0027] Figure 4 A rise time-amplitude relationship diagram before and after correction provided by an embodiment of the present invention;

[0028] Figure 5 A comparison diagram of the energy spectra before and after correction provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0029] To make the technical solution of the present invention more clear, a specific implementation example is clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiment is only a part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0030] The energy spectrum correction method for semiconductor gamma detectors described in the present invention implements the energy spectrum measurement correction process as follows: Figure 1 The steps are as follows:

[0031] Step 1: Use a fast-front, low-noise charge-sensitive amplifier to integrate and amplify the current pulse signal output by the detector, then digitize the waveform output by the amplifier, and then extract the rise time and amplitude information of the waveform. In this embodiment, the time interval between the waveform reaching a preset low threshold and high threshold is used as the rise time. Other methods can also be used to calculate the waveform rise time. A typical waveform is as follows: Figure 2 shown.

[0032] Step 2: Eliminate cases with too short rise time. The specific method is: given a parameter t0, if the waveform rise time t r If it is less than t0, the case is discarded.

[0033] Step 3: For the waveforms retained in step 2, they are divided into two categories according to their rise time. The classification method is as follows: For a given parameter t1 (t1>t0), if the waveform rise time t r If t is less than t1, it is a case of the first category; otherwise, it is a case of the second category.

[0034] Step 4: Use a radioactive source to calibrate the detector, and fit the relationship between the waveform amplitude and rise time of the two types of events obtained from the test. Through fitting, the amplitude compensation formulas for the two types of events are obtained. The amplitude values ​​of the two types of events are compensated accordingly, and then the amplitude values ​​of all compensated events are statistically analyzed to obtain the corrected energy spectrum. Specifically, it includes:

[0035] 1) Compensate the amplitude value of the first type of case, that is, for a given time t1 (t1>t0), if the waveform rise time t r Less than t1, compensating for the charge loss caused by the short electron drift path. The amplitude compensation formula is:

[0036] A′=A+A*f1(t r ) (1)

[0037] Where A′ is the amplitude after compensation, A is the original amplitude of the waveform, and f1 is the rise time t r This formula is derived by calibrating the gamma detector to be corrected using a radioactive source of known energy and then fitting the relationship between the waveform rise time and amplitude for each event. The amplitude compensation formula can be constructed using an analytical function or a lookup table in numerical form.

[0038] 2) For the second type of case, compensate for the charge loss caused by electron capture. The amplitude compensation formula is:

[0039] A′=A+A*f2(t r ) (2)

[0040] Where A′ is the amplitude after compensation, A is the original amplitude of the waveform, and f2 is the rise time t r This formula is derived by calibrating the gamma detector to be corrected using a radioactive source of known energy and then fitting the relationship between the waveform rise time and amplitude for each event. The amplitude compensation formula can be constructed using an analytical function or a lookup table in numerical form.

[0041] 3) Calculate the energy spectrum using the amplitude after discarding and compensating.

[0042] In order to further demonstrate the implementation and effect of the correction method, a quasi-hemispherical semiconductor detector with typical structural dimensions was set up using physical simulation software to generate simulation cases and corresponding signal waveforms, which were then processed using the technical solution proposed in this invention:

[0043] This implementation uses the commonly used Garfield++ software in the field of particle detection for Monte Carlo simulation. The detector material is set in the software to cadmium zinc telluride (CZT), with dimensions of 10mm*10mm*5mm. The detector structure is quasi-hemispherical, with the anode located in the center of the top (square) and measuring 1mm*1mm. The other five sides are cathodes. The structural model is as follows: Figure 3 Its electron mobility is set to a typical 1100 cm 2 / (V·s), the hole mobility is set to a typical 50 cm 2 / (V·s). A gamma point source with an energy of 511 keV is set 10 mm from the cathode at the bottom of the detector. Simulation reveals the current pulse signal generated by each gamma ray interacting with the detector. Simultaneously, a typical fast-lead, low-noise charge-sensitive amplifier is simulated using PSpice circuit simulation software to obtain its impulse response function. In this embodiment, the amplifier's intrinsic rise time is approximately 16 ns.

[0044] The following describes the specific process of gamma spectrum correction using simulation data:

[0045] First, the current pulse signal of the detector is convolved with the impulse response function of the amplifier to obtain the output waveform of the amplifier, and the waveform is digitized. The sampling point interval is set to 2ns, which is equivalent to a waveform digitization sampling rate of 500MSPS. Then, the waveform rise time and amplitude are calculated and recorded, and a two-dimensional scatter plot is made, as shown in the figure. Figure 4 As shown in the center left figure, the upper and lower thresholds are set to 90% and 10% of the waveform amplitude, respectively. The parameters t0 and t1 are set to 30ns and 50ns, respectively. This means that events with rise times less than 30ns are discarded, and the remaining events are divided into two categories: those with rise times between 30 and 50ns are classified as category one, and the rest are classified as category two.

[0046] from Figure 4 As can be seen from the middle left figure, for the first type of case, the amplitude decreases as the rise time decreases. This is the charge collection loss caused by the small weighted potential difference at both ends of the electron drift path; for the second type of case, the amplitude decreases as the rise time increases. This is the charge collection loss caused by electron capture. The first and second types of cases can be compensated for amplitude using the formulas (1) and (2) mentioned above, respectively. The amplitude compensation formulas (1) and (2) are obtained by using a gamma radiation source of known energy for calibration testing (in this embodiment, a single-energy gamma source is set in the simulation software to obtain simulation data) and fitting the relationship between the rise time and amplitude of its waveform.

[0047] The rise time-amplitude relationship after correction according to the above steps is as follows Figure 4 As shown in the middle right figure, we can see that both charge losses are effectively corrected. The energy spectrum before and after correction is compared. Figure 5 As shown in the figure, the full-energy peak half-maximum width (FWHM) decreased from 1.78keV to 1.68keV, the one-tenth width (FWTM) decreased from 4.65keV to 3.21keV, and the FWTM / FWHM ratio decreased from 2.61 to 1.91, approaching the theoretical value of 1.82 for Gaussian distribution. This indicates that the energy resolution has been significantly improved after the correction.

[0048] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications made within the spirit and principles of the present invention, including replacement of the semiconductor detector type (for example, replacing cadmium zinc telluride with a semiconductor detector based on other materials, or the performance parameters of the detector material differ from the simulation parameters set in the embodiment), or changes to the specific size or geometric structure of the detector, as well as equivalent replacements, improvements, etc., should be included in the scope of protection of the present invention.

Claims

1. A method for correcting energy spectrum of a semiconductor gamma detector, characterized in that: The method comprises the following steps: Step 1: Use a fast-front, low-noise charge-sensitive amplifier to perform analog integration on the current pulse signal output by the detector, then sample and digitize the waveform after integration and amplification, and extract the rise time and amplitude information of the waveform on this basis; Step 2: Eliminate cases with too short rise times, and the remaining cases are valid cases; Step 3: For valid cases, classify them into two categories based on the rise time characteristics of the waveform: the first category is cases where the electron drift path is too short, and the second category is cases where the electron drift path is too long; Step 4: Use a radioactive source to calibrate the detector and fit the relationship between the waveform amplitude and rise time of the two types of events obtained from the test. Through fitting, the amplitude compensation formulas for the two types of events are obtained. The amplitude values ​​of the two types of events are compensated accordingly. The amplitude values ​​of all compensated events are then statistically analyzed to obtain the corrected energy spectrum.

2. The energy spectrum correction method for a semiconductor gamma detector according to claim 1, wherein: In step 2, the case where the rise time is too short is eliminated, that is, a parameter t0 is set. If the waveform rise time t r If it is less than t0, the case is discarded.

3. The energy spectrum correction method for a semiconductor gamma detector according to claim 2, wherein: In step 4, the amplitude value of the first type of event is compensated, that is, for a given time t1, if the waveform rise time t r If it is less than t1, it compensates for the charge collection loss caused by the electron drift path being too short and the weighted potential difference between the end point and the starting point of the drift path being too small, where t1 > t0.

4. The energy spectrum correction method for a semiconductor gamma detector according to claim 2, wherein: In step 4, the amplitude value of the second type of event is compensated, that is, for a given time t1, if the waveform rise time t r If the time is greater than t1, the charge collection loss due to electron capture is compensated, where t1 > t0.

5. The energy spectrum correction method for a semiconductor gamma detector according to claim 3 or 4, characterized in that: The parameters of the amplitude compensation formula for compensating the amplitude value of the first type of event or the amplitude value of the second type of event are obtained by measuring a standard radioactive source of known energy and fitting the rise time and amplitude of its waveform; the amplitude compensation formula is constructed into a lookup table in the form of a function analytical expression or a numerical value.

Images