Estimation method of neutron energy spectrum
The method estimates neutron energy spectra using semiconductor memory devices' soft error rates, overcoming facility and safety challenges of traditional methods, allowing for safe and efficient neutron energy spectrum determination.
Patent Information
- Application Number
- JP2023102032
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-06-21
- Publication Date
- 2025-01-09
AI Technical Summary
Existing methods for measuring neutron energy spectra, such as the time-of-flight and activation methods, face limitations in availability of facilities, require high-resolution detectors, and pose risks due to handling radioactive materials.
A method utilizing the soft error occurrence rate in semiconductor memory devices with known neutron energy-dependent SEU cross sections to estimate neutron energy spectra, involving measurement and comparison of soft error rates to calculate the neutron energy spectrum.
Enables easy and safe estimation of neutron energy spectra without the need for specialized facilities or high-resolution detectors, reducing radiation exposure risks.
Smart Images
Figure 2025002104000001_ABST
Abstract
Description
[Technical field]
[0001] The present disclosure relates to a method for estimating a neutron energy spectrum. [Background technology]
[0002] One method for measuring a neutron energy spectrum is the time-of-flight method (Non-Patent Document 1). In the time-of-flight method, a neutron energy spectrum is measured based on the flight time from when a neutron beam is generated until it reaches a detector.
[0003] There is also an activation method that utilizes a threshold reaction (Non-Patent Document 2). In this activation method, a metal foil is activated by being irradiated with a neutron beam, and the radiation emitted when the atomic nuclei of the activated metal foil decay is measured to measure the energy spectrum of the neutron beam irradiated onto the metal foil. [Prior art documents] [Non-patent literature]
[0004] [Non-Patent Document 1] A. Masuda, et al., “Time-of-Flight Measurements for Low-Energy Components of 45-MeV Quasi-Monoenergetic High-Energy Neutron Field from 7Li(p,n) Reaction”, IEEE Transactions on Nuclear Science, vol.62, no.3, June 2015, p.1295-p.1300 [Non-Patent Document 2] V. Suman et al., “Measurement of Neutron Energy Distributions From p+Be Reaction at 20 MeV Using Threshold Activation Foils”, IEEE Transactions on Nuclear Science, vol.63, no.4, August 2016, p.2283-p.2292 Summary of the Invention [Problem to be solved by the invention]
[0005] However, when measuring the neutron energy spectrum using the time-of-flight method, only pulsed neutron sources are available, which are limited worldwide, and an instrument capable of detecting neutrons with high time resolution is required.
[0006] When measuring neutron energy spectra using the activation method, it takes a long time to measure the radiation from the activated metal foil, and there is a risk of exposure due to the nature of handling activated materials for the measurement.
[0007] The present disclosure has been made in consideration of the above circumstances, and aims to provide a technique that enables a neutron energy spectrum to be estimated simply and safely. [Means for solving the problem]
[0008] A neutron energy spectrum estimation method according to one embodiment of the present disclosure includes a first step in which a measurement device measures a soft error occurrence rate due to a neutron energy spectrum to be estimated for a semiconductor memory device having an SEU (Single Event Upset) cross-section that depends on a known neutron energy spectrum, and a second step in which an estimation device estimates and calculates the neutron energy spectrum to be estimated based on a result of comparing a provisional value of the soft error occurrence rate in the semiconductor memory device based on a provisional value of the neutron energy spectrum to be estimated, or based on a result of comparing true values of multiple soft error occurrence rates for multiple different semiconductor memory devices. Effect of the Invention
[0009] According to the present disclosure, a technique can be provided that enables a neutron energy spectrum to be estimated simply and safely. [Brief description of the drawings]
[0010] [Figure 1] FIG. 1 is a diagram showing an example of the configuration of an estimation system according to this embodiment. [Diagram 2] FIG. 2 is a diagram showing a method for estimating the intensity of a neutron energy spectrum. [Diagram 3] FIG. 3 is a diagram showing a specific example of a method for estimating the intensity of a neutron energy spectrum. [Figure 4] FIG. 4 is a diagram showing a method for estimating the shape of a neutron energy spectrum. [Diagram 5] FIG. 5 is a diagram showing a specific example of a method for estimating the shape of a neutron energy spectrum. [Figure 6] FIG. 6 is a diagram illustrating a hardware configuration of the estimation device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0011] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.
[0012] [overview] The present disclosure relates to a technique for measuring the energy distribution (neutron energy spectrum) of neutron rays, which are a type of radiation. In order to solve the above problems, the present disclosure discloses a method for estimating the neutron energy spectrum using the rate of soft errors caused by neutrons.
[0013] Specifically, a semiconductor memory device having a known neutron energy-dependent SEU (Single Event Upset) cross-section is placed in a neutron field, the soft error occurrence rate in the semiconductor memory device due to the neutron ray to be estimated is measured, and the neutron energy spectrum due to the neutron ray is estimated from the trend of the measured soft error occurrence rate.
[0014] The SEU cross section is the probability of soft errors occurring for each neutron energy in a semiconductor memory device. An SEU soft error is an event in which a single particle (neutron, proton, heavy particle, etc.) enters an LSI such as a semiconductor memory, and the charge generated by the nuclear reaction causes the bit data stored in the LSI (Large Scale Integration) to be inverted.
[0015] [Calculation method for soft error occurrence rate] The soft error occurrence rate in a semiconductor memory device can be calculated by multiplying the neutron energy spectrum by the SEU cross-section (soft error occurrence probability) of the semiconductor memory.
[0016] Specifically, the intensity of a neutron beam with energy En, which is the neutron energy spectrum, is expressed as φ(En) [1 / cm 2 / s] and the SEU cross section is σ SEU (En) [cm 2 / Mbit], the soft error rate, SER (Soft Error Rate) [1 / s / Mbit], can be calculated using equation (1).
[0017]
number
[0018] Therefore, in this embodiment, the neutron energy spectrum is estimated using the soft error occurrence rate according to formula (1).
[0019] [Estimation system] FIG. 1 is a diagram showing an example of the configuration of an estimation system 1 according to this embodiment.
[0020] The estimation system 1 includes a measurement device 11 that measures the soft error occurrence rate in a semiconductor memory device according to a neutron energy spectrum, and an estimation device 12 that estimates the neutron energy spectrum based on the trend of the measured soft error occurrence rate.
[0021] The measuring device 11 has a function of measuring the soft error occurrence rate due to a neutron energy spectrum to be estimated for a semiconductor memory device having an SEU cross-section (soft error occurrence probability) that depends on a known neutron energy spectrum.
[0022] The estimation device 12 has a function of estimating and calculating a true value of the soft error occurrence rate in the semiconductor memory device due to the neutron energy spectrum of the estimation target based on a comparison result between a provisional value of the soft error occurrence rate in the semiconductor memory device due to the provisional value of the neutron energy spectrum of the estimation target, or based on a comparison result between a plurality of true values of the soft error occurrence rates related to a plurality of different semiconductor memory devices.
[0023] For example, the estimation device 12 estimates and calculates the neutron energy spectrum of the estimation target by reflecting the difference between the true value and the provisional value of the soft error incidence rate in the provisional value of the neutron energy spectrum of the estimation target.
[0024] For example, the estimation unit 12 calculates the magnitude relationship of the soft error occurrence rate ratios when multiple true values of soft error occurrence rates related to multiple different semiconductor memory devices are normalized by one true value, searches for an energy region of the SEU cross-section that matches the magnitude relationship of the calculated soft error occurrence rate ratios, and estimates that the neutron energy spectrum of the searched energy region is large.
[0025] The measurement device 11 and the estimation device 12 may be realized by one device or functional unit.
[0026] [Estimation method] First, a method for estimating the intensity of a neutron energy spectrum will be described with reference to Fig. 2.
[0027] Step S101; First, the estimation device 12 determines a provisional value of the intensity of the neutron energy spectrum to be estimated, and calculates a provisional value of the soft error occurrence rate in the semiconductor memory to be used by multiplying the determined provisional value of the intensity of the neutron energy spectrum by the measured SEU cross-section of the semiconductor memory to be used.
[0028] For example, the estimation device 12 inputs provisional values of the intensity of the neutron energy spectrum to the graph of the neutron energy spectrum shown in Fig. 3(a).Then, the estimation device 12 multiplies the provisional values of the intensity of the neutron energy spectrum by the known SEU cross-sections of each of the field programmable gate arrays (FPGAs) 1 to 4 shown in Fig. 3(b), and inputs provisional values of the soft error occurrence rates in each of the FPGAs 1 to 4 obtained by the multiplication to the graph of the soft error occurrence rate shown in Fig. 3(c).
[0029] Step S102; Next, the user places a semiconductor memory having the SEU cross-section used in step S101 in a neutron field having a neutron energy spectrum to be estimated, and performs a neutron beam irradiation experiment on the semiconductor memory. After that, the measurement device 11 measures the soft error occurrence rate in the semiconductor memory in the neutron beam irradiation experiment. The estimation device 12 adds the measured soft error occurrence rate in the semiconductor memory as a true value to the graph in FIG. 3(c).
[0030] Step S103; Finally, the estimation device 12 compares the true value and the provisional value of the soft error occurrence rate in the semiconductor memory from the graph in Figure 3 (c) and compares the difference between the true value and the provisional value between semiconductor memories (FPGAs) to determine whether the provisional value of the neutron energy spectrum is close to the true value or not, and identifies the true value of the intensity of the neutron energy spectrum by performing fitting.
[0031] In the example shown in Fig. 3, since the true value of the soft error occurrence rate is about twice as large as the provisional value from the graph in Fig. 3(c), it can be estimated that the true intensity of the neutron energy spectrum is about twice as large as the provisional value. Therefore, the estimation device 12 adds twice the provisional value of the intensity of the neutron energy spectrum input in Fig. 3(a) to Fig. 3(a) as the true intensity of the neutron energy spectrum to be estimated.
[0032] Next, a method for estimating the shape of a neutron energy spectrum will be described with reference to Fig. 4, which is a diagram showing a method for estimating the shape of a neutron energy spectrum.
[0033] Step S201; First, the estimation device 12 determines a provisional value of the shape of the neutron energy spectrum to be estimated, and calculates a provisional value of the soft error occurrence rate in the semiconductor memory to be used by multiplying the determined provisional value of the shape of the neutron energy spectrum (more precisely, the provisional value of the intensity that forms the shape of the neutron energy spectrum) by the measured SEU cross-section of the semiconductor memory to be used.
[0034] For example, the estimation device 12 inputs provisional values of the shape of the neutron energy spectrum having a shape with a large component around 10 [MeV] to the graph of the neutron energy spectrum shown in Fig. 5(a). Then, the estimation device 12 multiplies the provisional values of the shape (intensity) of the neutron energy spectrum by the known SEU cross-sections of each of FPGAs 1 to 4 shown in Fig. 5(b), and inputs the provisional values of the soft error occurrence rates in each of FPGAs 1 to 4 obtained by the multiplication to the graph of the soft error occurrence rate shown in Fig. 5(c). At this time, the estimation device 12 calculates and inputs the soft error occurrence rate ratios obtained by normalizing the soft error occurrence rates of the other FPGAs 2 to 4 with the value of FPGA 1 set to 1.
[0035] Step S202; Next, the user places a semiconductor memory having the SEU cross section used in step S201 in a neutron field having a neutron energy spectrum to be estimated, and performs a neutron beam irradiation experiment on the semiconductor memory. After that, the measurement device 11 measures the soft error occurrence rate in the semiconductor memory by the neutron beam irradiation experiment. The estimation device 12 adds the measured soft error occurrence rate in the semiconductor memory as a true value to the graph in FIG. 5(c). At this time, the estimation device 12 calculates and inputs a soft error occurrence rate ratio in which the value of FPGA 1 is set to 1 and the soft error occurrence rates of the other FPGAs 2 to 4 are normalized, as in step S201.
[0036] Step S203; Finally, the estimation device 12 compares the true value and the provisional value of the soft error occurrence rate in the semiconductor memory from the graph in Figure 5 (c) and compares the difference between the true value and the provisional value between semiconductor memories (FPGAs) to determine whether the provisional value of the neutron energy spectrum is close to the true value or not, and identifies the true value of the shape of the neutron energy spectrum by performing fitting.
[0037] Here, the soft error occurrence rate ratio of each semiconductor memory (FPGA) depends on the SEU cross section. The size relationship of the SEU cross sections of each FPGA around 10 [MeV] where the components are large in the provisional shape values is "FPGA1 ≒ FPGA2>FPGA4>FPGA3" as shown in Figure 5 (b), so the provisional values of the soft error occurrence rate ratio shown in Figure 5 (c) have the same size relationship.
[0038] On the other hand, in the example shown in Figure 5(c), the true value of the soft error occurrence rate ratio measured in the neutron beam irradiation experiment is "FPGA2>FPGA4>FPGA3>FPGA1", so if we look for the energy region of the SEU cross-section that has a similar magnitude relationship in Figure 5(b), we can find that it is 1 [MeV].
[0039] Therefore, it can be estimated that the neutron energy spectrum to be estimated has a large component near 1 [MeV], not near 10 [MeV]. Therefore, the estimation device 12 adds the shape of the neutron energy spectrum having a large component near 1 [MeV] to Fig. 5(a) as the shape of the true value of the neutron energy spectrum to be estimated. For example, while maintaining the shape of the provisional value of the neutron energy spectrum, it slides so that the maximum value of the neutron energy spectrum becomes 1 [Mev].
[0040] As described above, the intensity and shape of the neutron energy spectrum can be estimated from the soft error occurrence rate in semiconductor memories.
[0041] [effect] According to this embodiment, the measuring device 11 measures the soft error occurrence rate due to the neutron energy spectrum to be estimated for a semiconductor memory device having an SEU cross-section that depends on a known neutron energy spectrum, and the estimation device 12 estimates and calculates the intensity of the neutron energy spectrum to be estimated based on the comparison result between the provisional value of the soft error occurrence rate in the semiconductor memory device due to the neutron energy spectrum to be estimated and the provisional value of the soft error occurrence rate in the semiconductor memory device due to the provisional value of the neutron energy spectrum to be estimated, thereby making it possible to estimate the neutron energy spectrum simply and safely.
[0042] According to this embodiment, the measurement device 11 measures the soft error occurrence rate due to the neutron energy spectrum to be estimated for a semiconductor memory device having an SEU cross-section that depends on a known neutron energy spectrum, and the estimation device 12 estimates and calculates the shape of the neutron energy spectrum to be estimated based on the comparison results between multiple true values of soft error occurrence rates for multiple different semiconductor memory devices, regarding the true value of the soft error occurrence rate in the semiconductor memory device due to the neutron energy spectrum to be estimated. Therefore, the neutron energy spectrum can be estimated simply and safely.
[0043] In other words, it is possible to measure neutron energy spectra at accelerator facilities where it is difficult to measure using the time-of-flight method. It also makes it possible to measure neutron energy spectra in the high-energy range without requiring a neutron detector with high time resolution. It also makes it possible to measure neutron energy spectra during irradiation experiments. In addition, because no radioactive materials are handled, the risk of exposure during measurements can be reduced.
[0044] [others] The present disclosure is not limited to the above-described embodiment, and various modifications are possible within the scope of the present disclosure.
[0045] The estimation device 12 of the present embodiment described above can be realized, for example, by using a general-purpose computer system including a CPU 901, a memory 902, a storage 903, a communication device 904, an input device 905, and an output device 906, as shown in FIG. 6.
[0046] The memory 902 and the storage 903 are storage devices. In the computer system, the CPU 901 executes a predetermined program loaded onto the memory 902, whereby each function of the estimation device 12 is realized.
[0047] The estimation device 12 may be implemented in one computer. The estimation device 12 may be implemented in multiple computers. The estimation device 12 may be a virtual machine implemented in a computer.
[0048] The program for the estimation device 12 can be stored in a computer-readable recording medium such as a HDD, an SSD, a USB memory, a CD, or a DVD. The computer-readable recording medium is, for example, a non-transitory recording medium. The program for the estimation device 12 can also be distributed via a communication network. [Explanation of symbols]
[0049] 1. Estimation System 11 Measuring equipment 12 Estimation device 901 CPU 902 Memory 903 Storage 904 Communication equipment 905 Input Device 906 Output Device
Claims
1. In a method for estimating a neutron energy spectrum, A first step of measuring a soft error occurrence rate due to a neutron energy spectrum of an estimation target for a semiconductor memory device having a single event upset (SEU) cross section depending on a known neutron energy spectrum by a measurement device; a second step in which the estimation device estimates and calculates a true value of a soft error occurrence rate in the semiconductor memory device according to the neutron energy spectrum of the estimation target based on a result of comparison between a provisional value of the neutron energy spectrum of the estimation target and a provisional value of the soft error occurrence rate in the semiconductor memory device, or based on a result of comparison between a plurality of true values of soft error occurrence rates related to a plurality of different semiconductor memory devices; A method for estimating the neutron energy spectrum.
2. In the second step, 2. The method for estimating a neutron energy spectrum according to claim 1, further comprising the step of: estimating and calculating the neutron energy spectrum of the estimation target by reflecting a difference between a true value and a provisional value of the soft error occurrence rate in the provisional value of the neutron energy spectrum of the estimation target.
3. In the second step, 2. The method for estimating a neutron energy spectrum according to claim 1, further comprising the steps of: calculating a magnitude relationship between soft error occurrence rate ratios when a plurality of true values of soft error occurrence rates relating to a plurality of different semiconductor memory devices are normalized by one true value; searching for an energy region of an SEU cross section that matches the magnitude relationship between the soft error occurrence rate ratios; and estimating that the neutron energy spectrum in the energy region is large.