Method for evaluating soft error rate of thermal neutrons in atmospheric environment

Abstract

The invention relates to the technical field of semiconductor device reliability evaluation, and discloses an atmospheric environment thermal neutron soft error rate evaluation method, which comprises the steps of chip microstructure reverse analysis, Monte Carlo simulation, accelerator test and LET spectrum correlation and error rate calculation. According to the method for evaluating the soft error rate of the thermal neutrons in the atmospheric environment, LET spectrums in the atmospheric environment and in the ground accelerator environment are generated through Monte Carlo simulation, and by taking a chip critical linear energy transfer value as a correlation factor, the soft error rate of the thermal neutrons in the atmospheric environment is evaluated according to the overlapping ratio of the LET spectrums in the atmospheric environment and in the ground accelerator environment. According to the method, the soft error rate in the actual atmospheric environment is obtained by combining test data in an accelerator test, large errors caused by LET spectrum differences are avoided, the accuracy of the soft error in the actual atmospheric environment is improved, and a quantifiable and verifiable evaluation technology is provided for the anti-radiation design of a chip.

Description

Technical Field

[0001] The present application relates to the technical field of semiconductor device reliability evaluation, and specifically to a method for evaluating the soft error rate of thermal neutrons in an atmospheric environment. Background Art

[0002] When evaluating the soft error rate of a chip in the prior art, the SER is usually extrapolated only through the simple ratio of the neutron flux of the accelerator test to the atmospheric flux, ignoring the difference between the atmospheric LET spectrum and the accelerator LET spectrum. For example, the moderation process of high-energy neutrons in the atmosphere results in a different LET spectrum distribution from that of accelerator neutrons. If no spectrum overlap correction is performed, it may lead to an increase in the SER calculation error, resulting in a significant deviation in the SER evaluation result in the atmospheric environment, unable to meet the requirements of high-reliability chip design, and increasing the chip failure risk and R & D cost.

[0003] Therefore, a more reliable SER evaluation technology is urgently needed. Summary of the Invention

[0004] The purpose of the present application is to provide a method for evaluating the soft error rate of thermal neutrons in an atmospheric environment to solve the technical problems proposed in the above background art.

[0005] To achieve the above purpose, the present application discloses the following technical solutions: A method for evaluating the soft error rate of thermal neutrons in an atmospheric environment, the method comprising the following steps:

[0006] Reverse analysis of the chip microstructure: Obtain the material composition, thickness, and geometric structure parameters of the sensitive area of the chip through microscopic analysis technology;

[0007] Monte Carlo simulation: Based on the material composition, the thickness, and the geometric structure parameters, use the Monte Carlo method to simulate the energy deposition of thermal neutrons in the chip in the atmospheric environment and the ground accelerator environment, and generate the LET spectra in the atmospheric environment and the ground accelerator environment;

[0008] Accelerator test: Perform a thermal neutron irradiation test on the chip through a ground accelerator, and record the soft error events of the chip at a set fluence;

[0009] LET spectrum correlation and error rate calculation: Use the critical line energy transfer value of the chip as the correlation factor, and calculate the soft error rate in the actual atmospheric environment through the overlap ratio of the LET spectra in the atmospheric environment and the ground accelerator environment, in combination with the test data in the accelerator test.

[0010] Preferably, in the reverse analysis of the chip microstructure, the microscopic analysis techniques include any one or more of scanning electron microscopy, focused ion beam, or transmission electron microscopy, and process node parameters of the sensitive area of the chip are recorded. The process node parameters include transistor channel thickness, Fin height, and SRAM cell layout.

[0011] Preferably, in the Monte Carlo simulation, the calculation of the LET spectrum is based on the energy deposition of the chip material, the particle path length, and the sensitivity coefficient of the chip structure to LET to quantify the LET distribution in different environments.

[0012] Preferably, the calculation of the LET value in the LET spectrum is specifically as follows:

[0013]

[0014] where ΔE is the energy deposition in the sensitive area; Δs is the particle path length; Data 结构参数 is any one of the values after quantification of material composition parameters, thickness, and geometric structure parameters, α is the sensitivity coefficient determined by the chip process node, LET th is the critical line performance energy transfer value of the chip, and E env is the environmental correction factor determined according to the humidity and temperature of the atmospheric environment.

[0015] Preferably, the sensitivity coefficient is defined as:

[0016]

[0017] where Δ Fin is the Fin pitch; Tch is the transistor channel thickness; D ox is the oxide layer thickness.

[0018] Preferably, the environmental correction factor is defined as:

[0019]

[0020] where H is the humidity of the atmospheric environment and T is the temperature of the atmospheric environment.

[0021] Preferably, in the accelerator test, the fluence rate of the accelerator test is determined according to the atmospheric neutron flux, the accelerator neutron flux, and the test time.

[0022] Preferably, in the LET spectrum correlation and error rate calculation, the soft error rate in the actual atmospheric environment is calculated by the following formula:

[0023]

[0024] where the SER加速器 is the soft error rate in the ground accelerator environment obtained from the accelerator test, F 大气 is the atmospheric neutron flux, F 加速器 is the accelerator neutron flux, P 大气LET is the overlap integral corresponding to the LET spectrum in the atmospheric environment, P 加速器LET is the overlap integral corresponding to the LET spectrum in the ground accelerator environment, T decay is the time decay factor determined according to the chip working time and the environmental radiation background.

[0025] Preferably, the time decay factor is defined as:

[0026]

[0027] where B 辐射 is the environmental radiation background, t work is the chip working time.

[0028] Advantageous effects: The evaluation method for the soft error rate of thermal neutrons in the atmospheric environment of the present application generates the LET spectra in the atmospheric environment and the ground accelerator environment through Monte Carlo simulation, and uses the critical line performance energy transfer value of the chip as the correlation factor. Through the overlap ratio of the LET spectra in the atmospheric environment and the ground accelerator environment, combined with the test data in the accelerator test, the soft error rate in the actual atmospheric environment is obtained, avoiding large errors caused by differences in LET spectra, improving the accuracy of soft errors in the actual atmospheric environment, and providing a quantifiable and verifiable evaluation technology for chip radiation resistance design. Description of the Drawings

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0030] Figure 1 is the flowchart of the evaluation method for the soft error rate of thermal neutrons in the atmospheric environment provided by the embodiments of the present application. Detailed Embodiments

[0031] The following will clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by the present application.

[0032] In this article, the term "comprising" is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a series of elements includes not only those elements but also other elements not expressly listed, or elements that are inherent to such process, method, article, or apparatus. Without further limitation, an element qualified by the statement "comprising..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0033] This embodiment provides an evaluation method for the soft error rate of thermal neutrons in the atmospheric environment as shown in Figure 1 which includes the following steps:

[0034] S1 - Reverse analysis of chip microstructure: Obtain the material composition, thickness, and geometric structure parameters of the sensitive area of the chip through microscopic analysis techniques. The microscopic analysis techniques include any one or more of scanning electron microscopy (SEM), focused ion beam (FIB), or transmission electron microscopy (TEM), and record the process node parameters of the sensitive area of the chip. The process node parameters include the transistor channel thickness Tch (measuring the vertical distance between the gate and the source / drain regions in the FinFET structure through SEM images), Fin height (determining the vertical height of the Fin structure through FIB cross-section analysis), and SRAM cell layout (such as cell area Sram_area, metal interconnect layer spacing M_inter): Extract the layout data through image processing software (such as OpenCV). Specifically, this step includes:

[0035] S11 - Microscopic photography - Use a high-resolution scanning electron microscope (SEM), focused ion beam (FIB), or transmission electron microscope (TEM) to image the chip layer by layer, with the magnification covering nanoscale details (such as transistor channel thickness, Fin height), and record the microscopic images of each layer;

[0036] S12 - Image processing and stitching: Denoise, binarize, and enhance the edges of the microscopic images through image processing software (such as ImageJ or MATLAB), and then generate a complete chip layout image through a stitching algorithm (such as Stitching);

[0037] S13 - Geometric parameter extraction: Use computer vision technology (such as OpenCV) to extract the transistor channel thickness Tch, Fin height, and SRAM cell layout from the processed image.

[0038] S2 - Monte Carlo Simulation: Based on the material composition, the thickness, and the geometric structure parameters, use the Monte Carlo method to simulate the energy deposition of thermal neutrons in the chip under the atmospheric environment and the ground accelerator environment, and generate the LET spectra under the atmospheric environment and the ground accelerator environment. This step specifically includes:

[0039] S21 - The Monte Carlo simulation is carried out based on the following parameters:

[0040] Energy distribution characteristics of thermal neutrons in the atmospheric environment: Adopt a thermal neutron energy of 0.025 eV;

[0041] Chip material parameters: Consist of a silicon-based substrate (density 2.33 g / cm 3 , atomic composition Si / O ratio 1:2), an oxide layer (SiO2, density 2.2 g / cm 3 ), and the density and atomic composition of the metal interconnection layer (Cu, density 8.96 g / cm 3 );

[0042] Geometric structure parameters of the sensitive area: Construct a geometric model based on the process node parameters;

[0043] S22 - Energy deposition calculation: Simulate the transport process of neutrons in the chip through a Monte Carlo code (such as MCNP or Geant4), record the path length and energy deposition of each particle, and generate the corresponding LET spectrum.

[0044] S3 - Accelerator test: Conduct a thermal neutron irradiation test on the chip through a ground accelerator. The accelerator type uses a spallation neutron source or a reactor neutron source. Record the soft error events of the chip under a set fluence, and based on any one of the existing technologies, obtain the soft error rate under the ground accelerator environment.

[0045] S4 - LET spectrum correlation and error rate calculation: Use the critical line energy transfer value of the chip (the minimum LET threshold for the chip sensitive area to trigger a soft error, which is determined by fitting the accelerator test data to obtain the LET sensitivity threshold of the chip) as the correlation factor, through the overlap ratio of the LET spectra under the atmospheric environment and the ground accelerator environment (the intersection part of the two LET spectra above the chip critical LET value), combined with the test data in the accelerator test (1. Accelerator test conditions: a. Irradiation parameters: neutron fluence (unit: neutrons / cm 2 ), energy (thermal neutrons or specific fast neutrons), flux (neutrons / cm 2 / s); b. Chip operating conditions: voltage (such as 5V), temperature (such as 25°C), functional mode (such as the communication status of the CANFD chip); 2. Test results: a. Soft error event record: the number of soft errors that occur in the chip under the set fluence (such as single event latch-up SEL, single event functional interrupt SEFI, etc.); b. LET threshold determination: determine the critical linear energy transfer value of the chip through experiments (such as the SIT1042AQ chip first triggers SEL at 37.5 MeV·cm 2 / mg), and calculate the soft error rate in the actual atmospheric environment

[0046] In this embodiment, in the Monte Carlo simulation, the calculation of the LET spectrum is based on the energy deposition of the chip material, the particle path length, and the sensitivity coefficient of the chip structure to LET to quantify the LET distribution in different environments. Specifically, the calculation of the LET value in the LET spectrum is as follows:

[0047]

[0048] Among them, ΔE is the energy deposition in the sensitive area, and the total energy deposition of the neutron trajectory is statistically calculated through a Monte Carlo code (such as MCNP); Δs is the particle path length, and the travel distance of the neutron in the chip is recorded through Monte Carlo simulation; Data 结构参数 is any one of the values obtained by quantifying the material composition parameters, thickness, and geometric structure parameters, α is the sensitivity coefficient determined by the chip process node, LET th is the critical linear energy transfer value of the chip, and E env is the environmental correction factor determined according to the humidity and temperature of the atmospheric environment. By introducing the sensitivity coefficient, the contribution of microstructures such as FinFET and SRAM layout to LET is quantified, and the calculation accuracy is improved. Secondly, by introducing the environmental correction factor, it is dynamically adjusted according to real-time humidity and temperature data, applicable to different geographical environments, and thus closer to the actual scenario.

[0049] In one implementation, the sensitivity coefficient is defined as:

[0050]

[0051] Among them, Δ Fin is the Fin pitch, measured by FIB; Tch is the transistor channel thickness; D ox is the oxide layer thickness, measured by an ellipsometer (such as J.A. Woollam VASE).

[0052] In one implementation, the environmental correction factor is defined as:

[0053]

[0054] Among them, H is the humidity of the atmospheric environment, and T is the temperature of the atmospheric environment.

[0055] With the above design of the LET spectrum calculation, by quantifying the influence of the structure and the environment, it provides more accurate technical support for the evaluation of the chip soft error rate, and significantly improves the reliability of semiconductor devices in complex environments.

[0056] In this embodiment, in the accelerator test, the fluence rate of the accelerator test is determined according to the atmospheric neutron flux, the accelerator neutron flux, and the test time. That is Preferably, by adjusting the parameters to compensate for the neutron flux difference in the actual application area of the chip. Therefore, the optimized fluence rate calculation formula is as follows:

[0057]

[0058] Among them, F 大气 is the atmospheric neutron flux, F 加速器 is the accelerator neutron flux, t time is the test time, D comp is the dynamic compensation factor, which is calculated according to the neutron flux fluctuation ΔF (unit: n / (cm 2 ·s)) monitored in real time and the temperature of the atmospheric environment during the test. ΔF is the neutron flux fluctuation, which is measured in real time by a scintillation detector (such as BC501A). By the ratio of the atmospheric and accelerator neutron fluxes, combined with the test time, the fluence of the accelerator test is adjusted to the level of the equivalent atmospheric environment. Then, by introducing the dynamic compensation factor, ΔF reflects the real-time fluctuation of the accelerator neutron flux. For example, if the accelerator output is unstable, resulting in an increase in ΔF, then the compensation factor D comp decreases, and thus the fluence rate is reduced to offset the fluctuation effect. With the introduction of temperature, when the temperature rises, the negative value of the exponential term increases, and D comp decreases, simulating the change in the neutron path caused by the thermal expansion of the chip material at high temperature, thereby dynamically adjusting the fluence rate. In this way, by monitoring ΔF and temperature in real time, the fluence rate is dynamically corrected to ensure that the test conditions are consistent with the actual atmospheric environment.

[0059] Furthermore, in order to ensure that the experimental data reflect the real single-event effect, in this embodiment, the fluence rate of the accelerator test includes a dynamic adjustment mechanism. By monitoring the cumulative fluence and setting a threshold upper limit, the dynamic fluence rate is calculated to correct the fluence rate calculated previously, avoiding the error accumulation caused by the particle overlap effect. Specifically, Among them, N cum is the cumulative fluence value, N th is the fluence threshold (unit n / cm 2) Ensure that the fluence does not exceed the cumulative radiation tolerance upper limit of the chip during its life cycle, calculated based on the chip life L (unit: year) and the typical working environment. γ is the non-linear attenuation coefficient, dynamically adjusted according to the chip process node. In the formula, constitutes a fluence attenuation factor. When the cumulative fluence N cum approaches the threshold N th the factor approaches 0, and the fluence rate drops sharply to avoid exceeding the tolerance limit of the chip life. With the introduction of the non-linear attenuation coefficient, the attenuation of advanced processes (≤7nm) is steeper, reflecting their sensitivity to radiation damage.

[0060] Through the calculation of the fluence rate in this embodiment and the combination of multi-parameter dynamic coupling, the accuracy and practicality of chip soft error rate evaluation are improved.

[0061] In this embodiment, in the LET spectrum correlation and error rate calculation, the soft error rate under the actual atmospheric environment is calculated by the following formula:

[0062]

[0063] Among them, the SER 加速器 is the soft error rate under the ground accelerator environment obtained based on the accelerator test, F 大气 is the atmospheric neutron flux, F 加速器 is the accelerator neutron flux, P 大气LET is the overlap integral corresponding to the LET spectrum under the atmospheric environment, P 加速器LET is the overlap integral corresponding to the LET spectrum under the ground accelerator environment, T decay is the time attenuation factor determined according to the chip working time and the environmental radiation background. The time attenuation factor is defined as:

[0064]

[0065] Among them, B 辐射 is the environmental radiation background (unit: n / (cm2·s)), t work is the chip working time (unit: hour).

[0066] The overlap integral of the LET spectrum under the atmospheric environment and the LET spectrum under the ground accelerator environment reflects the difference in their energy deposition distributions. If the proportion of high-LET particles in the atmospheric environment is higher, the overlap integral corresponding to the LET spectrum under the atmospheric environment increases, resulting in an increase in the SER result. Therefore, through P 大气LET and P 加速器LETThe introduction corrects the errors caused by the LET distribution in different environments and improves the calculation accuracy of SER. The introduction of the time decay factor directly correlates the chip lifespan with the environmental radiation background, enabling the model to adapt to environmental changes and improving the prediction accuracy.

[0067] In summary, for the method for evaluating the thermal neutron soft error rate in the atmospheric environment of this embodiment, by generating the LET spectra in the atmospheric environment and the LET spectra in the ground accelerator environment through Monte Carlo simulation, and combining dynamic fluence rate adjustment and time decay, the accuracy, environmental adaptability, and long-term reliability of the evaluation of the thermal neutron soft error rate in the atmospheric environment are improved, providing a quantifiable and verifiable evaluation support for the radiation-resistant design of the chip. Moreover, the method for evaluating the thermal neutron soft error rate in the atmospheric environment of this embodiment has the following technical characteristics:

[0068] (1) Improve the calculation accuracy of SER: By correcting the overlapping integral of the LET spectra in the atmospheric environment and the LET spectra in the ground accelerator environment, the errors caused by the LET spectrum differences are reduced, and the calculation accuracy of SER is improved;

[0069] (2) Dynamic environmental adaptability: By introducing the neutron flux fluctuation and temperature correction coefficients, the evaluation technology can adapt to temperature changes and accelerator flux fluctuations, reducing the result errors;

[0070] (3) The time decay factor enables the SER prediction to cover the entire life cycle of the chip, and has a high credibility in high-reliability scenarios.

[0071] In the embodiments provided in the present application, it should be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code, or any appropriate combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, microcontroller, microprocessor, or other electronic units designed to implement the functions described herein, or a combination thereof. For software implementation, part or all of the processes of the embodiments can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable storage medium or transmitted as one or more instructions or codes on a computer-readable storage medium. The computer-readable storage medium includes computer storage media and communication media, where the communication media includes any medium that facilitates the transmission of a computer program from one place to another. The storage media can be any available medium that can be accessed by a computer. The computer-readable storage medium can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM, or other optical disk storage, magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer.

[0072] Finally, it should be noted that the above are only the preferred embodiments of the present application and are not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An evaluation method for the soft error rate of thermal neutrons in the atmospheric environment, characterized in that The method includes the following steps: Reverse analysis of chip microstructure: Obtain the material composition, thickness, and geometric structure parameters of the sensitive area of the chip through microscopic analysis techniques; Monte Carlo simulation: Based on the material composition, the thickness, and the geometric structure parameters, use the Monte Carlo method to simulate the energy deposition of thermal neutrons in the chip in the atmospheric environment and the ground accelerator environment, and generate the LET spectra in the atmospheric environment and the ground accelerator environment; Accelerator test: Conduct a thermal neutron irradiation test on the chip through a ground accelerator, and record the soft error events of the chip at a set fluence; LET spectrum correlation and error rate calculation: Using the critical line energy transfer value of the chip as the correlation factor, calculate the soft error rate in the actual atmospheric environment through the overlap ratio of the LET spectra in the atmospheric environment and the ground accelerator environment, combined with the test data in the accelerator test.

2. The evaluation method of the thermal neutron soft error rate in the atmospheric environment according to claim 1, characterized in that In the reverse analysis of the chip microstructure, the microscopic analysis techniques include any one or more of a scanning electron microscope, a focused ion beam, or a transmission electron microscope, and record the process node parameters of the sensitive area of the chip. The process node parameters include the transistor channel thickness, the Fin height, and the SRAM cell layout.

3. The evaluation method of the soft error rate of thermal neutrons in the atmospheric environment according to claim 2, characterized in that, In the Monte Carlo simulation, the calculation of the LET spectrum is based on the energy deposition of the chip material, the particle path length, and the sensitivity coefficient of the chip structure to LET to quantify the LET distribution in different environments.

4. The method for evaluating the soft error rate of thermal neutrons in the atmospheric environment according to claim 3, wherein, The calculation of the LET value in the LET spectrum is specifically: Among them, ΔE is the energy deposition in the sensitive area; Δs is the particle path length; Data 结构参数 is any one of the values after quantization of the material composition parameters, thickness, and geometric structure parameters. α is the sensitivity coefficient determined by the chip process node. LET th is the critical line energy transfer value of the chip, and E env is the environmental correction factor determined according to the humidity and temperature of the atmospheric environment.

5. The evaluation method of the soft error rate of thermal neutrons in the atmospheric environment according to claim 4, characterized in that The sensitivity coefficient is defined as: Among them, Δ Fin is the Fin pitch; Tch is the transistor channel thickness; D ox is the oxide layer thickness.

6. The method for evaluating the soft error rate of thermal neutrons in the atmospheric environment according to claim 4 or 5, characterized in that, The environmental correction factor is defined as: Where H is the humidity of the atmospheric environment and T is the temperature of the atmospheric environment.

7. The method for evaluating the soft error rate of thermal neutrons in the atmospheric environment according to any one of claims 1-6, characterized in that In the accelerator test, the fluence rate of the accelerator test is determined according to the atmospheric neutron flux, the accelerator neutron flux, and the test time.

8. The method for evaluating the soft error rate of thermal neutrons in the atmospheric environment according to claim 1, wherein In the LET spectrum correlation and error rate calculation, the soft error rate in the actual atmospheric environment is calculated through the following formula: Among them, the SER 加速器 is the soft error rate in the ground accelerator environment obtained based on the accelerator test, F 大气 is the atmospheric neutron flux, F 加速器 is the accelerator neutron flux, P 大气LET is the overlap integral corresponding to the LET spectrum in the atmospheric environment, P 加速器LET is the overlap integral corresponding to the LET spectrum in the ground accelerator environment, T decay is the time decay factor determined according to the chip working time and the environmental radiation background.

9. The evaluation method of the soft error rate of thermal neutrons in the atmospheric environment according to claim 8, wherein The time decay factor is defined as: Among them, B 辐射 is the environmental radiation background, and t work is the chip working time.

Images