SiC power device in-orbit single-particle hard error risk probability estimation method

Abstract

The invention discloses a SiC power device in-orbit single-particle hard error risk probability estimation method, and relates to the technical field of anti-radiation guarantee of power device space application. Based on a ground radiation test and a Weibull data statistical law, a space radiation environment is simulated in combination with simulation software; the method can estimate the probability of single-particle hard errors during space on-orbit application of the SiC device at low cost, is low in implementation difficulty, reduces the cost of an on-orbit task of a spacecraft, enhances the reliability during the on-orbit period, guarantees the safety of on-orbit flight, belongs to the technical field of anti-radiation guarantee of space application of power devices, has very high economic value, and is suitable for popularization and application. The method is of great significance in promoting development of novel aerospace power supplies.

Description

Technical Field

[0001] The present invention relates to the field of radiation resistance assurance technology for space applications of power devices, and in particular to a method for estimating the risk probability of single particle hard errors of SiC power devices on-orbit. Background Art

[0002] Silicon carbide (SiC) is a third-generation semiconductor material. Compared with silicon (Si) materials, it has superior electrical and thermal properties such as high thermal conductivity, high breakdown field strength, high electron saturation rate, and strong radiation resistance. These excellent properties make it expected to become the preferred high-voltage power device material in aerospace system circuits. Therefore, in the field of aerospace, power SiC devices may gradually replace silicon-based power devices with more ideal performance advantages, and have the potential to be used in satellites, aircraft, detectors, and power electronic systems in space in the future.

[0003] However, there are a large number of high-energy particles in the space radiation environment composed of the Earth's capture zone, galactic cosmic rays, and solar cosmic rays. These high-energy space particles in the space radiation environment will cause radiation effects on the devices in the aerospace equipment. The radiation effects are mainly divided into total dose effects, displacement effects, and single particle effects. Among them, the single particle effect can cause leakage in the circuit system, resulting in reduced reliability, and in severe cases, hard error failures will cause the device to fail to work, which seriously threatens the normal operation of the aerospace equipment. Power SiC devices have a strong ability to resist total doses, but they are more sensitive to single particle effects when working in a high-voltage state. The devices are prone to single particle burnout (SEB) and single particle gate breakdown (SEGR). Therefore, it is necessary to carry out single particle effect testing, but ground simulation tests can only obtain single particle burnout and single particle gate breakdown phenomena, which cannot accurately reflect the on-orbit radiation environment and single particle risks. In the prior art, there is a problem that it is difficult to accurately estimate the probability of single particle hard errors when power SiC devices are used in space. Summary of the invention

[0004] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a method for estimating the risk probability of single-particle hard errors of SiC power devices on orbit. The present invention is based on ground heavy ion irradiation tests, combined with existing space radiation environment simulation software and device failure cross-section data statistical laws, to accurately predict the number of single-particle hard errors that occur in power SiC devices on orbit per unit time, thereby solving the reliability assessment problem of single-particle effects caused by heavy ion irradiation of power SiC devices during on-orbit operation in the prior art.

[0005] The present invention adopts the following technical solutions to solve the above technical problems:

[0006] A method for estimating the risk probability of a single-particle hard error of a SiC power device on-orbit is provided according to the present invention, comprising:

[0007] Step 1: Perform a single-particle hard error irradiation test on the SiC power device, monitor the relationship between the drain or gate current and the injection amount in real time during the irradiation process, and obtain n curve data of the relationship between the drain or gate current and the injection amount;

[0008] Step 2: Filter out the single-particle effect failure injection data when the SiC power device has single-particle burnout and single-particle gate penetration from the curve data;

[0009] Step 3: Perform Weibull data statistics on the single particle effect failure dose data to obtain a Weibull statistical distribution diagram of the single particle failure dose of the SiC power device, and perform linear fitting based on the Weibull statistical distribution diagram of the single particle failure dose, and use the fitting slope and intercept to obtain the Weibull shape parameter β and scale parameter η;

[0010] Step 4: Based on the Weibull shape parameter β and scale parameter η, the average failure fluence data F under a single linear energy transfer LET is obtained. avg ;

[0011] Step 5: Average failure dose data F under a single LET avg Take the inverse to get the failure cross section;

[0012] Repeat the single-event hard error irradiation test on other SiC power devices of the same model until the failure cross sections {σ1,σ2,σ3…σ j}, j>2, j is the number of LETs, one single-particle hard error irradiation test corresponds to one LET, σ j′ For the jth ′ The failure cross section under LET, 1≤j′≤j; and the failure cross section {σ1,σ2,σ3…σ j} is fitted to obtain the failure section function σ(LET);

[0013] Step 6: Multiply the failure cross-section function σ(LET) by cos(θ) to obtain the failure cross-section function σ(LET, θ) at different incident angles, where θ is the angle between the particle incident path and the particle perpendicular to the incident path of the SiC power device. According to σ(LET, θ), the failure cross-section formula is integrated over the entire space to obtain the average failure cross-section σ AVG (LET);

[0014] Step 7: Use the space radiation environment software to input the orbital data of the spacecraft, add the environment model, which includes solar particles, geomagnetic captured particles and galactic cosmic ray particles, and obtain the differential flux energy spectrum data h(LET) of the space radiation environment in the spacecraft cabin;

[0015] Step 8: According to the average failure cross section σ AVGThe on-orbit single-event hard error rate Fr is calculated based on the total flux data of the space radiation environment h(LET).

[0016] As a further optimization scheme of the method for estimating the risk probability of single-particle hard errors of SiC power devices on-orbit described in the present invention, the test method adopted for the single-particle hard error irradiation test in step one is based on the single-particle burnout and single-particle gate penetration tests described in the 1080 method in the U.S. military standard MIL-STD-750.

[0017] As a further optimization scheme of the method for estimating the probability of single-particle hard error risk of an on-orbit SiC power device according to the present invention, step 2 is specifically as follows:

[0018] Step 201: first set the reference current I r , from each curve data in step 1, obtain all the curves that are greater than the reference current I r data points; then find the slope of the selected data points and select the maximum slope β from the slopes max , get the injection data corresponding to the maximum slope;

[0019] Step 202: Repeat step 201 to obtain single-event failure injection data {f1, f2, f3…f n}, n>2, f n′ is the single event effect failure dose data of the n′th drain or gate current variation curve, 1≤n′≤n.

[0020] As a further optimization scheme of the method for estimating the risk probability of single-particle hard errors of SiC power devices on-orbit described in the present invention, the thickness of the shielding layer of the spacecraft is 3 mm aluminum.

[0021] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0022] The present invention combines ground irradiation tests and data statistical methods to propose a method for estimating the probability of single-particle hard errors occurring in power SiC devices on orbit, which is used to solve the reliability assessment problem of single-particle effects occurring in power SiC devices irradiated by heavy ions during on-orbit operation in the prior art, and further predict the performance degradation and device life of SiC devices in space applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flow chart for implementing the method of the present invention;

[0024] Figure 2 This is the result of a typical device single event hard error irradiation test;

[0025] Figure 3It is the Weibull statistical distribution diagram of the device single-particle failure dose;

[0026] Figure 4 is the average failure cross-section diagram of the device under different LET;

[0027] Figure 5 is the differential flux LET spectrum of the GEO orbit space radiation environment. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0029] The present invention provides a method for estimating the probability of single-particle hard errors occurring in SiC power devices on orbit, which belongs to the field of radiation protection technology for space applications of power devices. First, the device failure injection is obtained through ground heavy ion irradiation tests, and then the single-particle hard error failure cross-section statistics are obtained using Weibull data statistical analysis and the influence of angle effects. Combined with the particle flux spectrum in the spacecraft cabin on a specific orbit, the risk probability of single-particle hard error events occurring in SiC power devices on orbit is predicted, which will greatly reduce the cost of space missions and improve the reliability of spacecraft on-orbit operation.

[0030] The present invention is specific in that: based on the existing space radiation environment simulation database, the total particle flux spectrum in the spacecraft cabin on a typical orbit is obtained; the device failure cross-section data is obtained by performing Weibull statistical analysis and fitting on the ground heavy ion irradiation test data; since the particle movement direction has three-dimensional spatial randomness and equal probability in each direction, in order to ensure the omnidirectionality of space particles, the influence of the particle incident angle is added to the failure cross-section. Finally, the above data is integrated to obtain the probability of single-particle hard errors when SiC devices are used in space on-orbit, which will greatly reduce the cost of space missions, improve the reliability of spacecraft during on-orbit operation, and is of great significance to the development of new aerospace power sources.

[0031] A method for estimating the risk probability of a single-particle hard error of a SiC power device on-orbit, comprising:

[0032] Step 1: Perform a single-particle hard error irradiation test on the SiC power device, monitor the relationship between the drain or gate current and the injection amount in real time during the irradiation process, and obtain n curve data of the relationship between the drain or gate current and the injection amount;

[0033] Step 2: Filter out the fluence data when single particle burnout and single particle gate penetration of SiC power devices occur from the acquired curve data;

[0034] Step 201: first set the reference current I r , from each curve data in step 1, obtain all the curves that are greater than the reference current Ir data points; then find the slope of the selected data points and select the maximum slope β from the slopes max , get the injection data corresponding to the maximum slope;

[0035] Step 202: Repeat step 201 to obtain single-event failure fluence data of n drain or gate current variation curves, that is, {f1, f2, f3…f n}, n>2, f n， is the single event effect failure injection data of the n′th drain or gate current variation curve, 1≤n′≤n;

[0036] Step 3: Single-event effect failure dose data {f1,f2,f3…f n} Perform Weibull data statistics to obtain the Weibull statistical distribution diagram of the single particle failure dose of SiC power devices, and perform linear fitting based on the Weibull statistical distribution diagram of the single particle failure dose, and use the fitting slope and intercept to obtain the Weibull shape parameter β and scale parameter η;

[0037] The statistical formula for Weibull data is:

[0038]

[0039] Among them, F(f) is the median of failure data (related to n), f is the fluence data, β is the Weibull shape parameter, and η is the Weibull scale parameter;

[0040] Step 4: Based on the Weibull shape parameter β and scale parameter η, according to the formula Get the average failure dose data F under a single linear energy transfer LET avg , where Γ(*) is the gamma function;

[0041] Step 5: Calculate the average failure dose data F under a single LET avg Take the inverse to get the failure cross section σ L , repeat the single-event hard error irradiation test on other SiC power devices of the same model until the failure cross sections {η1,η2,η3…η j}, j>2, j is the number of LETs, one single-particle hard error irradiation test corresponds to one LET, σ j′ For the jth ′ The failure cross section under LET, 1≤j′≤j, and the failure cross section is fitted to obtain the failure cross section function σ(LET);

[0042] Step 6: Multiply the failure cross-section function σ(LET) by cos(θ) to obtain the failure cross-section function σ(LET, θ) at different incident angles, where θ is the angle between the particle incident path and the particle perpendicular to the incident path of the SiC power device. The failure section formula is integrated over the entire space, and finally the average failure section σ is obtained AVG (LET);

[0043] Step 7: Use the space radiation environment software to input the orbital data of the spacecraft, add the environment model, which includes solar particles, geomagnetic captured particles and galactic cosmic ray particles, and the shielding layer thickness is 3 mm aluminum. Finally, obtain the differential flux energy spectrum data h(LET) of the space radiation environment in the spacecraft cabin;

[0044] Step 8: According to the average failure cross section σ AVG (LET) and the total flux data of space radiation environment h(LET), through the formula Calculate the on-orbit single-event hard error rate Fr.

[0045] Aiming at the on-orbit application requirements of space SiC devices, this specification provides a method for estimating the probability of single-particle hard errors in power SiC devices on-orbit. The method can be used to solve the reliability assessment problem of single-particle effects caused by heavy ion irradiation in power SiC devices during on-orbit operation in the prior art, and belongs to the field of radiation resistance assurance of power devices.

[0046] like Figure 1 As shown in the flow chart, the present invention provides a method for estimating the probability of a single-particle hard error event occurring in a power SiC device on-orbit, comprising the following steps:

[0047] Step 1: Select a commercial SiC power device and perform a single-particle irradiation test on it. The test method is based on the single-particle burnout and single-particle gate penetration tests described in the 1080 method of the US military standard MIL-STD-750. During the irradiation process, the relationship between the drain or gate current and the flux is monitored in real time to obtain the relationship curve between the drain or gate current and the injection amount.

[0048] In the embodiment of the present invention, the source-drain voltage V DS SiC MOSFET at 900V, subjected to krypton (Kr) ion test: LET = 33 MeV cm 2 / mg irradiation test, the beam is perpendicular to the device surface, and the total fluence is 10 10 cm -2 The device gate current variation data under Figure 2 shown.

[0049] Step 2: Process the acquired curve to obtain the injection data {f1,f2,f3…fn}(n=9), as shown in Table 1.

[0050] The specific process is as follows:

[0051] against Figure 2 The relationship curve between gate current and injection volume, setting the reference current value I r 100nA (the data point where the gate current exceeds 100nA has a single-particle hard error phenomenon), select the gate current to exceed the reference current value I r Data points {P1, P2, P3…P i} (assuming there are i points), find the slope of each point {β1,β2,β3…β i}, select the maximum slope β mae , and obtain the injection data corresponding to the maximum slope data point. Figure 2 There are 9 curves in the equation, and finally the fluence data {f1, f2, f3…f9} are obtained. The fluence data are shown in Table 1.

[0052] Table 1 Statistics of median rank F(f) of injection data and failure data

[0053] Serial number <![CDATA[Dose to failure (cm -2 )]]> F(f) 1 0.7E+09 0.0745 2 1.47E+09 0.1809 3 1.73E+09 0.2872 4 3.09E+09 0.3936 5 4.72E+09 0.5000 6 5.04E+09 0.6064 7 9.08E+09 0.7128 8 1.37E+10 0.8191 9 3.18E+10 0.9255

[0054] Step 3: Figure 3 As shown, we perform Weibull data statistical analysis on the fluence data {f1, f2, f3...f9} to obtain the shape parameter η and scale parameter β in the Weibull parameters;

[0055] The specific process is as follows:

[0056] First, calculate the median rank of failure data F(f). The median rank of failure data is the middle number. The formula is as follows. The obtained data is shown in Table 1.

[0057]

[0058] Among them, f is the injection amount, k is the fault number. In the sample experiment, the fault numbers are sorted according to the size of the injection amount data {f1, f2, f3...f9}, and n is the total number of samples.

[0059] Secondly, the Weibull formula is transformed into:

[0060]

[0061] Where: β is the Weibull shape parameter and η is the Weibull scale parameter.

[0062] The formula transforms into a linear function of the x variable:

[0063] y=βx-βln(η)

[0064] in: x=ln(f).

[0065] according to Figure 3 The slope and intercept of the fitted linear function are combined with the linear function of the x variable deformed by the Weibull formula to calculate the specific values ​​of the shape parameter η and the scale parameter β.

[0066] Step 4: Based on the shape parameter η and the scale parameter β, we get LET = 33 MeV cm 2 Average failure dose data F at / mg avg .

[0067] The specific process is:

[0068] According to the shape parameter η and scale parameter β obtained in step 3, the average failure injection F is calculated. avg , the average failure dose formula is:

[0069]

[0070] Step 5: Based on the F under a single LET avg The failure section σ is calculated L Repeat the test and the above steps to obtain at least three failure cross-section data {σ1,σ2,σ3…σ j}(j>2), and fit the failure section data to obtain the fitted failure section σ(LET), Figure 4 is the fitted average failure cross-section diagram.

[0071] The specific process is:

[0072] According to the average failure injection data F avg Get the failure section σ L , the formula is:

[0073] σ L =F avg -1

[0074] Repeat the test process to obtain the failure cross section {σ1,σ2,σ3…σ j}(j>2) to fit the data, and the fitting formula is as follows:

[0075]

[0076] Where LET1 is the linear energy transfer value at the beginning of the single-particle hard error phenomenon, σ L (LET1)=0.

[0077] Step 6. Failure section angle correlation. Since the direction of particle movement has three-dimensional spatial randomness and equal probability in each direction, in order to ensure the omnidirectionality of spatial particles, the influence of the particle incident angle should be added to the failure section, and the incident angle influence factor should be added to σ(LET). It is usually believed that the single-particle hard error phenomenon caused by vertical incidence is more serious. As the incident angle increases, the probability of a single-particle hard error will gradually decrease. There is an inverse relationship between the directional cross section and the incident angle, which follows the cosine law, and θ represents the incident angle. Therefore, the formula for the failure section σ(LET, θ) is:

[0078]

[0079] According to the formula The failure section formula is integrated over the entire space, and finally the average failure section σ is obtained AVG (LET):

[0080]

[0081] Step 7. In order to simulate the space radiation environment for the on-orbit application of SiC devices, we use the SPENVIS radiation environment software tool of the European Space Agency, call its space radiation environment particle spectrum calculation module, input typical space orbit data, add solar particles, captured protons and galactic cosmic rays and other particle environments, set the shielding layer thickness of the spacecraft to 3 mm aluminum, and obtain the function h(LET) of the differential flux energy spectrum of the space radiation environment in the GEO orbit (geosynchronous orbit) after the spacecraft is shielded as a function of LET, as shown in Figure 5 It is the differential flux LET spectrum of the typical orbit GEO space radiation environment.

[0082] Step 8: Combine the average failure cross section σ AVG (LET) and the total flux data of the space radiation environment h(LET) are used to calculate the probability of a single particle hard error of the device on orbit to be 0.01658 / bit / day. The specific formula is:

[0083]

[0084] Where: h(LET) is the differential flux LET spectrum of the space radiation environment behind the shielding layer, σ AVG (LET) is the average failure cross section, Fr represents the number of single-particle hard errors that occur in one day, and the unit is: / bit / day.

[0085] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for estimating the risk probability of single-particle hard errors of SiC power devices on-orbit, characterized in that: include: Step 1: Perform a single-particle hard error irradiation test on the SiC power device, monitor the relationship between the drain or gate current and the injection amount in real time during the irradiation process, and obtain n curve data of the relationship between the drain or gate current and the injection amount; Step 2: Filter out the single-particle effect failure injection data when the SiC power device has single-particle burnout and single-particle gate penetration from the curve data; Step 3: Perform Weibull data statistics on the single particle effect failure dose data to obtain a Weibull statistical distribution diagram of the single particle failure dose of the SiC power device, and perform linear fitting based on the Weibull statistical distribution diagram of the single particle failure dose, and use the fitting slope and intercept to obtain the Weibull shape parameter β and scale parameter η; Step 4: Based on the Weibull shape parameter β and scale parameter η, obtain the average failure fluence data F under a single linear energy transfer LET avg ; Step 5: Average failure dose data F under a single LET avg Take the inverse to get the failure cross section; Repeat the single-event hard error irradiation test on other SiC power devices of the same model until the failure cross sections {σ1,σ2,σ3…σ j }, j>2, j is the number of LETs, one single-particle hard error irradiation test corresponds to one LET, σ j′ For the jth ′ The failure cross section under LET, 1≤j′≤j; and the failure cross section {σ1,σ2,σ3…σ j } is fitted to obtain the failure section function σ(LET); Step 6: Multiply the failure cross-section function σ(LET) by cos(θ) to obtain the failure cross-section function σ(LET, θ) at different incident angles, where θ is the angle between the particle incident path and the particle perpendicular to the incident path of the SiC power device. According to σ(LET, θ), the failure cross-section formula is integrated over the entire space to obtain the average failure cross-section σ AVG (LET); Step 7: Use the space radiation environment software to input the orbital data of the spacecraft, add the environment model, which includes solar particles, geomagnetic captured particles and galactic cosmic ray particles, and obtain the differential flux energy spectrum data h(LET) of the space radiation environment in the spacecraft cabin; Step 8: According to the average failure cross section σ AVG The on-orbit single-event hard error rate Fr is calculated based on the total flux data of the space radiation environment h(LET).

2. The method for estimating the risk probability of single-particle hard errors of SiC power devices on-orbit according to claim 1 is characterized in that: The test method used in the single-particle hard error irradiation test in step 1 is based on the single-particle burnout and single-particle gate penetration tests described in the 1080 method of the US military standard MIL-STD-750.

3. The method for estimating the risk probability of single-particle hard errors of SiC power devices on-orbit according to claim 1 is characterized in that: Step 2 is as follows: Step 201: first set the reference current I r , from each curve data in step 1, obtain all the curves that are greater than the reference current I r data points; Then find the slope of the selected data points and select the maximum slope β from the slopes max , get the injection data corresponding to the maximum slope; Step 202: Repeat step 201 to obtain single-event failure injection data {f1, f2, f3…f n }, n>2, f n′ is the single event effect failure dose data of the n′th drain or gate current variation curve, 1≤n′≤n.

4. The method for estimating the risk probability of single-particle hard errors of SiC power devices on-orbit according to claim 1 is characterized in that: The spacecraft's shielding is 3 mm thick aluminum.

Images