CN113109860B - Method for predicting heavy ion single event effect section curve of device - Google Patents

Abstract

The invention discloses a method for predicting the heavy ion single particle effect cross-section curve of a device, which includes the following steps: carrying out a proton or neutron single particle effect experiment on the device, obtaining the proton or neutron single particle effect cross-section data and performing Weibull fitting to obtain the proton or neutron single particle effect cross-section curve. Neutron single particle effect cross-section curve; construct a chip structure model, Monte Carlo calculation of the secondary particle LET spectrum produced in the sensitive layer due to nuclear reactions between protons or neutrons of different energies and device materials; estimate two heavy ion single particle effect cross-section data points And preliminarily fit a heavy ion single particle effect cross-section curve; integrate the heavy ion single particle effect cross-section curve with the secondary particle LET spectrum to obtain the proton or neutron single particle effect cross-section of the device; compare the integrated calculation data with the experimental data , adjust the fitting parameters of the heavy ion single particle effect cross-section curve, and repeat the integration until the deviation between the calculated data and the experimental data is within a certain range, and the required heavy ion single particle effect cross-section curve is obtained.

Description

A method for predicting the heavy ion single particle effect cross-section curve of devices

Technical field

The invention belongs to the research field of space radiation effect simulation test technology and radiation resistance reinforcement technology, and relates to a method for predicting the heavy ion single particle effect cross-section curve of a device based on the proton or neutron single particle effect cross-section.

Background technique

In the space radiation environment, single event effects are an important factor affecting the reliability of spacecraft electronic systems, and heavy ions and protons are the main sources of single event effects in electronic devices. Protons mainly cause single-particle effects by ionizing and depositing energy in the sensitive area through secondary particles produced by nuclear reactions with device materials, while heavy ions cause single-particle effects by depositing energy in the sensitive area through direct ionization. Ground simulation of heavy ion single particle effects usually uses heavy ion irradiation devices generated by accelerators. In the experiment, more than 5 heavy ion LET value points are selected to obtain the relationship curve between the single particle effect cross section of the device and the LET value, so as to carry out the device's anti-single particle effect. Evaluation of abilities.

With the improvement of device performance, increased integration and the development of packaging technology, the flip-chip process has become the main packaging form. Its process characteristics are: flipping the device onto the substrate, and connecting the two in the form of solder bumps , a substrate with a thickness of several hundred microns is located above the sensitive area of the device. Due to the limited ion energy and range of the heavy ion accelerator, it is difficult for heavy ions to penetrate the substrate and reach the sensitive area of the device. Therefore, when flip-chip devices are used for heavy ion accelerator experiments, the device usually needs to be opened and thinned, which can easily cause damage to the device during the operation. On the other hand, for heavy ions with high LET values and high atomic numbers, their single The nuclear energy is further reduced, and even for thinned devices, it is difficult to meet the requirements for ion range in experiments, which makes it extremely difficult to evaluate the heavy ion single particle effect of flip-chip devices. Medium and high-energy protons, due to their low LET value, have small energy loss when penetrating materials and have long range. They can effectively pass through the device package and substrate to reach the sensitive area of the device to induce single particle effects, and obtain the complete proton single particle effect cross-section curve of the device.

In view of the above problems and current situation, considering the correlation between the generation mechanisms of proton and heavy ion single particle effects, a method is proposed to predict the heavy ion single particle effect cross-section curve of the device based on the proton or neutron single particle effect experimental data. Through the proton experimental data and Corresponding simulation calculations can obtain the heavy ion single particle effect cross-section curve of the device, effectively solving the practical problem of difficult evaluation of the anti-heavy ion single particle performance of flip-chip devices.

Patent application number 200710177960.5, publication number CN100538378C, titled "Method for Obtaining the Relationship between Single Particle Effect Cross Section and Heavy Ion Linear Energy Transfer", provides an experimental method based on the heavy ion single particle effect cross section of a heavy ion accelerator test device; patent application No. 202010982765.5, the public number is CN112230081A, and the name is "A method for calculating the equivalent LET value of the pulsed laser single particle effect test". It provides a method for using the pulsed laser single particle experimental data to equivalent heavy ion single particle effect cross sections with different LET values. . The patent application number is 201711173677.5, the publication number is CN108008289B, and the name is "A method for obtaining proton single particle effect cross-section". It provides a method for obtaining proton single particle effect cross-section data based on the heavy ion single particle effect cross-section curve. None of these three methods involves the method of predicting the heavy ion single particle effect cross-section curve based on the device proton single particle effect cross section.

Contents of the invention

The invention provides a method for predicting the heavy ion single particle effect cross-section curve of a device. The complete heavy ion single particle effect cross-section curve of a flip-chip device can be obtained without carrying out a heavy ion single-particle effect experiment. This method provides a method for evaluating flip-chip devices. The device's ability to withstand heavy ion single particles provides an effective technical means, overcoming the existing technology's shortcomings in the difficulty of evaluating the heavy ion single particle effect of flip-chip devices.

The technical solution of the present invention is:

A method for predicting the heavy ion single particle effect cross-section curve of a device is special in that it includes the following steps:

Step 1: Conduct a proton or neutron single particle effect experiment on the device to be tested, and obtain the proton or neutron single particle effect cross-section curve of the device;

Step 2: Conduct longitudinal analysis of the device, construct a device structure model, and simulate and calculate the LET spectrum of secondary particles produced by the nuclear reaction of protons or neutrons of different energies with the device material;

Step 3: Estimate two heavy ion single particle effect cross-section data points through proton data, and initially fit a heavy ion single particle effect cross-section curve;

Step 4: Integrate the heavy ion single particle effect cross-section curve obtained in step 3 and the secondary particle LET spectra of different proton or neutron energies calculated in step 2 to obtain the proton or neutron single particle effect at different energies. Particle Effect Cross Section;

Step 5: Compare the integral calculation data in Step 4 with the proton or neutron single particle effect experimental data in Step 1. When the deviation exceeds the set range, continuously adjust the fitting parameters of the device heavy ion single particle effect cross-section curve, and repeat Step 4: Until the deviation between the integral calculation data and the proton or neutron experimental data is within the set range, the heavy ion single particle effect cross-section curve at this time is the desired one.

Further, the step one is specifically:

Carry out device proton or neutron single particle effect experiments, perform Weibull fitting on the obtained proton or neutron experimental data, and obtain the fitted proton or neutron single particle effect cross-section curve σ _p (E _p ):

In the formula, σ _sat-p is the saturation cross section of proton or neutron single particle effect, unit cm ² ; E _p0 is the energy threshold of proton or neutron single particle effect, unit MeV; W is the scale parameter; S is the shape parameter; E _p It is the energy of proton or neutron in MeV.

Further, the second step is specifically:

2.1) Conduct a longitudinal cross-sectional analysis of the device, obtain the thickness and material composition of its packaging, heat dissipation silicone grease, substrate, and sensitive layer, and construct a sensitive volume model of the device;

2.2) Use Monte Carlo particle transport simulation software to calculate the probability that protons or neutrons with energy E _p will react with the device material to produce secondary particles with LET value L in the sensitive layer, and obtain the probability function p (E _p ,L ) versus LET value.

Further, the third step is specifically:

3.1) Based on the proton single particle effect saturation cross section combined with formula (1-2), initially estimate the heavy ion single particle effect saturation cross section data point σ _sat-ion :

σ _sat-ion =10 ⁶ ×σ _sat-p (1-2)

3.2) Calculate the equivalent LET value of the secondary particles generated in the sensitive layer by the proton with energy E _p near the inflection point of the proton or neutron single particle effect cross-section curve in step 1, and statistically the equivalent LET value is greater than the LET threshold The probability p(L>L ₀ ), another heavy ion single particle effect cross-section data point σ _ion (L) is estimated by formula (1-3):

In the formula, L is the equivalent LET value, L ₀ is the LET threshold;

3.3) Combine the two heavy ion single particle effect cross sections estimated in 3.1) and 3.2) and initially fit a heavy ion single particle effect cross section curve according to formula (1-4)

Further, the fourth step is specifically:

The integral expression used in step four is:

In the formula, σ _p (E _p ) is the single particle effect cross section of a proton or neutron with energy E _p ; p (E _p ,L) is the secondary particle LET generated by the nuclear reaction of protons or neutrons with energy E _p with the device material. The probability that the value is L; σ _ion (L) is the heavy ion single particle effect cross section with a LET value of L.

Further, the device under test is a flip-chip device.

The beneficial effects of the present invention are:

1. The present invention can obtain the complete heavy ion single particle effect cross-section curve of the flip-chip device without carrying out heavy ion single particle effect experiments, solves the technical bottleneck of difficult evaluation of the anti-heavy ion single particle effect capability of the flip-chip device, and greatly improves the efficiency of the flip-chip device. Greatly reduces experimental costs.

2. The proton or neutron single particle effect test can be carried out in the air, and the proton or neutron single particle effect cross section can be tested without unpacking or thinning the device, reducing damage to the device and lowering the difficulty of the test.

3. This invention starts from the fundamental mechanism of the nuclear reaction of protons or neutrons with device materials to produce secondary particles that trigger single particle effects. The physical concept is clear, and the calculation time and data accuracy meet the needs of practical applications.

Description of drawings

Figure 1 is a flow chart of an embodiment of the present invention;

Figure 2 is a sensitive volume structure model containing multi-layer material information of the device;

Figure 3 is the relationship curve between the probability function of secondary particles produced in the sensitive layer due to nuclear reactions between protons of different energies and device materials and the LET value;

Figure 4 is a preliminary fitting cross-section curve of heavy ion single particle effect;

Figure 5 is a comparison chart when the deviation between proton experimental data and simulation calculation data meets the requirements;

Figure 6 is a comparison diagram between the final calibrated heavy ion single particle effect cross-section curve and the heavy ion experimental data.

Detailed ways

Taking a flip-chip FPGA device as an example, specific embodiments of the present invention will be further described in detail with reference to the accompanying drawings. The following embodiments are only used to illustrate the present invention, but are not intended to limit the scope of the present invention.

Figure 1 is a flow chart of a method of predicting the heavy ion single particle effect cross section curve of a device based on the proton single particle effect cross section of the present invention. The steps of this method are described in detail with reference to Figure 1.

S1] Carry out proton single particle effect experiments on flip-chip FPGA devices, and perform Weibull fitting on the experimental data to obtain the fitted proton single particle effect cross-section function, whose expression is:

S2] Conduct longitudinal analysis on the flip-chip FPGA, and construct a sensitive volume structure model of the device based on the longitudinal material and process information of the device, as shown in Figure 2. Use the Monte Carlo particle transport simulation software Geant4 to calculate the probability that a proton with energy E _p will react with the device material to produce a secondary particle with an LET value L in the sensitive layer, and obtain the probability function p(E _p ,L) and the LET value. The relationship curve is shown in Figure 3.

S3】Preliminary estimation of heavy ion single particle flipping saturation cross section σ _sat-ion =10 ⁶ ×σ _sat-p =10 ⁶ ×2.1×10 ^-15 cm ² /bit=2.1×10 ^-9 cm based on formula (1-2) ² /bit. Using Geant4 to calculate, the equivalent LET value of 40MeV protons in the sensitive layer is 1.71MeV·cm ² /mg. At this time, the probability p (L>L ₀ )=3.17×10 ^-5 ; according to formula (1-3) Combined with the proton single particle effect cross section at 40 MeV, the heavy ion single particle flipping cross section with a LET value of 1.71 MeV·cm ² /mg is estimated to be . From the above two points, a preliminary heavy ion single particle effect cross-section curve is obtained through Weibull fitting, as shown in Figure 4. Its expression is:

S4] Compare the expression (1-7) in S3] with the secondary particle LET spectrum p(E _p ,L) produced in the sensitive layer by the nuclear reaction between protons of different energies and the device material in S2] according to the formula (1-5) Perform integral calculations to obtain proton single particle flipping cross sections at different energies.

S5] Compare the integral calculation data in S4] with the proton single particle effect experimental data in S1]. If the deviation is large at this time, continuously adjust the fitting parameters of the device heavy ion single particle effect cross-section curve, and repeat S4] until the integration When the deviation between the calculated data and the proton experimental data is reduced to a certain range, as shown in Figure 5, the heavy ion single particle effect cross-section curve at this time is the required curve, see Figure 6. Among them, the heavy ion single particle effect cross-section Weibull curve expression (1-8) obtained after multiple adjustments:

Perform Weibull fitting on the heavy ion single particle effect experimental data in Figure 6. The expression is shown in (1-9):

It can be seen that there is good consistency between the calibration calculation results and the heavy ion single particle experimental results.

In another embodiment, the method of the present invention can also be used to predict the heavy ion single particle effect cross section curve of the device based on the neutron single particle effect cross section.

Claims (4)

1. A method for predicting the heavy ion single particle effect cross-section curve of a device, which is characterized by including the following steps: Step 1: Conduct a proton or neutron single particle effect experiment on the device to be tested, and obtain the proton or neutron single particle effect cross-section curve of the device; Carry out device proton or neutron single particle effect experiments, perform Weibull fitting on the obtained proton or neutron experimental data, and obtain the fitted proton or neutron single particle effect cross-section curve σ _p (E _p ): In the formula, σ _sat-p is the saturation cross section of proton or neutron single particle effect, unit cm ² ; E _p0 is the energy threshold of proton or neutron single particle effect, unit MeV; W is the scale parameter; S is the shape parameter; E _p is the energy of proton or neutron in MeV; Step 2: Conduct longitudinal analysis of the device, construct a device structure model, and simulate and calculate the LET spectrum of secondary particles produced by the nuclear reaction of protons or neutrons of different energies with the device material; 2.2) Use Monte Carlo particle transport simulation software to calculate the probability that protons or neutrons with energy E _p will react with the device material to produce secondary particles with LET value L in the sensitive layer, and obtain the probability function p (E _p ,L ) and the relationship curve between LET value; Step 3: Estimate two heavy ion single particle effect cross-section data points through proton data, and initially fit a heavy ion single particle effect cross-section curve; 3.1) Based on the proton single particle effect saturation cross section combined with formula (1-2), initially estimate the heavy ion single particle effect saturation cross section data point σ _sat-ion : σ _sat-ion =10 ⁶ ×σ _sat-p (1–2) σ _sat-p is the single particle effect saturation cross section of proton or neutron; 3.2) Calculate the equivalent LET value L _avg of the secondary particles produced by the proton with energy Ep near the inflection point of the proton or neutron single particle effect section curve in step 1 in the sensitive layer through Monte Carlo calculation, and count the LET values greater than the LET threshold Probability p(L>L ₀ ), estimate another heavy ion single particle effect cross-section data point σ _ion (L _avg ) according to formula (1-3): In the formula, L _avg is the equivalent LET value, L is the secondary particle LET value, L ₀ is the LET threshold, σ _p (Ep) is the proton or neutron single particle effect cross section; 3.3) Combine the two heavy ion single particle effect cross sections estimated in 3.1) and 3.2) and initially fit a heavy ion single particle effect cross section curve according to formula (1-4) Step 4: Integrate the heavy ion single particle effect cross-section curve obtained in step 3 and the secondary particle LET spectra of different proton or neutron energies calculated in step 2 to obtain the proton or neutron single particle effect at different energies. Particle Effect Cross Section; Step 5: Compare the integral calculation data in Step 4 with the proton or neutron single particle effect experimental data in Step 1. When the deviation exceeds the set range, continuously adjust the fitting parameters of the device heavy ion single particle effect cross-section curve, and repeat Step 4: Until the deviation between the integral calculation data and the proton or neutron experimental data is within the set range, the heavy ion single particle effect cross-section curve at this time is the desired one. 2. The method for predicting the heavy ion single particle effect cross-section curve of a device as claimed in claim 1, characterized in that the step two is specifically: 2.1) Conduct a longitudinal cross-section analysis of the device to obtain the thickness and material composition of its packaging, heat dissipation silicone grease, substrate, and sensitive layer, and construct a sensitive volume model of the device. 3. The method for predicting the heavy ion single particle effect cross-section curve of a device as claimed in claim 1, characterized in that the step four is specifically: The integral expression used in step four is: In the formula, σ _p (E _p ) is the single particle effect cross section of a proton or neutron with energy E _p ; p (E _p ,L) is the secondary particle LET generated by the nuclear reaction of protons or neutrons with energy E _p with the device material. The probability that the value is L; σ _ion (L) is the heavy ion single particle effect cross section with a LET value of L. 4. The method for predicting the heavy ion single event effect cross-section curve of a device according to any one of claims 1 to 3, characterized in that the device to be tested is a flip-chip device.