Single event effective data acquisition method for flip chip circuit

Abstract

The present application relates to a kind of single particle effective data acquisition methods of flip welding circuit, comprising: test sample substrate thinning;Test heavy ion selection;Test circuit temperature monitoring;Data validity determination;Thinning thickness measurement and effective data calculation.Compared with the traditional test method, using the single particle test evaluation method in the present application, the single particle test efficiency of flip welding process circuit can be effectively improved.Through temperature control, multiple verification and error calibration, etc., the accuracy and effectiveness of the test data of the flip welding process circuit in the single particle test are provided, effectively filling the blank of the single particle evaluation method of the flip welding circuit used in spaceflight, and providing strong support for the radiation hardening of flip welding process spaceflight integrated circuit.

Description

Technical Field

[0001] This invention relates to a method for acquiring single-event effective data of flip-chip circuits, belonging to the field of single-event effect testing of aerospace integrated circuits. Background Technology

[0002] With the completion and successful operation of China's space station, my country has joined the ranks of the world's leading space powers. Facing major space projects such as deep space exploration and manned lunar landing, there is an urgent need for various high-performance aerospace integrated circuits to meet increasingly complex application requirements. Faced with increasingly complex circuit functions, flip-chip bonding has become one of the mainstream packaging technologies for advanced process nodes of 28nm and below. However, due to the special characteristics of flip-chip bonding and the limited ion range of domestic accelerators, the samples required for single-event experiments must undergo substrate thinning to ensure that ions can reach the active region of the chip. Therefore, for today's complex aerospace integrated circuits, there is an urgent need for a single-event testing procedure to evaluate flip-chip bonding circuits, characterize the single-event effect characteristics of target circuits under ion irradiation, and provide guidance for radiation hardening of flip-chip bonding aerospace integrated circuits. Summary of the Invention

[0003] The technical problem solved by this invention is to overcome the shortcomings of the prior art and provide a method for acquiring effective single-event data of flip-chip circuits, thereby improving the effectiveness of single-event test data of flip-chip circuits and providing technical support for radiation resistance test evaluation of aerospace integrated circuits.

[0004] The technical solution of this invention is: a method for acquiring single-particle effective data of flip-chip circuits, the method comprising the following steps:

[0005] S1. Thin the substrate of the flip-chip soldering circuit sample and record the thinning thickness;

[0006] S2. Select heavy ions suitable for flip-chip bonding process circuits to conduct single-particle tests on the thinned flip-chip bonding process circuit samples to obtain the single-particle error number of different samples under the same heavy ion irradiation and the single-particle error number of the same sample under different heavy ion irradiation.

[0007] S3. After the single-particle experiment is completed, the experimental data is analyzed to determine the validity of the experimental data. If the experimental data is valid, proceed to step S4; otherwise, replace the flip-chip soldering process circuit sample and repeat steps S1 to S3.

[0008] S4. Perform thickness measurement and calibration on the flip-chip bonding process circuit samples. Based on the calibration results, select valid flip-chip bonding process circuit samples from the experimental samples. Then, execute steps S5 and S6 to obtain the single-particle effective data corresponding to heavy ions. The single-particle effective data includes the effective LEETe and the single-particle error cross section σ.m ;

[0009] S5. Based on the single-particle test results of the effective flip-chip soldering circuit samples, calculate the single-particle error cross section σ of heavy ions. m ;

[0010] S6. Based on the average thickness d of the substrate thinning of the effective flip-chip bonding process circuit sample... r-mean The effective LET of heavy ions was obtained by simulation using heavy ion LET calculation software.

[0011] Preferably, during the single-particle test, the operating temperature of the flip-chip soldering circuit sample is monitored in real time. If the operating temperature of the flip-chip soldering circuit sample is greater than or equal to the temperature alarm threshold, the test is stopped immediately, and the test is resumed after the operating temperature of the flip-chip soldering circuit sample returns to between [T1, T2], where [T1, T2] is the temperature range of the flip-chip soldering circuit under rated operation.

[0012] Preferably, the thickness l of the thinned flip-chip bonding process circuit sample satisfies 20μm < l < 40μm, and the uniformity of the substrate surface meets the preset conditions.

[0013] Preferably, the range of the heavy ions selected for the flip-chip bonding process circuit is greater than 40 μm.

[0014] Preferably, the LET values ​​of the selected heavy ions suitable for flip-chip bonding circuits cover three energy ranges:

[0015] 0~10MeV·cm 2 / mg, 30-40MeV·cm 2 / mg, 60~100MeV·cm 2 / mg.

[0016] Preferably, the validity of experimental data is determined by the simultaneous fulfillment of the following two conditions: if the experimental data is considered valid, then the experimental data is considered invalid.

[0017] The first condition is: the single-particle error number of different samples under the same ion irradiation satisfies:

[0018]

[0019] Where, α i This represents the single-particle error number in the i-th experiment;

[0020] n represents the total number of test samples;

[0021] α max Indicates the maximum number of single-particle errors for the test sample;

[0022] α minRepresents the minimum single-particle error number for the test sample;

[0023] ε and τ are constants;

[0024] The second condition is that the number of single-particle errors (SPOIs) of the same sample under different ion irradiations must satisfy the requirement that the number of SPOIs caused by irradiation with high LET values ​​is greater than that caused by irradiation with low LET values.

[0025] Preferably, the method for thickness measurement and calibration of the flip-chip bonding process circuit sample is as follows: the flip-chip bonding process circuit sample is longitudinally cut along the diagonal, and the maximum substrate thickness d of the flip-chip bonding process circuit sample is determined. r-max and minimum thickness d r-min Calculate the average thickness d of the substrate thinning in the flip-chip soldering process circuit sample. r-mean If d r-max -d r-min Greater than d r-min If 10% of the samples are invalid, the flip-chip soldering circuit sample is considered invalid; otherwise, the flip-chip soldering circuit sample is considered valid.

[0026] Preferably, the average thickness d of the substrate thinning for effective flip-chip bonding process circuit samples r-mean =(d r-max +d r-min ) / 2.

[0027] Preferably, the effective LETe is obtained by the following method:

[0028] S6.1 Substitute the surface energy Es of the ions incident on the flip-chip bonding process circuit sample into the heavy ion LET calculation software to obtain the surface range μ of the incident ions;

[0029] S6.2. Subtract the substrate thinning thickness from the surface range μ obtained in S6.1 to obtain the effective range μ-d of the incident ions. r-mean ;

[0030] S6.3. Using heavy ion LET calculation software, the effective range was iteratively calculated to be μ-d. r-mean The energy and LET value of the heavy ion at this time are: the energy at this time is the effective energy Ee, and the corresponding LET is the effective LET.

[0031] Preferably, the single-particle error cross section σ of the heavy ion m It is obtained through the following method:

[0032] Single-event error cross section = Total number of single-event errors / Total irradiation flux

[0033] The total number of single-event errors refers to the total number of single-event errors generated in single-event experiments corresponding to heavy ions;

[0034] Total irradiation flux refers to the total irradiation flux of heavy ions in a single-particle experiment corresponding to heavy ions.

[0035] The advantages of this invention compared to the prior art are:

[0036] (1) The single-event test result evaluation method for flip-chip bonding process circuits proposed in this invention provides accurate test data for single-event effective data through multiple verifications and error calibrations, and provides strong support for the evaluation of single-event effects of flip-chip bonding process aerospace integrated circuits.

[0037] (2) This invention clarifies the selection of single-event test and heavy ion selection for flip-chip bonding process circuits, effectively improving the efficiency of single-event test. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the single-particle effective data acquisition method for flip-chip soldering circuits in an embodiment of the present invention. Detailed Implementation

[0039] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0040] like Figure 1 As shown, the present invention provides a method for acquiring single-event effective data of flip-chip circuits, the method comprising the following steps:

[0041] S1. Thin the substrate of the flip-chip soldering circuit sample and record the thinning thickness;

[0042] The specific steps for thinning are as follows:

[0043] (1.1) Remove the thermal layer metal cover plate from the surface of the circuit sample;

[0044] (1.2) Because the flip-chip circuit is back-facing, during single-particle experiments, accelerated ions can only enter from the back of the chip, passing through the substrate to reach the active region. However, due to the limited heavy ion energy and range of domestic accelerators, substrate thinning is required. The substrate thickness of the circuit is reduced from several hundred micrometers to the target thickness d using a grinding device, ranging from 30μm to 50μm. The distance from the substrate to the active region of the aerospace-grade flip-chip circuit is M, and the thinning thickness is K. Therefore, the actual thinned thickness l = MK, and the thickness l of the thinned flip-chip circuit sample must satisfy 20μm < l < 40μm.

[0045] (1.3) During the subsequent thinning, it is necessary to ensure that the uniformity of the substrate surface meets the preset conditions and that the thinning does not cause damage to the circuit. In a specific embodiment of the present invention, the uniformity of the substrate surface meets the following conditions: points in the thinned flip-chip bonding process circuit sample are selected as base points, the step size is set to x, and points in the front, back, left, and right directions with a distance of x from the base point are defined as uniformity measurement points. The thickness error between these uniformity measurement points and the base point is less than 5%. In this embodiment, the step size is selected as 1 μm.

[0046] S2. Select heavy ions suitable for flip-chip bonding process circuits to conduct single-particle tests on the thinned flip-chip bonding process circuit samples to obtain the single-particle error number of different samples under the same heavy ion irradiation and the single-particle error number of the same sample under different heavy ion irradiation.

[0047] Due to the unique characteristic of flip-chip bonding circuits with the substrate facing upwards, the range of heavy ions selected for flip-chip bonding circuits must be greater than 40 μm, and the ion range provided by the accelerator must be greater than 40 μm.

[0048] To facilitate subsequent data analysis and on-orbit prediction, the LET values ​​of the heavy ions selected for flip-chip bonding circuits cover three energy ranges:

[0049] 0~10MeV·cm 2 / mg, 30-40MeV·cm 2 / mg, 60~100MeV·cm 2 / mg.

[0050] Based on the actual beam output capabilities of domestic heavy ion accelerators, the available heavy ion accelerators in China include C, O, F, Al, and Si (LET values ​​ranging from 0 to 10 MeV·cm). 2 / mg), Kr (LET value between 30 and 40 MeV·cm 2 (mg) and Xe, Ta, Bi, etc. (LET values ​​in the range of 60–100 MeV·cm⁻¹) 2 / mg).

[0051] To conduct single-event experiments on flip-chip circuits, at least five different heavy ion types with varying LET values ​​are required to accurately characterize their single-event effects. To ensure the effectiveness of ion incident radiation and effective coverage of the single-event error cross-section, appropriate irradiation ions are selected based on the actual beam output capabilities of domestic heavy ion accelerators.

[0052] In single-event experiments, the experiments are conducted in ascending order of LET values.

[0053] To prevent short circuits or burnouts caused by continuous heat during the operation of flip-chip integrated circuits, which would shorten their lifespan, heat sinks are typically used to dissipate heat and cool the circuits. However, conducting single-event experiments requires thinning the substrate, which damages this structure and reduces the circuit's heat dissipation capacity. Furthermore, the tandem accelerator at the China Institute of Atomic Energy operates within a vacuum chamber, where temperature rises become more frequent. Therefore, temperature monitoring of the irradiation circuitry is crucial to prevent single-event latch-up or circuit malfunctions caused by high temperatures, ensuring the accuracy of experimental data.

[0054] During the single-particle test, a temperature detection device is set up to monitor the operating temperature of the flip-chip soldering circuit sample in real time, and a temperature alarm threshold is set. If the operating temperature of the flip-chip soldering circuit sample is greater than or equal to the temperature alarm threshold, the test is stopped immediately and the test is resumed after the operating temperature of the flip-chip soldering circuit sample returns to between [T1, T2], where [T1, T2] is the temperature range of the flip-chip soldering circuit under rated operation.

[0055] S3. After the single-particle experiment is completed, the experimental data is analyzed to determine the validity of the experimental data. If the experimental data is valid, proceed to step S4; otherwise, replace the flip-chip soldering process circuit sample and repeat steps S1 to S3.

[0056] During substrate thinning in flip-chip bonding circuits, surface inhomogeneity may occur due to technological limitations. This can lead to invalidation of single-event test (SET) data. For example, the SET cross-section at high LET values ​​may be lower than that at low LET values. At least three samples are required for SET testing. When analyzing data at the test site, both lateral and longitudinal comparisons of the test data should be performed. The validity of test data is determined by the simultaneous fulfillment of two conditions; otherwise, the test data is considered invalid.

[0057] The first condition is: A horizontal comparison of the single-particle error number of different samples under the same ion irradiation must satisfy the following:

[0058]

[0059] Where, α i This represents the single-particle error number in the i-th experiment;

[0060] n represents the total number of test samples;

[0061] α max Indicates the maximum number of single-particle errors for the test sample;

[0062] α min Represents the minimum single-particle error number for the test sample;

[0063] ε and τ are constants; based on practical engineering experience, ε is 1.5 and τ is 5.

[0064] The second condition is: when comparing the number of single-particle errors (SPOIs) of the same sample under different ion irradiations in a longitudinal manner, the number of SPOIs caused by irradiation with high LET values ​​must be greater than that caused by ion data with low LET values.

[0065] When processing single-particle data, invalid data needs to be marked, and the experiment should be carried out again with heavy ions at the same LET value after changing the test sample.

[0066] S4. Perform thickness measurement and calibration on the flip-chip bonding process circuit samples. Based on the calibration results, select valid flip-chip bonding process circuit samples from the experimental samples. Then, execute steps S5 and S6 to obtain the single-particle effective data corresponding to heavy ions. The single-particle effective data includes the effective LEETe and the single-particle error cross section σ. m ;

[0067] After completing the single-particle experiment, in order to verify the substrate thinning thickness and the uniformity of the incident surface of the flip-chip bonding process circuit, it is necessary to measure and calibrate the thinning thickness of the test sample.

[0068] The method for thickness measurement and calibration of flip-chip bonding process circuit samples is as follows: the flip-chip bonding process circuit sample is longitudinally cut along the diagonal, and the maximum substrate thickness d of the flip-chip bonding process circuit sample is measured. r-max and minimum thickness d r-min Calculate the average thickness d of the substrate thinning in the flip-chip soldering process circuit sample. r-mean If d r-max -d r-min Greater than d r-min If 10% of the samples are invalid, the flip-chip soldering circuit sample is considered invalid; otherwise, the flip-chip soldering circuit sample is considered valid.

[0069] S5. Based on the single-particle test results of the effective flip-chip soldering circuit samples, calculate the single-particle error cross section σ of heavy ions. m ;

[0070] The single-particle error cross section σ of the heavy ion m It is obtained through the following method:

[0071] Single-event error cross section = Total number of single-event errors / Total irradiation flux

[0072] The total number of single-event errors refers to the total number of single-event errors generated in single-event experiments corresponding to heavy ions;

[0073] Total irradiation flux refers to the total irradiation flux of heavy ions in a single-particle experiment corresponding to heavy ions.

[0074] S6. Based on the average thickness d of the substrate thinning of the effective flip-chip bonding process circuit sample... r-mean The effective LET of heavy ions was obtained by simulation using heavy ion LET calculation software (such as SRIM).

[0075] Average thickness d of substrate thinning for effective flip-chip bonding process circuit samples r-mean =(d r-max +d r-min ) / 2.

[0076] A valid LETE is obtained through the following method:

[0077] S6.1 Substitute the surface energy Es of the ions incident on the flip-chip bonding process circuit sample into the heavy ion LET calculation software to obtain the surface range μ of the incident ions;

[0078] S6.2. Subtract the substrate thinning thickness from the surface range μ obtained in S6.1 to obtain the effective range μ-d of the incident ions. r-mean ;

[0079] S6.3. Using heavy ion LET calculation software, the effective range was iteratively calculated to be μ-d. r-mean The energy and LET value of the heavy ion at this time are: the energy at this time is the effective energy Ee, and the corresponding LET is the effective LET.

[0080] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make possible changes and modifications to the technical solutions of the present invention by utilizing the methods and techniques disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention shall fall within the protection scope of the technical solutions of the present invention.

Claims

1. A method for acquiring single-particle effective data in a flip-chip circuit, characterized in that... Includes the following steps: S1. Thin the substrate of the flip-chip soldering circuit sample and record the thinning thickness; S2. Select heavy ions suitable for flip-chip bonding process circuits to conduct single-particle tests on the thinned flip-chip bonding process circuit samples to obtain the single-particle error number of different samples under the same heavy ion irradiation and the single-particle error number of the same sample under different heavy ion irradiation. S3. After the single-particle experiment is completed, the experimental data is analyzed to determine the validity of the experimental data. If the experimental data is valid, proceed to step S4; otherwise, replace the flip-chip soldering process circuit sample and repeat steps S1 to S3. S4. Perform thickness measurement and calibration on the flip-chip bonding process circuit samples. Based on the calibration results, select valid flip-chip bonding process circuit samples from the experimental samples. Then, execute steps S5 and S6 to obtain the single-particle effective data corresponding to heavy ions. The single-particle effective data includes the effective LEETe and the single-particle error cross section σ. m ; S5. Based on the single-particle test results of the effective flip-chip soldering circuit samples, calculate the single-particle error cross section σ of heavy ions. m ; S6. Based on the average thickness d of the substrate thinning of the effective flip-chip bonding process circuit sample... r-mean The effective LET of heavy ions was obtained by simulation using heavy ion LET calculation software.

2. The method for acquiring single-particle effective data of a flip-chip circuit according to claim 1, characterized in that, During the single-particle test, the operating temperature of the flip-chip soldering circuit sample is monitored in real time. If the operating temperature of the flip-chip soldering circuit sample is greater than or equal to the temperature alarm threshold, the test is stopped immediately and the test is resumed after the operating temperature of the flip-chip soldering circuit sample returns to the range of [T1, T2], where [T1, T2] is the temperature range of the flip-chip soldering circuit under rated operation.

3. The method for acquiring single-particle effective data of a flip-chip circuit according to claim 1, characterized in that, The thickness l of the thinned flip-chip bonding process circuit sample satisfies 20μm < l < 40μm, and the uniformity of the substrate surface meets the preset conditions.

4. The method for acquiring single-particle effective data of a flip-chip circuit according to claim 1, characterized in that, The selected heavy ions suitable for flip-chip bonding circuits have a range greater than 40 μm.

5. The method for acquiring single-particle effective data of a flip-chip circuit according to claim 1, characterized in that, The selected heavy ion LET values ​​suitable for flip-chip bonding circuits cover three energy ranges: 0~10MeV·cm 2 / mg、30~40MeV·cm 2 / mg、60~100MeV·cm 2 / mg。 6. The method for acquiring single-particle effective data of a flip-chip circuit according to claim 1, characterized in that, The validity of experimental data is determined by the simultaneous fulfillment of the following two conditions: if the experimental data is valid, it is considered valid; otherwise, the experimental data is considered invalid. The first condition is: the single-particle error number of different samples under the same ion irradiation satisfies: Where, α i This represents the single-particle error number in the i-th experiment; n represents the total number of test samples; α max Indicates the maximum number of single-particle errors for the test sample; α min Represents the minimum single-particle error number for the test sample; ε and τ are constants; The second condition is that the number of single-particle errors (SPOIs) of the same sample under different ion irradiations must satisfy the requirement that the number of SPOIs caused by irradiation with high LET values ​​is greater than that caused by irradiation with low LET values.

7. The method for acquiring single-particle effective data of a flip-chip circuit according to claim 1, characterized in that, The method for thickness measurement and calibration of flip-chip bonding process circuit samples is as follows: the flip-chip bonding process circuit sample is longitudinally cut along the diagonal, and the maximum substrate thickness d of the flip-chip bonding process circuit sample is measured. r-max and minimum thickness d r-min Calculate the average thickness d of the substrate thinning in the flip-chip soldering process circuit sample. r-mean If d r-max -d r-min Greater than d r-min If 10% of the samples are invalid, the flip-chip soldering circuit sample is considered invalid; otherwise, the flip-chip soldering circuit sample is considered valid.

8. The method for acquiring single-particle effective data of a flip-chip circuit according to claim 7, characterized in that, Average thickness d of substrate thinning for effective flip-chip bonding process circuit samples r-mean =(d r-max +d r-min ) / 2.

9. The method for acquiring single-particle effective data of a flip-chip circuit according to claim 1, characterized in that, A valid LETE is obtained through the following method: S6.1 Substitute the surface energy Es of the ions incident on the flip-chip bonding process circuit sample into the heavy ion LET calculation software to obtain the surface range μ of the incident ions; S6.

2. Subtract the substrate thinning thickness from the surface range μ obtained in S6.1 to obtain the effective range μ-d of the incident ions. r-mean ; S6.

3. Using heavy ion LET calculation software, the effective range was iteratively calculated to be μ-d. r-mean The energy and LET value of the heavy ion at this time are: the energy at this time is the effective energy Ee, and the corresponding LET is the effective LET.

10. The method for acquiring single-particle effective data of a flip-chip circuit according to claim 1, characterized in that, The single-particle error cross section σ of the heavy ion m It is obtained through the following method: Single-event error cross section = Total number of single-event errors / Total irradiation flux The total number of single-event errors refers to the total number of single-event errors generated in single-event experiments corresponding to heavy ions; Total irradiation flux refers to the total irradiation flux of heavy ions in a single-particle experiment corresponding to heavy ions.

Images