A method for predicting reliable life of transistors based on β-value degradation

Abstract

The present invention discloses a method for predicting the reliable life of a transistor based on β value degradation. The reliable life prediction of the transistor is based on a failure physics model. Compared with the traditional reliable life prediction method based on mathematical statistics, it can start from the microscopic failure physics of the transistor and more accurately reflect the degradation of the sensitive parameters of the transistor, thereby improving the accuracy of the reliable life prediction of the transistor; the reliable life prediction of the transistor takes into account the initial distribution of the sensitive parameter β, which is more in line with the actual situation of the external characteristic parameters of the transistor, and based on this, gives a life prediction under a specified reliability, which is more instructive for the engineering application of the transistor.

Description

Technical Field

[0001] The present invention belongs to the field of component reliability and relates to transistor reliability prediction, and in particular to a transistor reliable life prediction method based on β value degradation, which is suitable for transistor reliable life prediction based on parameter degradation, combined with a failure physical model and considering parameter consistency. Background Art

[0002] Military semiconductor discrete devices are the core and foundation of modern military electronic technology. Due to their large usage and wide application, they are crucial in the reliability evaluation of single machines and are related to the lifeline of national security and national defense construction.

[0003] Reliability prediction methods based on the physics of failure represent a new phase in the scientific development of reliability analysis methods, evolving from mathematical statistics to a new approach based on physical and chemical analysis. They analyze the causes of failure from the very nature of electronic product failure. Therefore, transistor lifetime prediction methods based on sensitive parameter degradation, combined with failure physics models, and considering parameter consistency can provide a scientific basis and theoretical support for the production of high-reliability electronic products. Summary of the Invention

[0004] The technology of the present invention solves the problem of overcoming the shortcomings of the traditional reliable life prediction method based on mathematical statistics, providing a transistor reliable life prediction method based on β value degradation, and improving the accuracy of transistor reliable life prediction.

[0005] A method for predicting the reliable life of a transistor based on β value degradation, comprising:

[0006] Step 1: randomly selecting multiple transistors from the same batch of transistors as test subjects for a high-temperature reverse bias test; randomly selecting multiple transistors from the same batch of transistors as test subjects for a steady-state power test; and randomly selecting multiple transistors from the same batch of transistors as test subjects for an initial β value distribution test.

[0007] Step 2: setting at least four different test conditions for high-temperature reverse bias tests based on the inherent characteristics and application characteristics of the transistor; and setting at least four different test conditions for steady-state power tests based on the inherent characteristics and application characteristics of the transistor.

[0008] Step 3: Carry out a high-temperature reverse bias test of the transistor and a steady-state power test of the transistor according to the set test conditions;

[0009] Step 4: During the test, monitor the change of the transistor's β value;

[0010] Step 5: After the test is completed, a curve of the change of the β value of the transistor over time under different high-temperature reverse bias test conditions and a curve of the change of the β value of the transistor over time under different steady-state power test conditions are plotted;

[0011] Step 6: Determine the failure physical model form of the transistor PN junction characteristic degradation under high temperature reverse bias test conditions, and the failure physical model form of the transistor hot carrier injection under steady-state power test conditions;

[0012] Among them, the failure physical model of PN junction characteristic degradation and the failure physical model of hot carrier injection are the degradation forms of the transistor β value over time; based on the curves of the change of β value over time under multiple sets of high-temperature reverse bias test conditions, the coefficients in the failure physical model of PN junction characteristic degradation are solved, and the specific form of the PN junction characteristic degradation failure physical model is determined; based on the curves of the change of β value over time under multiple sets of steady-state power test conditions, the coefficients in the failure physical model of hot carrier injection are solved, and the specific form of the hot carrier injection failure physical model is determined, which is specifically:

[0013] PN junction characteristic degradation model:

[0014] Hot carrier injection failure physical model:

[0015] In formula (1), when calculating the coefficients in the failure physical model based on the curve of the change of β value with time under different test conditions, the following physical quantities are known: current gain β, initial value of current gain β0, collector-base voltage VCB, Boltzmann constant k, temperature T, time t; the variables to be solved are: coefficient A, coefficient m, coefficient a, and activation energy E a ;

[0016] In formula (2), when calculating the coefficients in the failure physical model based on the curve of the change of β value with time under different test conditions, the following physical quantities are known: current gain β, initial value of current gain β0, collector current IC, Boltzmann constant k, temperature T, time t, and the variables to be solved are: coefficient A, coefficient n, coefficient a, and activation energy E a ;

[0017] Step 7, testing the β values ​​of multiple transistors and providing an initial distribution of β values;

[0018] Step 8: Under the specified transistor operating conditions, calculate the degradation trajectory of the initial distribution of β value under the failure physical model of PN junction characteristic degradation and the failure physical model of hot carrier injection, and combine it with the set device failure threshold β th, give the reliable life of the transistor based on the degradation of PN junction characteristics and the reliable life based on hot carrier injection under specified reliability, and select the shorter one, that is, the reliable life of the transistor under the above conditions.

[0019] Preferably, the high temperature reverse bias test conditions include the collector base voltage V CB and ambient temperature T A ; V CB 、T A The condition combinations are 4 groups, 3 of which are fixed T A， Change V CB , another group of T A With the previous 3 groups of T A Different, V CB With 1 V in the previous 3 groups CB same.

[0020] Preferably, the steady-state power test conditions include the collector current I C , collector-emitter voltage V CE , shell temperature T C ;I C 、V CE 、T C The condition combinations are 4 groups, 3 of which are fixed T C , change I C 、V CE , another group of T C With the previous 3 groups of T C Different, I C 、V CE With 1 I in the previous 3 groups C 、V CE same.

[0021] Preferably, 10 transistors are randomly selected from the same batch of transistors as test subjects for the high-temperature reverse bias test; 10 transistors are randomly selected from the same batch of transistors as test subjects for the steady-state power test; and 100 transistors are randomly selected from the same batch of transistors as test subjects for the initial β value distribution test.

[0022] Preferably, the test time for conducting the high temperature reverse bias test and the steady-state power test is 1000 hours.

[0023] Preferably, in step 5, the curve of the change of β value over time provides the original data points, the best fitting curve and the 95% confidence interval.

[0024] Preferably, in step 4, when monitoring the change of the β value of the transistor, the monitoring time is 0 hours, 100 hours, 200 hours, 300 hours, 400 hours, 500 hours, 600 hours, 700 hours, 800 hours, 900 hours, and 1000 hours.

[0025] The present invention has the following beneficial effects:

[0026] (1) In the present invention, the reliable life prediction of transistors is based on a failure physics model. Compared with the traditional reliable life prediction method based on mathematical statistics, it can start from the microscopic failure physics of transistors and more accurately reflect the degradation of sensitive parameters of transistors, thereby improving the accuracy of transistor reliable life prediction.

[0027] (2) In the present invention, the reliable life prediction of the transistor takes into account the initial distribution of the sensitive parameter β, which is more in line with the actual situation of the transistor's external characteristic parameters. Based on this, a life prediction under a specified reliability is given, which is more instructive for the engineering application of the transistor. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a flowchart of the steps of a method for predicting the reliable life of a transistor based on β value degradation in an embodiment of the present invention;

[0029] FIG2( a ) and FIG2( b ) are respectively the structure and appearance of a 3CK2907 transistor according to an embodiment of the present invention;

[0030] Figure 3 β value variation curves over time under four sets of high-temperature reverse bias test conditions in the embodiment of the present invention;

[0031] Figure 4 1 is a curve showing the change of β value over time under four groups of steady-state power test conditions in the embodiment of the present invention. DETAILED DESCRIPTION

[0032] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0033] Reference Figure 1 , shows a flow chart of a method for predicting the reliable life of a transistor based on β value degradation in an embodiment of the present invention. In this embodiment, the method for predicting the reliable life of a transistor based on β value degradation includes:

[0034] Step 1: randomly select 10 transistors from the same batch of transistors as test objects for the high-temperature reverse bias test; randomly select 10 transistors from the same batch of transistors as test objects for the steady-state power test; and randomly select 100 transistors from the same batch of transistors as test objects for the initial β value distribution test.

[0035] In the embodiment of the present invention, the transistor model specification is 3CK2907, the wafer material is silicon, the cap material is nickel alloy, and the A3-01B package is used. The device collector junction rating at room temperature is V CBO=-60V, rated power is 500mW, the device structure and appearance are shown in Figure 2.

[0036] Step 2: Set the test conditions for the high-temperature reverse bias test based on the inherent characteristics and application characteristics of the transistor; wherein the high-temperature reverse bias test conditions include the collector-base voltage V CB and ambient temperature T A ; V CB 、T A The condition combinations are 4 groups, 3 of which are fixed T A， Change V CB , another group of T A With the previous 3 groups of T A Different, V CB With 1 V in the previous 3 groups CB Same; according to the inherent characteristics and application characteristics of the transistor, the test conditions of the steady-state power test are set; among them, the steady-state power test conditions include the collector current I C , collector-emitter voltage V CE , shell temperature T C ;I C 、V CE 、T C The condition combinations are 4 groups, 3 of which are fixed T C , change I C 、V CE , another group of T C With the previous 3 groups of T C Different, I C 、V CE With 1 I in the previous 3 groups C 、V CE same.

[0037] In the embodiment of the present invention, the high temperature reverse bias test conditions are: V CB =-48V, T A =448K, V CB =-48V, T A =423K, V CB =-38V, T A =423K, V CB =-30V, T A =423K; steady-state power test conditions are: I C =-50mA, V CE =-10V, T C =346K, I C =-20mA, V CE =-25V, T C =346K, I C =-10mA, V CE =-45V, TC =346K, I C =-20mA, V CE =-25V, T C =331K.

[0038] Step 3: Carry out high-temperature reverse bias tests on 10 transistors and steady-state power tests on 10 transistors according to the set test conditions; the test time is 1000 hours.

[0039] In the embodiment of the present invention, the test time is 1000 hours.

[0040] Step 4: During the test, monitor the change in the β value of the transistor; wherein the monitoring time is 0 hour, 100 hours, 200 hours, 300 hours, 400 hours, 500 hours, 600 hours, 700 hours, 800 hours, 900 hours, and 1000 hours.

[0041] In the embodiment of the present invention, the monitoring time is 0 hours, 100 hours, 200 hours, 300 hours, 400 hours, 500 hours, 600 hours, 700 hours, 800 hours, 900 hours, and 1000 hours.

[0042] Step 5. After the test is completed, a curve of the change of the β value of the transistor over time under different high-temperature reverse bias test conditions and a curve of the change of the β value of the transistor over time under different steady-state power test conditions are plotted; among them, four groups of curves of the change of the β value over time under high-temperature reverse bias test conditions are plotted respectively; four groups of curves of the change of the β value over time under steady-state power test conditions are plotted respectively; the curve of the change of the β value over time should provide the original data points, the best fit curve and the 95% confidence interval.

[0043] In the embodiment of the present invention, the curves of the change of β value over time under the four groups of high temperature reverse bias test conditions are as follows: Figure 3 As shown in the figure, the curves of β value changing with time under the four groups of steady-state power test conditions are as follows: Figure 4 shown.

[0044] Step 6: Determine the failure physical model form of the transistor PN junction characteristic degradation under high-temperature reverse bias test conditions, and the failure physical model form of the transistor hot carrier injection under steady-state power test conditions; wherein, the failure physical model of PN junction characteristic degradation and the failure physical model of hot carrier injection are the degradation forms of the transistor β value over time; based on the 4 sets of β value change curves over time under high-temperature reverse bias test conditions, solve the coefficients in the failure physical model of PN junction characteristic degradation to determine the specific form of the PN junction characteristic degradation failure physical model; based on the 4 sets of β value change curves over time under steady-state power test conditions, solve the coefficients in the failure physical model of hot carrier injection to determine the specific form of the hot carrier injection failure physical model.

[0045] In the embodiment of the present invention, based on the research and analysis of existing models, the model forms determined are:

[0046] PN junction characteristic degradation model:

[0047] Hot carrier injection failure physical model:

[0048] In formula (1), when calculating the coefficients in the failure physical model based on the curve of the change of β value with time under different test conditions, the following physical quantities can be considered as known quantities: current gain β, initial value of current gain β0, collector-base voltage VCB, Boltzmann constant k, temperature T, time t; the variables to be solved are: coefficient A, coefficient m, coefficient a, and activation energy E a .

[0049] In formula (2), when calculating the coefficients in the failure physical model based on the curve of the change of β value with time under different test conditions, the following physical quantities can be considered as known quantities: current gain β, initial value of current gain β0, collector current IC, Boltzmann constant k, temperature T, time t, and the following variables are solved: coefficient A, coefficient n, coefficient a, and activation energy E a .

[0050] Step 7: Test the β values ​​of 100 transistors and give an initial distribution of β values.

[0051] In the embodiment of the present invention, the initial distribution of β values ​​is a normal distribution.

[0052] Step 8: Under the specified transistor operating conditions, calculate the degradation trajectory of the initial distribution of β value under the failure physical model of PN junction characteristic degradation and the failure physical model of hot carrier injection, and combine the device failure threshold (β th ), give the reliable life of the transistor based on the degradation of PN junction characteristics and the reliable life based on hot carrier injection under specified reliability, and select the shorter one, that is, the reliable life of the transistor under the above conditions.

[0053] In the embodiment of the present invention, the transistor operating conditions are set as follows: high temperature reverse bias: V CB =-48V, T A =448K, steady-state power: I C =-50mA, V CE =-10V, T C =346K, combined with the device failure threshold (β th) Calculations show that, at a reliability of 95%, the transistor's reliable lifetime due to PN junction degradation is 28,524 hours, and its reliable lifetime due to hot carrier injection is 17,358 hours. This confirms that hot carrier injection is a sensitive failure mechanism for the 3CK2907 transistor. Under these conditions, the device lifetime can reach 17,358 hours.

[0054] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for predicting the reliable life of a transistor based on β value degradation, characterized in that: include: Step 1, randomly selecting multiple transistors from the same batch of transistors as test objects for a high-temperature reverse bias test; Randomly select multiple transistors from the same batch of transistors as test objects for a steady-state power test; randomly select multiple transistors from the same batch of transistors as test objects for an initial β value distribution test; Step 2: setting at least four different test conditions for high-temperature reverse bias tests based on the inherent characteristics and application characteristics of the transistor; and setting at least four different test conditions for steady-state power tests based on the inherent characteristics and application characteristics of the transistor. Step 3: Carry out a high-temperature reverse bias test of the transistor and a steady-state power test of the transistor according to the set test conditions; Step 4: During the test, monitor the change of the transistor's β value; Step 5: After the test is completed, a curve of the change of the β value of the transistor over time under different high-temperature reverse bias test conditions and a curve of the change of the β value of the transistor over time under different steady-state power test conditions are plotted; Step 6: Determine the failure physical model form of the transistor PN junction characteristic degradation under high temperature reverse bias test conditions, and the failure physical model form of the transistor hot carrier injection under steady-state power test conditions; Among them, the failure physical model of PN junction characteristic degradation and the failure physical model of hot carrier injection are the degradation forms of the transistor β value over time; based on the curves of the change of β value over time under multiple sets of high-temperature reverse bias test conditions, the coefficients in the failure physical model of PN junction characteristic degradation are solved, and the specific form of the PN junction characteristic degradation failure physical model is determined; based on the curves of the change of β value over time under multiple sets of steady-state power test conditions, the coefficients in the failure physical model of hot carrier injection are solved, and the specific form of the hot carrier injection failure physical model is determined, which is specifically: PN junction characteristic degradation model: Hot carrier injection failure physical model: In formula (1), when calculating the coefficients in the failure physical model based on the curve of the change of β value with time under different test conditions, the following physical quantities are known: current gain β, initial value of current gain β0, collector-base voltage VCB, Boltzmann constant k, temperature T, time t; the variables to be solved are: coefficient A, coefficient m, coefficient a, and activation energy E a ; In formula (2), when calculating the coefficients in the failure physical model based on the curve of the change of β value with time under different test conditions, the following physical quantities are known: current gain β, initial value of current gain β0, collector current IC, Boltzmann constant k, temperature T, time t, and the variables to be solved are: coefficient A, coefficient n, coefficient a, and activation energy E a ; Step 7, testing the β values ​​of multiple transistors and providing an initial distribution of β values; Step 8: Under the specified transistor operating conditions, calculate the degradation trajectory of the initial distribution of β value under the failure physical model of PN junction characteristic degradation and the failure physical model of hot carrier injection, and combine it with the set device failure threshold β th , give the reliable life of the transistor based on the degradation of PN junction characteristics and the reliable life based on hot carrier injection under specified reliability, and select the shorter one, that is, the reliable life of the transistor under the above conditions.

2. The method for predicting the reliable life of a transistor based on β value degradation according to claim 1, wherein: High temperature reverse bias test conditions include collector base voltage V CB and ambient temperature T A ; V CB 、T A The condition combinations are 4 groups, 3 of which are fixed T A， Change V CB , another group of T A With the previous 3 groups of T A Different, V CB With 1 V in the previous 3 groups CB same.

3. A method for predicting the reliable life of a transistor based on β value degradation according to claim 1 or 2, characterized in that: Steady-state power test conditions include collector current I C , collector-emitter voltage V CE , shell temperature T C ;I C 、V CE 、T C The condition combinations are 4 groups, 3 of which are fixed T C , change I C 、V CE , another group of T C With the previous 3 groups of T C Different, I C 、V CE With 1 I in the previous 3 groups C 、V CE same.

4. The method for predicting the reliable life of a transistor based on β value degradation according to claim 1, wherein: 10 transistors are randomly selected from the same batch of transistors as test objects for the high-temperature reverse bias test; 10 transistors are randomly selected from the same batch of transistors as test objects for the steady-state power test; and 100 transistors are randomly selected from the same batch of transistors as test objects for the initial β value distribution test.

5. The method for predicting the reliable life of a transistor based on β value degradation according to claim 1, wherein: The test time for high temperature reverse bias test and steady state power test is 1000 hours.

6. The method for predicting the reliable life of a transistor based on β value degradation according to claim 1, wherein: In step 5, the curve of the change of β value over time provides the original data points, the best fitting curve and the 95% confidence interval.

7. The method for predicting the reliable life of a transistor based on β value degradation according to claim 1, wherein: In step 4, when monitoring the change in the β value of the transistor, the monitoring time is 0 hour, 100 hours, 200 hours, 300 hours, 400 hours, 500 hours, 600 hours, 700 hours, 800 hours, 900 hours, and 1000 hours.

Images