Test Method

Abstract

To perform cooling or the like of a semiconductor device in a screening test at a low temperature or the like.SOLUTION: There is provided a testing method comprising a preparation stage for preparing multiple setting groups including multiple setting semiconductor devices each and for allocating different inspection voltages to the respective setting groups, a first testing stage for applying an allocated inspection voltage to the setting semiconductor devices and testing them at a first temperature, a second testing stage for testing a setting semiconductor device determined to be conforming in the first testing stage at a second temperature and for detecting a breakdown voltage at which the setting semiconductor device is broken, a relationship acquisition stage for acquiring a relationship between an inspection voltage and a breakdown voltage, and an application voltage setting stage for setting an application voltage when testing a tested semiconductor device at a first temperature on the basis of the relationship acquired in the relationship acquisition stage.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method for testing a semiconductor device. [Background technology]

[0002] BACKGROUND ART Conventionally, a method for screening semiconductor devices such as transistors is known (see, for example, Patent Document 1). Patent Document 1: Japanese Patent Application Laid-Open No. 2007-327918 Summary of the Invention [Problem to be solved by the invention]

[0003] For screening tests at temperatures other than room temperature, the temperature of the semiconductor device must be adjusted. [Means for solving the problem]

[0004] To solve the above problem, one aspect of the present invention provides a semiconductor device testing method. The testing method may include a preparation step of preparing multiple setting groups, each including multiple setting semiconductor devices, and assigning a different test voltage to each setting group. The testing method may include a first testing step of applying the assigned test voltage to the multiple setting semiconductor devices included in each setting group and testing them at a first temperature. The testing method may include a second testing step of testing setting semiconductor devices determined to be non-defective in the first testing step at a second temperature different from the first temperature to detect a breakdown voltage at which the setting semiconductor devices are broken. The testing method may include a relationship acquisition step of acquiring a relationship between the test voltage and the breakdown voltage for each setting group. The testing method may include an application voltage setting step of setting an application voltage to be applied when testing the semiconductor device under test at the first temperature based on the relationship acquired in the relationship acquisition step.

[0005] The second temperature may be lower than the first temperature.

[0006] In the step of setting the applied voltage, the applied voltage may be set based on the relationship between the breakdown voltage that the semiconductor device under test should have at the second temperature and the relationship between the inspection voltage and the breakdown voltage.

[0007] In a second test stage, the minimum breakdown voltage in each configuration group may be detected.

[0008] In the second test stage, one or more setting semiconductor devices may be tested in sequence. In the second test stage, in testing each setting semiconductor device, the voltage applied to the setting semiconductor device may be gradually increased to detect the breakdown voltage of the setting semiconductor device. In the second test stage, when the voltage applied to each setting semiconductor device reaches the breakdown voltage of the setting semiconductor device already tested, the next setting semiconductor device may be tested.

[0009] In the second test stage, the distribution of breakdown voltages in each of the configuration groups may be detected.

[0010] The test method may include a gradient acquisition step of acquiring a gradient of a waveform of a test voltage when the test voltage is applied to the setting semiconductor device, the gradient of the waveform of the test voltage being the smallest at which the setting semiconductor device is destroyed. In the first test step and the second test step, the setting semiconductor device may be tested using the gradient of the waveform acquired in the gradient acquisition step.

[0011] A first test stage, a second test stage, and a relationship acquisition stage may be performed on a first group including two or more setting groups to acquire an approximate value of an applied voltage when testing the semiconductor device under test at a first temperature. A test voltage according to the approximate value of the applied voltage may be assigned to a second group including two or more setting groups different from the first group, and the first test stage, a second test stage, and a relationship acquisition stage may be performed. In the applied voltage setting stage, the applied voltage to the semiconductor device under test may be set based on the relationship acquired in the relationship acquisition stage for the second group.

[0012] The above summary of the invention does not list all of the necessary features of the present invention, and subcombinations of these features may also constitute inventions. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a flowchart outlining a semiconductor device testing method according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram illustrating an example of a preparation stage S102. [Figure 3] 10 shows an example of distribution of withstand voltages of the setting semiconductor devices 210 in the setting group 200-1. [Figure 4] FIG. 10 is a diagram illustrating the second test stage S106. [Figure 5] FIG. 10 is a diagram showing an example of a method for obtaining the minimum breakdown voltage Vb1 in the second test stage S106. [Figure 6] 10 is a diagram for explaining an outline of a relationship acquisition step S108 and an applied voltage setting step S110. FIG. [Figure 7] FIG. 10 is a diagram illustrating another embodiment of the test method. [Figure 8] FIG. 10 is a diagram illustrating processing for a second group. [Figure 9] 1 is a diagram illustrating an overview of a test device 100 that tests a setting semiconductor device 210. FIG. [Figure 10] FIG. 10 is a diagram showing an example of measurement of leakage current. [Figure 11] 10 is a diagram showing a case where the setting semiconductor device 210 is broken during leakage current measurement. FIG. [Figure 12] FIG. 10 is a diagram showing the relationship between the rising slope (dv / dt) of the applied voltage and the breakdown rate of the semiconductor device. [Figure 13] 10 is a flowchart showing another example of a testing method. DETAILED DESCRIPTION OF THE INVENTION

[0014] The present invention will be described below through embodiments of the invention, but the following embodiments do not limit the scope of the invention according to the claims. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0015] FIG. 1 is a flowchart showing an outline of a semiconductor device testing method according to one embodiment of the present invention. For example, the semiconductor device may be, but is not limited to, an IGBT (Insulated Gate Bipolar Transistor). In this example, the testing method screens semiconductor devices that do not have a predetermined breakdown voltage. The breakdown voltage at a predetermined temperature, such as -40°C, may be set as a semiconductor device specification. In this case, it may be possible to test the semiconductor device while it is cooled to -40°C, but this would reduce testing efficiency.

[0016] In the test method of this example, the relationship between the breakdown voltage of a semiconductor device at a first temperature (e.g., room temperature such as 25°C) and the breakdown voltage of the semiconductor device at a second temperature (e.g., -40°C) is estimated in advance. From this relationship, the breakdown voltage at the first temperature, which corresponds to the required value of the breakdown voltage that the semiconductor device should have at the second temperature, is estimated. By testing the semiconductor device at the first temperature using conditions according to the estimated breakdown voltage, screening at the second temperature is performed in a pseudo manner at the first temperature. This can improve test efficiency.

[0017] As shown in FIG. 1 , the test method of this example includes a preparation step S102, a first test step S104, a second test step S106, a relationship acquisition step S108, an applied voltage setting step S110, and a DUT test step S112. In the preparation step S102 through the applied voltage setting step S110, a setting semiconductor device is used to estimate a withstand voltage at a first temperature, which corresponds to the required withstand voltage value that the semiconductor device should have at a second temperature. In the DUT test step S112, a semiconductor device under test (DUT) to be tested is tested under conditions corresponding to the estimated withstand voltage. The setting semiconductor device and the semiconductor device under test are different devices. It is preferable that the setting semiconductor device and the semiconductor device under test are devices manufactured with the same design for part or all of their device structures. Note that the steps S102 through S110 do not need to be performed each time S112 is performed. The conditions acquired in the steps S102 through S110 can be continuously used in the step S112 for a certain period of time.

[0018] 2 is a diagram illustrating an example of the preparation stage S102. In the preparation stage S102, a plurality of setting groups 200 are prepared, each including a plurality of setting semiconductor devices 210. A different test voltage Va is assigned to each setting group 200. In the example of FIG. 2, test voltages Va1, Va2, Va3, etc. are assigned to the setting groups 200-1, 200-2, 200-3, etc.

[0019] FIG. 3 shows an example of the breakdown voltage distribution of the setting semiconductor devices 210 in the setting group 200-1. The horizontal axis in FIG. 3 represents the applied voltage to the setting semiconductor devices 210, and the vertical axis represents the number of setting semiconductor devices 210 that have a breakdown voltage corresponding to each applied voltage. In this specification, the applied voltage refers to the voltage applied between the main terminals of a semiconductor device. Applying a voltage to a semiconductor device refers to applying a voltage between the main terminals of a semiconductor device. The main terminals are terminals through which a main current flows when the semiconductor device is in the on state. In the case of an IGBT, the main terminals are the collector terminal and the emitter terminal. Breakdown of a semiconductor device refers to an irreversible change in the characteristics of the semiconductor device before and after the breakdown. For example, if a voltage equal to or greater than the breakdown voltage is applied to the insulating film or semiconductor region of a semiconductor device, the structure of the insulating film or semiconductor changes irreversibly, resulting in changes in characteristics such as insulation.

[0020] In the first test stage S104, an assigned test voltage Va is applied to the multiple setting semiconductor devices 210 included in each setting group 200, and the setting semiconductor devices 210 are tested at a first temperature T1. In the first test stage S104, the test voltage Va may be applied to the setting semiconductor device 210 with the gate terminal and the emitter terminal of the setting semiconductor device 210 shorted. In the first test stage S104, the setting semiconductor devices 210 that are destroyed by applying the test voltage Va are separated from the setting semiconductor devices 210 that are not destroyed and are determined to be non-defective.

[0021] 3 shows an example of test results for the setting group 200-1. A test voltage Va1 is applied to the setting semiconductor devices 210 in the setting group 200-1 at a temperature T1. Similarly, a test voltage Van is applied to the setting semiconductor devices 210 in the setting group 200-n at a temperature T1. The temperature T1 is, for example, 25° C., but is not limited to this.

[0022] 3, among the setting semiconductor devices 210 in the setting group 200-1, those with a withstand voltage lower than the test voltage Va1 are destroyed, and those with a withstand voltage equal to or higher than the test voltage Va1 are not destroyed and are determined to be good products. In the first test stage S104, the setting semiconductor devices 210 are sorted into a defective product group 230 that are destroyed and a good product group 220 that are not destroyed.

[0023] FIG. 4 is a diagram illustrating the second test step S106. In the second test step S106, the setting semiconductor devices 210 in the non-defective group 220 determined to be non-defective in the first test step S104 are tested at a second temperature T2 different from the first temperature T1. The second temperature T2 may be a temperature used to define the specification or required value of the breakdown voltage of the semiconductor device. The second temperature T2 may be a temperature lower than the first temperature T1. The second temperature T2 is, for example, −40° C., but is not limited thereto. The breakdown voltage of a semiconductor device may decrease as the temperature decreases. Therefore, even if a predetermined breakdown voltage is met at the first temperature T1, the breakdown voltage may not be met at the second temperature T2. In the test method of this example, the first test step S104 and the second test step S106 are used to previously obtain a correlation between the breakdown voltage at the second temperature T2 and the breakdown voltage at the first temperature T1. Then, by converting the required value of the withstand voltage at the second temperature T2 into the withstand voltage at the first temperature T1, it is possible to carry out a withstand voltage test at the second temperature T2 in a pseudo manner at the first temperature T1.

[0024] In the second test stage S106, a breakdown voltage Vb at which the setting semiconductor devices 210 in the non-defective group 220 are broken down is detected for each setting group 200. In the example of FIG. 4, a breakdown voltage Vb1 is detected for the setting group 200-1. Similarly, a breakdown voltage Vbn is detected for the setting group 200-n. In the second test stage S106, the voltage applied to the setting semiconductor devices 210 may be gradually increased to detect the breakdown voltage at which each setting semiconductor device 210 is broken down. In the second test stage S106, it is preferable to detect the minimum breakdown voltage Vb1 among the breakdown voltages Vb1 of the setting semiconductor devices 210 in the non-defective group 220.

[0025] The setting semiconductor device 210 exhibiting the minimum breakdown voltage Vb1 is considered to be the setting semiconductor device 210 with the smallest breakdown voltage among the non-defective device group 220. The relationship between the breakdown voltages of the multiple setting semiconductor devices 210 does not change even when the test temperature changes. In other words, if the breakdown voltage of the first setting semiconductor device 210-1 is higher than that of the second setting semiconductor device 210-2 at the first temperature T1, the breakdown voltage of the first setting semiconductor device 210-1 will also be higher than that of the second setting semiconductor device 210-2 at the second temperature T2. Therefore, the setting semiconductor device 210 that is destroyed by the minimum breakdown voltage Vb1 at the second temperature T2 is considered to have a breakdown voltage at the first temperature T1 that is approximately equal to or slightly higher than the test voltage Va1. Therefore, by obtaining the test voltage Va1 at the first temperature T1 and the breakdown voltage Vb1 at the second temperature T2, the relationship between the breakdown voltage at the first temperature T1 and the breakdown voltage at the second temperature T2 can be estimated.

[0026] Note that the breakdown voltage Vb1 acquired in the second test stage S106 is not limited to the minimum breakdown voltage Vb1. In the second test stage S106, any breakdown voltage Vb1 that can estimate the relationship between the breakdown voltage at the first temperature T1 and the breakdown voltage at the second temperature T2 may be acquired. In the second test stage S106, the second smallest breakdown voltage Vb1 may be acquired, the mth smallest breakdown voltage Vb1 may be acquired, the average value of the breakdown voltages Vb1 of the non-defective group 220 may be acquired, or a value obtained by other processing may be acquired. When acquiring the minimum breakdown voltage Vb1, the minimum breakdown voltage Vb1 may be detected after measuring the breakdown voltages Vb1 of all the setting semiconductor devices 210 in the non-defective group 220, or the minimum breakdown voltage Vb1 may be acquired by other methods.

[0027] 5 is a diagram showing an example of a method for obtaining the minimum breakdown voltage Vb1 in the second test stage S106. In the second test stage S106 of this example, one or more setting semiconductor devices 210 included in the non-defective product group 220 are tested in sequence. In addition, in testing each setting semiconductor device 210, the voltage applied to the setting semiconductor device 210 is gradually increased to detect the breakdown voltage of the setting semiconductor device 210. In addition, when the voltage applied to each setting semiconductor device 210 reaches the minimum breakdown voltage Vb1 of the setting semiconductor device 210 that has already been tested, the next setting semiconductor device 210 is tested even if that setting semiconductor device 210 has not been destroyed.

[0028] Specifically, first, in S502, one of the setting semiconductor devices 210 included in the non-defective group 220 is selected. Next, in S504, the voltage applied to the setting semiconductor device 210 is increased. In S504, the applied voltage may be increased by a predetermined unit increment. Also, in S504, an initial value of the applied voltage may be set. In S504, the applied voltage may be gradually increased from the initial value.

[0029] Next, in S506, it is determined whether the setting semiconductor device 210 has been destroyed. If the setting semiconductor device 210 has been destroyed, the applied voltage at that time is saved as the minimum breakdown voltage Vb1 (S508). If the setting semiconductor device S210 has not been destroyed, it is determined whether the applied voltage has reached the saved minimum breakdown voltage Vb1 (S510). If the applied voltage has not reached the saved minimum breakdown voltage Vb1, the process from S504 is repeated. Note that if the first setting semiconductor device 210 in the non-defective group 220 is being tested, the minimum breakdown voltage Vb1 has not yet been saved, so the processes of S504 and S506 are repeated until the setting semiconductor device 210 is destroyed. If the applied voltage has reached the saved minimum breakdown voltage Vb1, it is determined that there is no longer any possibility of updating the minimum breakdown voltage Vb1 of the setting semiconductor device 210, and the test for the setting semiconductor device 210 is terminated.

[0030] Then, it is determined whether all the setting semiconductor devices 210 have been tested (S512), and if there are any setting semiconductor devices 210 that have not yet been tested, the next setting semiconductor device 210 is selected in S502, and the process is repeated from S502. If testing of all the setting semiconductor devices 210 has been completed, the second test stage S106 is completed.

[0031] 5 is performed on the non-defective group 220 of each setting group 200. This makes it possible to detect the minimum breakdown voltage Vb1 in each setting group 200 without detecting the breakdown voltages of all setting semiconductor devices 210.

[0032] FIG. 6 is a diagram illustrating an overview of the relationship acquisition step S108 and the applied voltage setting step S110. In the relationship acquisition step S108, the relationship between the test voltage Van and the breakdown voltage Vbn for each setting group 200-n is acquired. In the relationship acquisition step S108 of this example, the measurement points (Van, Van) for each setting group 200-n are plotted on a graph with the test voltage Va on the horizontal axis and the breakdown voltage Vb on the vertical axis. In FIG. 6, each measurement point is indicated by a circle. In the relationship acquisition step S108, an approximate straight line 300 that approximates each measurement point may be acquired. The approximate straight line 300 may be calculated using a computer, for example, using a method such as the least squares method.

[0033] An approximate straight line 300 shows the relationship between the test voltage Va at the first temperature T1 and the breakdown voltage Vb at the second temperature T2. Using the approximate straight line 300, it is possible to estimate which test voltage Va at the first temperature T1 corresponds to the breakdown voltage Vb at the second temperature T2.

[0034] In the applied voltage setting step S110, the applied voltage when the semiconductor device under test is tested at the first temperature T1 is set based on the relationship (e.g., the approximate straight line 300) acquired in the relationship acquisition step S108. For example, in the applied voltage setting step S110, the applied voltage is set based on the breakdown voltage that the semiconductor device under test should have at the second temperature T2 and the approximate straight line 300. For example, as shown in Fig. 6, when testing whether the semiconductor device under test has a breakdown voltage of Vc at the second temperature T2, the test voltage Vd corresponding to the breakdown voltage Vc is extracted from the approximate straight line 300 and set as the applied voltage at the first temperature T1.

[0035] In the DUT testing stage S112, the semiconductor device under test is tested using the applied voltage Vd at the first temperature T1, thereby screening out semiconductor devices under test that do not have the withstand voltage Vc at the second temperature T2. In the DUT testing stage S112, the applied voltage Vd may be applied to the semiconductor device under test at the first temperature T1 to exclude semiconductor devices under test that are destroyed.

[0036] In the second test stage S106, the applied voltage at which the setting semiconductor device 210 was destroyed may be used as the breakdown voltage, or the voltage applied immediately before the breakdown may be used as the breakdown voltage. The voltage applied immediately before the breakdown refers to the applied voltage in the step immediately before the step at which the setting semiconductor device 210 was actually destroyed, when the applied voltage is gradually increased step by step, as in S504 of FIG. 5 . The setting semiconductor device 210 tested in the second test stage S106 is the device that was not destroyed by the test voltage Va in the first test stage S104. Therefore, in the second test stage S106, the maximum voltage at which the setting semiconductor device 210 is not destroyed is set as the breakdown voltage Vb, thereby enabling accurate setting of the relationship between the test voltage Va and the breakdown voltage Vb. Furthermore, by setting the breakdown voltage Vb to a voltage lower than the voltage at which the setting semiconductor device 210 was actually destroyed in the second test stage S106, a margin is provided in screening, preventing semiconductor devices under test with low breakdown voltages from being shipped.

[0037] Furthermore, in the second test step S106, the distribution of breakdown voltages of the setting semiconductor devices 210 may be detected in each setting group 200, as shown by the solid line in FIG. 4. As shown in FIG. 4, this distribution may be approximated, for example, by a portion of a Gaussian distribution. In the second test step S106, the minimum value of the breakdown voltage Vb at the second temperature T2 may be determined based on the obtained distribution. For example, in the second test step S106, if the actual number of breakdowns of the setting semiconductor devices 210 at the extracted minimum breakdown voltage Vb1 deviates from the distribution approximated by the Gaussian distribution, this breakdown voltage need not be adopted as the minimum value. In the second test step S106, the next smallest breakdown voltage Vb1 may be adopted as the minimum value. This allows the minimum breakdown voltage Vb1 to be set while excluding test results of specific individuals.

[0038] FIG. 7 is a diagram illustrating another embodiment of the testing method. In this example, the first testing step S104, the second testing step S106, and the relationship acquisition step S108 are performed on a first group including two or more setting groups 200 to obtain an approximate value Vd1 of the applied voltage when the semiconductor device under test is tested at a first temperature T1. That is, the same procedures as those described with reference to FIGS. 1 to 6 are performed to obtain an approximate value Vd1 of the applied voltage when the semiconductor device under test is tested at the first temperature T1. In this example, an approximate straight line 301 for the first group is obtained, similar to the example shown in FIG. 6. FIG. 7 shows the relationship between the test voltage Va and the breakdown voltage Vb obtained using the first group, and the approximate straight line 301.

[0039] Next, a first test stage S104, a second test stage S106, and a relationship acquisition stage S108 are performed on a second group including two or more setting groups 200 different from the first group. In the first test stage S104 for the second group, a test voltage Va corresponding to the above-mentioned approximate value Vd1 of the applied voltage is assigned to each setting group 200. For example, a test voltage Va within a predetermined voltage range 303 including the approximate value Vd1 is assigned to each setting group 200. In other words, a test voltage Va near the approximate value Vd1 is assigned to each setting group 200.

[0040] 8 is a diagram illustrating the processing for the second group. As described above, the test voltage Va in the voltage range 303 near the estimated value Vd1 is set for the setting group 200 of the second group. Then, the first test stage S104 and the second test stage S106 are performed to obtain the relationship between the test voltage and the breakdown voltage. In FIG. 8, each relationship is shown by a plot 305.

[0041] Then, in the relationship acquisition step S108, an approximate straight line 304 of the multiple plots 305 is acquired. Furthermore, based on the approximate straight line 304 and the predetermined withstand voltage Vc, an applied voltage Vd2 to be applied to the semiconductor device under test at the first temperature T1 is acquired.

[0042] In this way, by determining the applied voltage Vd2 in two stages, many plots 305 can be arranged near the estimated value Vd1, allowing the applied voltage Vd2 to be set with high precision. The number of setting groups 200 included in the second group may be greater than the number of setting groups 200 included in the first group. Furthermore, the number of setting semiconductor devices included in one setting group 200 in the second group may be greater than the number of setting semiconductor devices included in one setting group 200 in the first group. This allows the applied voltage Vd2 to be set with even higher precision.

[0043] 9 is a diagram illustrating an overview of a test apparatus 100 that tests a setting semiconductor device 210. The test apparatus 100 may be used in a first test stage S104 and a second test stage S106. The test apparatus 100 may also be used to test a semiconductor device under test. In this example, the setting semiconductor device 210 is an IGBT. The setting semiconductor device 210 has a gate terminal G, a collector terminal C, and an emitter terminal E.

[0044] In at least one of the first test stage S104 and the second test stage S106, the test apparatus 100 of this example may further measure the leakage current (ICES) of the setting semiconductor device 210. In other words, the test apparatus 100 may perform a destructive test on the setting semiconductor device 210 while measuring the leakage current.

[0045] The test device 100 of this example has a power supply 4, a voltmeter 6, an ammeter 8, a resistor 10, and a data processing unit 12. The collector terminal C of the setting semiconductor device 210 is connected to the voltmeter 6 and the resistor 10. The gate terminal G and the emitter terminal E of the setting semiconductor device 210 are connected to the ammeter 8. The test device 100 applies a predetermined voltage V between the collector terminal C and the emitter terminal of the semiconductor device 210 with the gate terminal G and the emitter terminal E of the setting semiconductor device 210 short-circuited. CE When a voltage is applied, the leakage current (ICES) that flows between the collector terminal C and the emitter terminal E is measured.

[0046] The power supply 4 applies a voltage to the setting semiconductor device 210. In this example, the power supply 4 is a variable voltage power supply that can change the voltage (applied voltage) applied to the setting semiconductor device 210. In this specification, the applied voltage may refer to the voltage between the collector terminal C and the emitter terminal E of the semiconductor device. Furthermore, applying a voltage to a semiconductor device refers to applying a voltage between the collector terminal C and the emitter terminal E of the semiconductor device, unless otherwise specified. When the power supply 4 applies a voltage to the setting semiconductor device 210, a collector current flows through the collector terminal C of the setting semiconductor device 210.

[0047] The voltmeter 6 measures the voltage applied to the setting semiconductor device 210. In this example, the voltmeter 6 measures the voltage between the gate terminal G and the collector terminal C of the setting semiconductor device 210 (or between the emitter terminal E and the collector terminal C of the setting semiconductor device 210). The voltmeter 6 is connected to the gate terminal G, the collector terminal C, and the emitter terminal E of the setting semiconductor device 210.

[0048] The ammeter 8 measures the collector current flowing through the collector terminal C of the setting semiconductor device 210. The ammeter 8 in this example is provided between the power supply 4 and the gate terminal G and emitter terminal E of the setting semiconductor device 210.

[0049] The resistor 10 limits the current flowing through the setting semiconductor device 210. In this example, the resistor 10 is provided between the power supply 4 and the collector terminal C of the setting semiconductor device 210. The resistor 10 is also connected to the data processing unit 12, and changes its resistance value based on the output of the data processing unit 12. In other words, the data processing unit 12 controls the resistance value of the resistor 10.

[0050] The data processing unit 12 is connected to the voltmeter 6 and the ammeter 8, and records the relationship between the applied voltage and the collector current of the setting semiconductor device 210. The data processing unit 12 may calculate the leakage current of the setting semiconductor device 210 based on the magnitude of the collector current when a predetermined voltage is applied to the setting semiconductor device 210. The data processing unit 12 may determine whether the setting semiconductor device 210 is good or bad based on the leakage current of the setting semiconductor device 210.

[0051] The data processing unit 12 may adjust the waveform slope of the rising edge of the applied voltage when applying the voltage to the setting semiconductor device 210. The waveform slope of the applied voltage can be adjusted, for example, by the resistance value of the resistor 10 or the rate of change of the output voltage of the power supply 4. The waveform slope of the rising edge of the applied voltage refers to the magnitude of the voltage increase per unit time. During an ICES test to measure leakage current, current or electric field may concentrate at a specific point on the setting semiconductor device 210, resulting in destruction of the setting semiconductor device 210. It has been experimentally confirmed that the ratio of the number of breakdowns to the number of semiconductor device tests (referred to as the breakdown rate) varies depending on the waveform slope of the rising edge of the applied voltage. The data processing unit 12 may adjust the breakdown rate of the setting semiconductor device 210 during leakage current testing by adjusting the waveform slope of the rising edge of the applied voltage. For example, increasing the breakdown rate of the setting semiconductor device 210 during leakage current testing allows setting semiconductor devices 210 that are relatively susceptible to breakdown to be removed in advance, thereby reducing the breakdown of setting semiconductor devices 210 after they are released to the market.

[0052] Figure 10 shows an example of measuring leakage current. Figure 10 shows the time changes of applied voltage and leakage current (collector current). In the graphs of applied voltage and collector current, the time when voltage application started is set to t0.

[0053] In the leakage current measurement, the voltage of power supply 4 is set to the specified voltage V CE For example, in the first test stage S104, the specified voltage V CE In the second test step S106, as shown in S504 of FIG. 5, the assigned test voltage V CE is gradually increased, and the designated voltage V CE , or the specified voltage V in the previous step CE is detected as the breakdown voltage Vb.

[0054] The voltage applied to the setting semiconductor device 210 ranges from 0V to a specified voltage V CEIn Figure 10, the applied voltage increases with a specified waveform slope until it reaches a specified voltage V CE The time when the specified voltage V is reached is defined as t1. CE When the voltage is kept at 100 V, the displacement current to the parasitic capacitance of the test device 100 decreases, and the collector current value stabilizes. At this timing, the collector current is measured as the leakage current. The leakage current is measured as I CES It is expressed as:

[0055] The waveform slope (dv / dt) of the rising edge of the applied voltage is expressed by the following formula: dV / dt=ΔV / Δt ΔV and Δt are the amounts of change in the applied voltage (V) and time (t) shown in FIG. 10. Δt may be the entire period from time t0 to time t1 shown in FIG. 10, or a part of it. Δt may be a period in which the waveform slope of the rising edge of the applied voltage is approximately uniform. Because the waveform slope of the rising edge of the applied voltage may not be uniform near time t0 and near time t1, Δt may be a period that does not include time t0 and time t1. Δt may be a period that includes the central time between time t0 and time t1.

[0056] FIG. 11 shows a case where the setting semiconductor device 210 breaks down during leakage current measurement. The bold line in FIG. 11 indicates the waveform when the setting semiconductor device 210 breaks down. When the applied voltage is increased, dielectric breakdown may occur between the collector and emitter of the setting semiconductor device 210. This causes the collector-emitter voltage to approach 0 V, causing a large collector current to flow. In the first test stage S104 and the second test stage S106, broken setting semiconductor devices 210 are screened. The same applies to the testing of semiconductor devices under test.

[0057] Specified voltage V CEIt has been experimentally confirmed that even if the slope of the rising waveform of the applied voltage is the same, the setting semiconductor device 210 will be destroyed or not depending on the slope of the rising waveform of the applied voltage. Therefore, if a leakage current test is performed without adjusting the slope of the rising waveform of the applied voltage, a setting semiconductor device 210 with a breakdown tolerance that should not be distributed to the market may survive the leakage current test but be destroyed after distribution to the market. In testing a semiconductor device under test, it is preferable to set the slope of the rising waveform of the applied voltage to a value that easily destroys the semiconductor device under test. Furthermore, in order to align the test conditions of the semiconductor device under test and the setting semiconductor device 210, it is also preferable to set the slope of the rising waveform of the applied voltage to a value that easily destroys the setting semiconductor device 210 in the first test step S104 and the second test step S106. In other words, it is preferable that the set value of the slope dv / dt of the rising waveform of the applied voltage be the same in testing the semiconductor device under test and testing the setting semiconductor device 210.

[0058] 12 is a diagram showing the relationship between the waveform slope (dv / dt) of the rising edge of the applied voltage and the breakdown rate of the semiconductor device. The horizontal axis of FIG. 12 is the waveform slope (dv / dt) of the rising edge of the applied voltage, and the vertical axis is the applied voltage V CE In addition, each plot in FIG. 12 shows the number of semiconductor devices inspected (denominator) and the number of semiconductor devices destroyed (numerator). In FIG. 12, the voltage applied to the semiconductor device is increased up to 1400 V for each waveform slope (dv / dt), and the applied voltage at which the semiconductor device first destroyed is plotted. In other words, the vertical axis in FIG. 12 corresponds to the breakdown voltage of the semiconductor device.

[0059] In Figure 12, black circles indicate that there was damage, and white circles indicate that there was no damage. Each black circle and white circle is assigned a damage rate. In the damage rate, the numerator and denominator represent the number of damages and the number of inspections. For example, 0 / 22 indicates that there were 0 damages out of 22 inspections (i.e., there was no damage). 8 / 22 indicates that there were 8 damages out of 22 inspections (i.e., there was damage).

[0060] When the waveform slope dv / dt of the applied voltage is in the range of 1 or less and in the range of 100 or more, the applied voltage V CE Even if the applied voltage V reaches the upper limit of 1400V, the number of semiconductor devices destroyed is 0. On the other hand, in the region where the waveform slope dv / dt is 10 to 100 [V / ms], CE The semiconductor device is destroyed when the waveform slope dv / dt is smaller than 1400 V. As shown in Fig. 12, when the waveform slope dv / dt reaches a specific value S1, the breakdown voltage of the semiconductor device reaches a minimum value and the breakdown rate of the semiconductor device reaches a maximum value.

[0061] Fig. 13 is a flowchart showing another example of the test method. The test method of this example further includes a gradient acquisition step S101 in addition to the test methods described with reference to Figs. 1 to 12. The other steps are the same as those of the test methods described with reference to Figs. 1 to 12.

[0062] In the gradient acquisition step S101, a specific value S1 of the waveform gradient dv / dt described in FIG. 12 is acquired. The specific value S1 is the waveform gradient at which the voltage at which the setting semiconductor device 210 breaks down is at its smallest. The specific value S1 does not need to strictly match the waveform gradient at which the withstand voltage of the semiconductor device becomes a minimum value or the breakdown rate of the semiconductor device becomes a maximum value. The specific value S1 may have an error of 10% or less with respect to the waveform gradient at which the withstand voltage of the semiconductor device becomes a minimum value or the breakdown rate of the semiconductor device becomes a maximum value.

[0063] In the gradient acquisition step S101, the specific value S1 may be acquired by testing a plurality of semiconductor devices, as described in Fig. 12. In this case, the semiconductor device used to acquire the specific value S1 is an individual device different from the setting semiconductor device 210. It is preferable that the semiconductor device is manufactured with the same design as the setting semiconductor device 210. Also, in the gradient acquisition step S101, the specific value S1 that is set in advance for each type of setting semiconductor device 210 may be acquired.

[0064] In the first test stage S104 and the second test stage S106, the specific value S1 is used as the waveform gradient dv / dt of the voltage applied to the setting semiconductor device 210. Also, in the DUT test stage S112, the specific value S1 is used as the waveform gradient dv / dt of the voltage applied to the semiconductor device under test. This allows screening of semiconductor devices under test that are prone to breakdown. Furthermore, the conditions of the first test stage S104 and the second test stage S106 can be matched with the conditions of the DUT test stage S112.

[0065] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the claims that such modifications and improvements can also be included within the technical scope of the present invention. [Explanation of symbols]

[0066] 4 power supply, 6 voltmeter, 8 ammeter, 10 resistor, 12 data processing unit, 100 test equipment, 200 setting group, 210 setting semiconductor device, 220 good product group, 230 defective product group, 300 approximation line, 301 approximation line, 303 voltage range, 304 approximation line, 305 plot

Claims

1. A method for testing a semiconductor device, comprising: a preparation step of preparing a plurality of setting groups each including a plurality of setting semiconductor devices and allocating different test voltages to the respective setting groups; a first test stage in which the assigned test voltage is applied to the plurality of setting semiconductor devices included in each of the setting groups and the setting semiconductor devices are tested at a first temperature; a second test stage in which the setting semiconductor device determined to be a non-defective product in the first test stage is tested at a second temperature different from the first temperature to detect a breakdown voltage at which the setting semiconductor device is broken; a relationship acquisition step of acquiring a relationship between the test voltage and the breakdown voltage for each of the setting groups; an applied voltage setting step of setting an applied voltage when testing the semiconductor device under test at the first temperature based on the relationship acquired in the relationship acquisition step; A test method comprising:

2. The second temperature is lower than the first temperature. The test method of claim 1.

3. In the step of setting the applied voltage, the applied voltage is set based on the relationship between the breakdown voltage that the semiconductor device under test should have at the second temperature and the relationship between the inspection voltage and the breakdown voltage.

3. The test method according to claim 1 or 2.

4. In the second test stage, the minimum breakdown voltage in each of the setting groups is detected.

4. The test method according to claim 1.

5. In the second test stage, one or more of the setting semiconductor devices are tested in order, and in the test of each of the setting semiconductor devices, the voltage applied to the setting semiconductor device is gradually increased to detect the breakdown voltage of the setting semiconductor device, and when the voltage applied to each of the setting semiconductor devices reaches the breakdown voltage of the setting semiconductor device that has already been tested, the next setting semiconductor device is tested. The test method according to claim 4.

6. In the second test stage, the distribution of the breakdown voltages in each of the setting groups is detected.

4. The test method according to claim 1.

7. The method further includes a gradient acquisition step of acquiring a gradient of the waveform of the test voltage when the test voltage is applied to the setting semiconductor device, the gradient of the waveform of the test voltage being the smallest at which the voltage at which the setting semiconductor device is broken down is smallest, In the first test stage and the second test stage, the setting semiconductor device is tested using the waveform gradient acquired in the gradient acquisition stage.

7. The test method according to any one of claims 1 to 6.

8. performing the first test stage, the second test stage, and the relationship acquisition stage on a first group including two or more of the setting groups to acquire an approximate value of the applied voltage when the semiconductor device under test is tested at the first temperature; assigning the test voltage according to the estimated value of the applied voltage to a second group including two or more setting groups different from the first group, and performing the first test stage, the second test stage, and the relationship acquisition stage; In the step of setting the applied voltage, the applied voltage to the semiconductor device under test is set based on the relationship acquired in the step of acquiring the relationship for the second group.

8. The test method according to any one of claims 1 to 7.

Images