JPS61253479A - Selection of semiconductor device - Google Patents

Abstract

PURPOSE:To select a device with a provable deterioration by radiation, by measuring changes in the current amplification factor at a fine current area of a silicon bipolar transistor within a semiconductor device or characteristic standard of the device determined by the changes in the current amplification. CONSTITUTION:Some semiconductors made up of a silicon bipolar transistor or the like may be deteriorated by radiation in the use under a radiation environment. In the low dose irradiation, the current amplification factor at a small current area of the bipolar transistor changes greater than that at a large current area. So changes in the current amplification factor at the fine current area of the bipolar transistor or changes in the characteristic standard of the semiconductor device based on the changes in the current amplification factor are measured. This enables the selection and removal of the semiconductor with a provable deterioration by radiation from the measured data.

Description

[Detailed Description of the Invention] [Summary] Semiconductor devices composed of silicon (Si) bipolar transistors and the like may suffer significant radiation deterioration when used in a radiation environment. Changes in the current amplification factor or changes in the current amplification factor in the minute current region of the Si bipolar transistor in the semiconductor device when In order to be able to select and remove certain semiconductor devices in advance (2).

[Industrial Application Field] The present invention relates to a method for sorting semiconductor devices according to their quality. The present invention relates to a semiconductor device sorting method for preliminarily selecting and removing semiconductor devices that are likely to undergo radiation deterioration.

[Conventional technology]

Conventionally, this type of screening method for radiation damage and deterioration focuses on differences in production lots, and samples are taken from a specific lot, irradiated with a dose corresponding to the actual use environment C, and the lot with less deterioration is selected for practical use. It was served to

Further, the conditions for measuring the electrical characteristics at this time were transistor standard values, ie, mass production measurement conditions, or bias conditions close to the standard values.

[Problem that the invention seeks to solve]

As described above, conventionally, a sampling test has been conducted for each lot based on the measurement judgment conditions in the medium current region of the collector current.

For this reason, in order to obtain the necessary radiation resistance (secondly, a sufficient margin is required for the lot input into manufacturing). When actually used in a radiation environment for sampling tests (2) there was a drawback that semiconductor devices with large characteristic fluctuations were mixed in.

In order to solve these drawbacks and minimize the occurrence of damage due to screening, the present invention performs selection based on characteristic changes C2 caused by low-dose irradiation.

[Means for solving problems]

The present invention is based on the fact that the current amplification factor in the small current region changes more with low dose irradiation than in the large current region. In other words, the dependence of the current amplification factor on the collector current before and after radiation irradiation is as shown in Figure 1C2. This is because the cause of the change in current amplification factor is an increase in pace current due to the occurrence of surface damage in the pace region, and this effect is greater in the small current region.

Here, the correlation between the rate of change in the current amplification factor in a large current region close to actual operating condition C2 and the rate of change in the current amplification factor for small currents and large currents will be quantitatively shown using experimental data. Table 1 shows how each correlation coefficient changes with increasing dose.
This shows the types of transistors. With 10' rad (Si) as the standard for correlation, the correlation coefficient between current amplification factors for large currents is naturally 1, which is larger than that for small currents, but from 10' to 1Q' rad (Si
), conversely, the correlation coefficient is larger for small currents than for large currents.

From this, it can be said that the estimation accuracy of the rate of change of the current amplification factor for high doses and large currents (; is higher for small currents).

Table 1 Correlation number of rate of change in current amplification factor (Alx) [Example] The screening method of the present invention will be described in detail below based on Example (: Example).

FIG. 2C is an embodiment of the method of the present invention. Dose 'C1
0” rad (Small current for S irradiation (100,4A) -
The distribution of the rate of change in the current amplification factor and the rate of change in the large current (1077!A) and current amplification factor for irradiation with a dose of 1Q' rad (Si) is shown. The characteristics of semiconductor components and their rate of change are generally C:
It shows a normal distribution, and the spread of the distribution can be expressed as the variance (σ).

Here, as a specific example, as shown in Fig. 2, 14 samples on the upper limit side of the distribution with 1 knee iff or more in the case of 1Q' rad (Si) irradiation are selected as defective samples from the population of samples (85 samples). Remove. Next, in order to evaluate whether this screening method is effective or not, high-dose irradiation was performed on all samples, including those that were screened out, and the samples that were screened out were high-dose irradiated and the upper limit of the distribution was 10 or more. We evaluated the extent to which it was included in the side. As a result, 10 “r
For at(Si), 12 samples had a value of 1σ or more, and the sini-cleaning effect was observed in the majority of samples.

Here, the loss of function of a semiconductor device means that the characteristic standard is broken, and in this example, the minimum value of the current amplification factor of the standard, 4o, is broken by irradiation of 10" rad (Si) or more. A test to determine whether or not this is the case can be carried out, for example, by subjecting a large number of samples to high-dose irradiation, numbering them, and measuring the characteristics of the obtained samples.

The screening method described above was applied to two types of transistors and evaluated. As shown in t'fi2, the screening effect was recognized even at high doses as shown in Table 2, and more than 80% of defective samples could be selected by this screening method. can.

Table 21 Example of application of this selection method From these results, it is found that semiconductor devices with impaired functionality are 1/10
It can be seen that it is possible to predict the characteristic deterioration at a dose of 0.00.

Note that the dose for estimation needs to be a dose that causes cadaverization to be greater than the measurement accuracy of the measurement system (in this example, the current amplification factor is approximately 0.1). Furthermore, characteristic deterioration due to estimated dose can be fully recovered to near its initial value by heat treatment at the maximum storage temperature of the transistor (e.g. 175℃) for about one day, and even when used for a long time without recovery. , there is no problem at all in practice.

Furthermore, instead of evaluating the change itself in the current amplification factor in the minute current region of the Si bipolar transistor in the semiconductor device, the evaluation is performed based on the change in the current amplification factor in the minute current region of the Si bipolar transistor in the semiconductor device. Measuring changes in the characteristic specifications of a semiconductor device.

〔Effect of the invention〕

As explained above in detail based on the examples, the method of the present invention focuses on the change in current amplification factor and selects the sample with the upper limit of the distribution, and measures the radiation resistance of the target Si bipolar transistor. Regardless of the presence or absence of the method, lot sorting (;) can also be applied, and it can also be applied to semiconductor devices including Si bipolar transistors.

[Brief explanation of the drawing]

Figure 1 is a graph showing the collector current dependence of the current amplification factor before and after radiation irradiation, and Figure 2 is the change rate distribution of the current amplification factor after radiation irradiation, showing the current amplification factor of small current during low-dose irradiation. It is a graph showing the distribution position occupied by the sample removed by screening in the distribution of current amplification factor of large current in high dose irradiation. Patent applicant Nippon Telegraph and Telephone Corporation agent Patent attorney Gobe Tamamushi (2 others) 1st grade)
Room 1 Medium M Car

Claims (1)

[Claims] For a Si bipolar transistor or a semiconductor device configured using a Si bipolar transistor, the semiconductor device causes a change in the characteristic specifications that is sufficiently smaller than the dose that impairs the function and is greater than or equal to the measurement accuracy of the measurement system. By irradiating a dose of radiation and measuring the change in the current amplification factor in the microcurrent region of the bipolar transistor in the semiconductor device before and after the irradiation, or the change in the characteristic specification of the semiconductor device determined by the change in the current amplification factor. Therefore, a method for sorting semiconductor devices, characterized in that the semiconductor devices are screened for radiation deterioration.