JPS63177076A - Method for feeding compound semiconductor avalanche photodiode equipped with inp cap layer - Google Patents

Abstract

PURPOSE:To accurately remove an inferior ternary light receiving element (APD) and to select an element stably operating over a long period of time, by performing a reverse direction current driving test of 100+ or -5muA corresponding to a reverse direction current value at the time of light irradiation to ternary APD at 200+ or -20 deg.C for a predetermined time. CONSTITUTION:A reverse direction current driving test of 100+ or -50muA corresponding to a reverse direction current value at the time of light irradiation is performed to ternary APD at 200+ or -20 deg.C for a predetermined time. Before and after said test, by removing ternary APD wherein a dark current exceeds a certain tolerant value by surface deterioration, the quality of ternary APD is judged. By this method, the element stably operating over a long period of time is selected.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides an I
The present invention relates to a method for easily selecting compound semiconductor avalanche photodiodes with nP cap layers.

Conventional technology Conventionally, Si or Ge, which operates stably over long periods of time, has been used.
Empirically, the temperature is set at 100 to select the photodetector (APD).
A high-temperature bias test in which a reverse bias voltage (20 to 30 V) is applied at ~125° C. is performed for a certain period of time, and light-receiving elements with little change in dark current before and after the test are selected as non-defective products. At this time, the important selection conditions are test temperature,
These are driving conditions, test time, and criteria for determining pass/fail. For example, in the case of GeAPD, the empirically determined temperature is 100°C,
A high temperature bias test with an applied reverse bias voltage of 20 to 30 V was conducted for 70 hours, and by selecting elements with a small increase in dark current, it was possible to obtain good products with sufficient reliability.

However, the deterioration mode of ternary Kiyadao (hereinafter referred to as ternary APD) is completely different from that of 3i and Ge photodetectors, and it is not possible to apply the same selection conditions as 3i and Ge photodetectors. The exact selection conditions are not clear.

In general, the factors that accelerate the deterioration of ternary APDs are temperature and bias applied in the opposite direction, and the observed deterioration modes include an increase in dark current caused by the surface of the ternary APD and characteristic deterioration due to the reaction between the electrode and the semiconductor. There is. The latter mode of deterioration is usually observed in temperature tests of 200°C or higher, and the rate of deterioration at actual operating temperatures is extremely small and does not pose a problem. On the other hand, the former deterioration mode is caused by pinholes in the surface protective film, and is a serious deterioration mode that causes random failures and determines the actual service life.

Since such surface deterioration progresses due to the combined effects of temperature and bias, a high temperature bias test method can be applied to eliminate this deterioration mode in advance. However, at that time, it was unclear how to set the test temperature, drive method, test time, and pass/fail criteria.

. Therefore, an object of the present invention is to provide a selection method for accurately removing defective ternary APDs and selecting elements that operate stably over a long period of time.

According to the present invention, for a ternary APD, at a temperature of 200±20°C, the reverse current value during light irradiation is 100±5.
A 0 μA reverse current drive test is performed for a predetermined period of time, and ternary APDs whose dark current exceeds a certain tolerance value due to surface deterioration are removed before and after the test to determine whether the ternary APDs are good or not.

100 to 12, which has been conventionally implemented with 3i and Qe photodetectors.
The selection method using a constant bias voltage drive test at a temperature of 5° C. is different from the test temperature, bias application method, and pass/fail judgment criteria.

As described above, it is mainly surface deterioration failures that govern the actual service life of ternary APDs and cause random failures. This degradation mode has a temperature activation energy of about 18V and 200V.
It becomes noticeable at temperatures above ℃. For example, as shown in the example below, under the conditions of 200°C and 100 μA, 1 ooh
Initial failures appear by then, and wear failures due to surface deterioration become dominant after 1000 hours or more. A test temperature of 220° C. or higher is not preferable because wear failure starts after about 400 hours and electrode deterioration is also expected.

On the other hand, at a temperature of 180° C. or lower, a minimum test time of 300 hours or more is required, which is not realistic in terms of cost.

Also, consider the optimal value of the reverse direction current. Since a ternary APD is actually used in a breakdown state and a reverse current of several tens of microamperes flows during optical input, a current value of at least 50 microamperes or more is required. However, if the current exceeds 200 μA, a short circuit failure due to heat generation due to local concentration of current is observed, so it is necessary to limit the current to 150 μA at most.

Therefore, the appropriate test temperature range is 180 to 220°C, and the appropriate reverse current range is 50 to 150 μA.

Embodiment FIG. 1 is a diagram for explaining the present invention in detail.
When a constant current drive test was performed on a PD with a reverse current of 10 CIA at a temperature of 200°C, the distribution of the time (life) until failure due to increase in dark current was plotted on so-called lognormal probability paper. In the figure, measurement point 1-1 indicates the time until an initial failure occurs, area 1-2 indicates an initial failure area, and 1-3 indicates a wear-out failure area.

From this, when performing the original high temperature constant current drive test at a temperature of 200° C. and a reverse current of 100 μA, the initial failure region and the wear-out failure region can be clearly separated after several hundred hours. In this example, an initial failure occurs within 100 hours. Therefore, aging for at least 100 hours is required to ensure that wear-out failures do not appear and early failure elements are reliably removed. In fact, when this selection condition (temperature 200°C, 100μA constant current drive, test time 100h) was applied, the increase in dark current (current value when reverse bias was applied to 90% of breakdown voltage) after the test was 30μ8 or less. and the absolute value of the dark current is 100 μ8 or less.
FIG. 4 shows an example of dark current change over time when a high temperature constant current drive test was conducted under the condition of 00 μA. After approximately 3,500 hours, there were no failures or deteriorated elements, and the system was operating extremely stably. On the other hand, when the temperature is set higher than 200° C., failure in the wear mode is accelerated and progresses, making it impossible to detect the initial failure mode and the wear mode separately. or,
FIG. 2 shows an example of the change in dark current over time when the temperature was lowered to 150° C. and a constant voltage drive test with a reverse bias of 70 V, similar to that performed with conventional Si and Qe light receiving elements, was performed. 40
Although it is stable up to 0fi, when the same element was subjected to a constant current drive test of 100 μA at a temperature of 150°C, many of the elements failed due to an increase in dark current within just 100 hours. actually 3
When the original APD is used, it is in a breakdown state and a current of several tens of μA flows during optical input.

Therefore, the high temperature constant voltage drive test below the breakdown voltage, which has been commonly used in the past, is not appropriate. Also, Figure 3 shows the temperature 15
A device with an excellent dark current increase of 30 μA or less and an absolute value of 100 μ8 or less after performing a constant current drive test of 100 μA at 0°C for 72 hours is considered to be a good product.
An example of the change in dark current over time when a high-current single current drive test of 100 μA at 150° C. is performed is shown. Some of the elements failed within 1000 hours, and this condition is inappropriate as a method for reliably selecting only non-defective elements. From the above, as a selection condition, performing a high temperature constant current drive test at 200° C. and 100 μA for 100 hours to select elements with a small increase in dark current is appropriate as a method for selecting good products in a short time. It should be noted that if the screening test is about 100 oh, there is no problem since at most 10% of the lifespan is due to the inherent wear of the element.

In this case, in the above embodiment, the acceptance criterion is 1ooh.
The increase in dark current after the test was set to 30 μA. However, the test time and the dark current increase are closely related, and the test time can be shortened by setting the 1lIN current increase to a small value, and slight changes in the test time and judgment criteria are within the scope of the present invention.

As explained above, according to the present invention, a 3-element APD is subjected to a reverse constant current test of 100 μA at a temperature of 200° C. for a predetermined period of time, and a 3-element APD with a large dark current increase due to the test.
It has the advantage of being able to easily select ternary APDs that operate stably for a long period of time, except for D, and the practical effect is extremely large.

Note that the above description merely shows one embodiment of the present invention, and various modifications and changes can be made without departing from the spirit of the present invention.

[Brief explanation of the drawing]

FIG. 1 shows an example of the lifetime distribution of ternary APDs, which is used to explain the ternary APD selection method according to the present invention. Point 1-1...... Time point area when initial failure occurs 1-2...... Initial failure distribution area Area 1-
3...Wear failure distribution area Figure 2 shows temperature 1
An example of a change in dark current over time is shown when a 70 V reverse constant voltage test (solid line) and a 100 μA reverse constant current test (broken line) were performed at 50°C. FIG. 3 shows an example of the change in dark current over time when a device that passed a reverse constant current test of 100 μA for 72 hours at a temperature of 150° C. was aged under the same test conditions. Figure 4 shows the dark current when a 100 μA reverse constant current test was performed for 100 hours at a temperature of 200°C, and a device that was found to be good was subjected to a 100 μA reverse constant current test at a temperature of 150°C. This is an example of change over time. Applicant: Nippon Telegraph and Telephone Co., Ltd.
Figure 0
Written amendment (method) % formula % 1. Indication of the case Japanese Patent Application No. 62-008654 2. Title of the invention Method for selecting compound semiconductor avalanche photodiodes with 1nP cap layer 3. Relationship with the person making the amendment Patent Applicant Address: 1-1-6 Uchisaiwai-cho, Chiyoda-ku, Tokyo Name:
(422) Shinfuji, Representative of Nippon Telegraph and Telephone Corporation
Kou 4, Agent address: 5-7 Kojimachi, Chiyoda-ku, Tokyo 102 Hidekazu Kioicho TBR No. 820 No. 5, Date of amendment order: March 31, 1986 (shipment date)
6. The number of inventions will increase due to amendments? 51) G, 7, Subject of amendment The name of the invention in the application Full text of the description and drawings 8, Contents of amendment (1) Application As attached (2) The name of the invention in the description "Selection method for compound semiconductor avalanche photodiodes"
The text has been corrected to read, "Selection method for compound semiconductor avalanche photodiodes with rlnP cap layer." (3) Engraving of the specification and drawings originally attached to the application As shown in the attached sheet (no changes to the contents) 9. Written amendment to the attached documents declaration procedure June 2, 1988 Commissioner of the Japan Patent Office Black 1) Mr. Yu Aki 1. Indication of the case Japanese Patent Application No. 62-008654 2. Name of the invention Method for selecting avalanche photodiodes 3. Relationship with the person making the amendment Patent applicant address 1-1-6 Uchisaiwai-cho, Chiyoda-ku, Tokyo Name Name
(422) Shinfuji, Representative of Nippon Telegraph and Telephone Corporation
Kou 4, Agent address: 5-7 Kojimachi, Chiyoda-ku, Tokyo 102 Hidekazu Kioi-cho TBR No. 820 No. 5, Date of amendment order Voluntary amendment 6, Detailed explanation of the invention increased by the amendment Document 1, Name of the invention Avalanche Photodiode selection method 2, claims A compound semiconductor avalanche photodiode with an InP cap layer has a value corresponding to the reverse current value during light irradiation in actual use at a temperature of 180 to 220°C. A reverse direction current is applied for a predetermined period of time, and the reverse direction ff1 of the compound semiconductor avalanche photodiode is
A method for selecting an avalanche photodiode, the method comprising selecting whether the compound semiconductor avalanche photodiode is good or bad based on the amount of variation in the dark current value before and after the FR is energized, and the dark current value after the reverse current is energized. . 3. DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for selecting the quality of a compound semiconductor avalanche photodiode with an InP cap layer. Conventionally, for Si or Ge avalanche photodiodes, an empirically determined temperature of 10
A reverse voltage having a value of 20 to 30 V is applied for a predetermined time at 0 to 125°C, and 3i or Ge
The quality of the Si or Ge avalanche photodiode can be judged from the amount of variation in the dark current value of the avalanche photodiode before and after the application of the reverse voltage described above.If the amount of variation in the dark current value is small, it is a good product, and if the amount of variation in the dark current value is large, it is a defective product. A method for selecting avalanche photodiodes has been proposed in which the selection method is as follows. Problems to be Solved by the Invention However, even if such an avalanche photodiode selection method is applied to a compound semiconductor avalanche photodiode with an InP cap layer (hereinafter referred to as a compound semiconductor avalanche photodiode for simplicity), It is not possible to determine whether the compound semiconductor avalanche photodiode is good or bad in a short time and with high reliability. The reason is as follows. That is, like 3i or Ge avalanche photodiodes, compound semiconductor avalanche photodiodes deteriorate due to temperature and reverse bias (voltage or current) as accelerating factors, and the 11ffi current value increases. The quality of the compound semiconductor avalanche photodiode can be determined from the amount of change in the dark current value of the compound semiconductor avalanche photodiode before and after applying a predetermined reverse bias at a temperature of . However, compound semiconductor avalanche photodiodes have a different deterioration mode from Si or Ge avalanche photodiodes, especially because their cap layers are made of InP. For this reason, in order to select the quality of compound semiconductor avalanche photodiodes by applying the above-described avalanche photodiode selection method, a reverse voltage having a value of 20 to 30V is applied to the compound semiconductor avalanche photodiodes, and the temperature at that time is If it is 100 to 125℃, the temperature is low, so
Unless the reverse voltage is applied to the compound semiconductor avalanche photodiode for an impractically long time such as 300 hours or more, it is not possible to reliably determine whether the compound semiconductor avalanche photodiode is good or bad. In addition, by applying the conventional avalanche photodiode screening method described above, the compound semiconductor avalanche photodiode was tested at a temperature of 100 to 150 to screen the quality of the compound semiconductor avalanche photodiode.
If the voltage value when applying a reverse voltage at ℃ is 20 to 30 V, the reverse current value (several tens of μAf & degrees) during light irradiation to the compound semiconductor avalanche photodiode during actual use. (50~150μ
Since it is not possible to flow a reverse current having A), it is not possible to reliably determine whether a compound semiconductor avalanche photodiode is good or bad. By means of solving the n points, the present invention provides a novel avalanche photodiode screening method that can screen compound semiconductor avalanche photodiodes as to whether they are good or bad in a short time and with high reliability. This is what we would like to propose. According to the method for selecting avalanche photodiodes according to the present invention, a reverse current value corresponding to a reverse current value during light irradiation in actual use is selected for a compound semiconductor avalanche photodiode at a temperature of 180 to 220°C.
The current is applied for a predetermined period of time, and from the amount of variation in the dark current value of the compound semiconductor avalanche photodiode before and after the reverse current is applied, and the maximum dark current value when the reverse current is applied, the compound semiconductor avalanche photodiode is determined. Select whether the avalanche photodiode is good or bad. The reason why the temperature is set at 180 to 220°C and the reverse current value is set to a value corresponding to the reverse current value during light irradiation during actual use is as follows. In other words, when these two conditions are satisfied at the same time, the deterioration that determines the quality of the compound semiconductor avalanche photodiode at the temperature during actual use (which determines the lifespan during actual use) is 300 hours or less. , within a relatively short practical period of time, it becomes noticeable as a wear-out failure mode mainly due to surface deterioration in a manner that is temporally distinct from the initial failure mode. However, although the reverse current value satisfies the above-mentioned conditions, the temperature does not satisfy the above-mentioned conditions.
When the temperature is set to a value below .degree. C., the above-mentioned deterioration hardly occurs unless an impractically long period of time, such as 300 hours or more, is used. Furthermore, when the reverse current value satisfies the above-mentioned conditions, but the temperature does not satisfy the above-mentioned conditions and is set to a value higher than 220°C, the above-mentioned deterioration occurs as the above-mentioned wear failure mode. Even if the wear mode is developed, the wear mode only appears in a manner that is temporally indistinguishable from the initial failure mode. Furthermore, when the temperature satisfies the above-mentioned conditions or the reverse current does not satisfy the above-mentioned conditions, the reverse current of the compound semiconductor avalanche photodiode naturally decreases to the reverse current value at the time of light irradiation during actual use. (Several micro
It does not flow at a value corresponding to A) (50 to 150 μA). From the above, unless both the temperature and the reverse current value satisfy the above-mentioned conditions at the same time, the amount of variation in the dark current value of the compound semiconductor avalanche photodiode before and after the application of reverse current indicates that the compound semiconductor avalanche photodiode It is not possible to easily determine whether the photodiode is good or bad. These are the reasons why, in the method for selecting avalanche photodiodes according to the present invention, the temperature is set at 180 to 220°C, and the reverse current value is set to a value corresponding to the reverse current value during light irradiation during actual use. It is. In addition, in the avalanche photodiode screening method according to the present invention, the quality of the compound semiconductor avalanche photodiode is determined by the amount of variation in the dark current value before and after the reverse current is applied, and the dark N value after the reverse current is applied.
The reason why compound semiconductor avalanche photodiodes are selected using the current value is that unless the compound semiconductor avalanche photodiode whose dark current value is higher than the expected dark current value is selected as a defective product, compound semiconductor avalanche photodiodes are This is because it is not possible to select with higher reliability whether the product is good or bad. 1. Nijinmaru Presentation According to the method for selecting avalanche photodiodes according to the present invention, from the above, it is possible to easily determine the quality of compound semiconductor avalanche photodiodes in a short period of time and with high reliability. It is clear that selection is possible. Embodiment Next, an embodiment of the method for selecting avalanche photodiodes according to the present invention will be described with reference to the drawings. Now, for a compound semiconductor avalanche photodiode, at a temperature of 200°C between 180 and 220°C, a current of 100 μA between 50 and 150 μA corresponds to the reverse current value (several tens of μA) during light irradiation in actual use. A reverse current with a value of is applied. Then, the amount of change in the dark current value of the compound semiconductor avalanche photodiode before and after energization, and therefore the increase in dark current, is measured. The results shown in FIG. 1 are obtained by plotting the distribution of the lifespan (lifetime) on so-called lognormal probability paper, with the horizontal axis representing the lifespan (hours) and the vertical axis representing the cumulative failure rate (%). In FIG. 1, measurement point 1-1 indicates the time until the initial failure mode occurs in the compound semiconductor avalanche photodiode, and area 1-2 indicates the time until the initial failure mode occurs in the compound semiconductor avalanche photodiode. In addition, regions 1-3 indicate regions where a wear-out failure mode can occur in a compound semiconductor avalanche photodiode. From the above, for a compound semiconductor avalanche photodiode, at a temperature of 200°C between 180 and 220°C, 50 to 150 μA, which corresponds to the reverse current value (several tens of μA) during light irradiation in actual use. If a reverse current with a value of 100 μA between It is clear that it is possible to clearly separate the area where this can occur. As described above, according to the embodiment of the method for selecting avalanche photodiodes according to the present invention, a reverse current having the above-mentioned value of 100 μA is applied to the compound semiconductor avalanche photodiode at the above-mentioned temperature of 200°C. Electrification was carried out for 100 hours, and the dark current values before and after the above-mentioned reverse current energization of the compound semiconductor avalanche photodiode (the current when the reverse bias voltage was applied at a value of 90% of the breakdown voltage) A compound semiconductor avalanche photodiode with a variation amount of 30μ8 or less and a dark current value of 100μA or less after the above-mentioned reverse current is applied to the compound semiconductor avalanche photodiode is considered to be a non-defective product. Select. The above is an embodiment of the method for selecting avalanche photodiodes according to the present invention. According to the method for selecting avalanche photodiodes according to the present invention, compound semiconductor avalanche photodiodes can be easily selected in a short time and with high reliability. This means that the compound semiconductor avalanche photodiode selected as a non-defective product according to the embodiment of the avalanche photodiode selection method according to the present invention has a value of 100 μA at a temperature of 150°C. When we verified and measured the change in dark current (A) over time of a compound semiconductor avalanche photodiode that was selected as a non-defective product by passing a reverse current with It is clear from the results obtained that all of the compound semiconductor avalanche photodiodes tested were operating extremely stably without failure or deterioration even after approximately 3,500 hours. Incidentally, irrespective of the method for selecting avalanche photodiodes according to the present invention, compound semiconductor avalanche photodiodes may be exposed to temperatures of 150° C. outside the temperature range of 180 to 220° C.
A reverse current with a value of 100 μA is applied at 7
The current was applied for 2 hours, and the amount of variation in the dark current value before and after the reverse current application to the compound semiconductor avalanche photodiode was measured in the same manner as in the example of the avalanche photodiode selection method according to the present invention described above. , and, as in the case of the embodiment of the avalanche photodiode selection method according to the present invention described above, the amount of variation is 30μ8 or less, and the dark current value of the compound semiconductor avalanche photodiode after the reverse current is energized as described above. Compound semiconductor avalanche photodiodes with a value of 100 μA or less were selected as non-defective products. Then, the compound semiconductor avalanche photodiodes selected as non-defective products in this way were verified and measured at a temperature of 150°C in the same way as the compound semiconductor avalanche photodiodes selected by the embodiment of the avalanche photodiode selection method according to the present invention described above.
When we verified and measured the change in dark current (A>) of a compound semiconductor avalanche photodiode that was selected as a good product by passing a current in the reverse direction with a value of 100 μA at ℃, we obtained the results shown in Figure 3. Therefore, an unreasonable amount of compound semiconductor avalanche photodiodes selected as good products fail or deteriorate within 1000 hours, even though they were selected as good products. In the above, the compound semiconductor avalanche photodiode was tested at 100 μA at a temperature of 200°C.
We have described the case where a compound semiconductor avalanche photodiode is judged as good or bad by passing a reverse current with a value of Even when the compound semiconductor avalanche photodiodes were judged to be good or bad by passing a reverse current, the same effects as in the embodiment of the avalanche photodiode screening method according to the present invention described above were obtained. 4. Brief Description of the Drawings FIG. 1 is a life distribution diagram of a compound semiconductor avalanche photodiode, which is used to explain the method for selecting compound semiconductor avalanche photodiodes according to the present invention. FIG. 2 shows a compound semiconductor avalanche photodiode in which a reverse current having a value of 100 μA is applied for 100 hours at a temperature of 200° C. to the compound semiconductor avalanche photodiode according to the avalanche photodiode selection method according to the present invention. sort out the quality of the
Regarding the compound semiconductor avalanche photodiode selected as a good product, at a temperature of 150°C,
FIG. 3 is a diagram showing the results of verifying and measuring the change in dark current of the compound semiconductor avalanche photodiode over time by passing a reverse current having a value of 100 μA. FIG. 3 shows that a reverse current having a value of 100 μA was applied to a compound semiconductor avalanche photodiode at a temperature of 150° C. for 72 hours without using the avalanche photodiode selection method according to the present invention. Semiconductor avalanche photodiodes were screened for quality, and the compound semiconductor avalanche photodiodes selected as good were tested at a temperature of 150°C.
FIG. 3 is a diagram showing the results of verification and measurement of the change in dark current of the compound semiconductor avalanche photodiode over time by passing a reverse current having a value of 00 μA. Applicant: Nippon Telegraph and Telephone Corporation

Claims (1)

[Claims] A compound semiconductor avalanche photodiode with an InP cap layer is subjected to a reverse current conduction test at a temperature of 200±50°C with a value of 100±50 μA corresponding to the reverse current value during light irradiation for a predetermined time, and the results are shown before and after the conduction test. A method for selecting a compound semiconductor avalanche photodiode, the method comprising selecting whether the compound semiconductor avalanche photodiode is good or bad based on the amount of variation in the dark current value of the compound semiconductor avalanche photodiode and the maximum dark current value after the test.