JPH0675094B2 - Semiconductor device testing equipment - Google Patents

Description

The present invention relates to an apparatus for testing malfunction of a semiconductor element due to α-ray irradiation.

[Conventional technology]

A soft error is one of the malfunctions of semiconductor devices. The source of α-rays that cause this constitutes a semiconductor element.
Si chips, Al wirings, glass protective films and other materials, and plastics, ceramics and other materials for packaging these, which are contained in trace amounts on the order of ppb, U and Th.
Is a naturally radioactive substance such as. U, Th, etc. emit α particles (He nuclei) by spontaneous decay.

2 Fig. 10 shows α emitted from the package material of the semiconductor device.
It is the energy spectrum of the alpha ray showing the relationship between the number N of particles and the energy E. These α-ray energy distributions continuously show a wide distribution, and when these α-particles enter a semiconductor device, electron-hole pairs are generated along this range. In the MOS dynamic memory, a soft error occurs when these generated electrons accumulate in the empty potential well portion of the memory cell for a certain amount or more. Further, in the high speed bipolar static memory, the generated electrons flow as a noise current to cause a soft error. Such soft errors tend to increase with higher integration and higher speed of semiconductor devices.

Therefore, it became necessary to evaluate the soft error due to the α ray of the semiconductor element, which led to the creation of the present invention. When this evaluation is performed, it takes a long time for the test only with α particles naturally emitted from the package material, because the α ray density is very small. Therefore, generally, an accelerated test is performed using a natural radioactive material or an artificial radioactive material with a high α-ray density.
The α-ray source used in the acceleration test has a single energy distribution having a peak at a certain energy value (Fig. 2, Fig. 11), which is greatly different from the energy spectrum 10 emitted from the actual package material. When the soft error of the semiconductor device was evaluated using such an α-ray source, there was a problem, which was significantly different from the actual result. Further, the soft error rate is considered to depend on the α-ray energy with which the semiconductor element is irradiated, but in the conventional device, the irradiation α-ray energy could not be varied.

As an apparatus for solving the above problem, as described in Japanese Utility Model Laid-Open No. 55-118175, a metal thin plate is chemically etched on the α-ray passing path between the α-ray source and the semiconductor device under test so that the surface is irregular. The α-ray energy moderator that is formed by laminating a thin metal material or the like is used as the α-ray energy moderator, and α-rays pass through the moderator to generate a continuous and wide energy distribution.

As another device, as described in Japanese Patent Application No. 59-264459, a plurality of α-ray energy moderators having different thicknesses are provided on the α-ray passing path between the α-ray source and the semiconductor device under test in vacuum. By providing the pieces on the same rotation plane to form a disk-shaped moderator, and passing the α-rays from the α-ray source into the rotated moderator, it is possible to make the moderator continuous and wide. The energy distribution was generated.

Further, as an apparatus capable of varying the α-ray energy applied to a semiconductor element under test, as described in JP-A-56-48146, an α-ray between an α-ray source and a semiconductor element under test in vacuum is used.
The α-energy moderator film is arranged on the ray passage, and the α-ray energy incident on the semiconductor device under test can be varied by changing at least one of the thickness, the number and the material of the moderator film. There were also things I was able to do.

[Problems to be solved by the invention]

 The above conventional technique has the following problems.

1. Metals and organic thin films are used as moderators for α-ray energy, but 1) these cannot be permanently used for α-ray irradiation and deteriorate over time and their characteristics change. , It is necessary to compensate for reproducible characteristics. 2)
As a means for obtaining a desired amount of attenuation of α-ray energy, it is necessary to change the thickness, number, and material of the moderator,
The relationship between each of these parameters and the amount of energy attenuation needs to be obtained in advance, and the compensation becomes temporary.

2.When a thin metal plate is chemically etched, the surface is made irregular, or a thin metal material is laminated as an α-ray energy moderator, in the physical location of the attenuator, The only difference is the amount of energy attenuation of α rays passing through this. That is, the same physical position of the semiconductor device under test is always irradiated with a constant value of α-ray energy. That is, for dimensional accuracy of α-ray irradiation symmetry,
Irregularity of the surface becomes remarkable and uniform irradiation cannot be performed.

3. When chemical etching is used as a moderator processing method, it is difficult to make the moderator surface with the same etching degree and shape. That is, the energy spectrum varies depending on each moderator.

4. When evaluating the sost error of a semiconductor device with large power consumption such as ECL (Emitter Coupled Logic) bipolar RAM in vacuum, the electrical characteristics of the device will change due to self-heating from the semiconductor device. Sometimes. Therefore, it is necessary to cool the semiconductor element,
Since it is in a vacuum, it is technically difficult to perform sufficient cooling even if a heat pipe, a Peltier effect element, or the like is used.

5. When evaluating in vacuum, vacuum chamber, vacuum pump,
Since various devices such as vacuum gauges are required, and the vacuum chamber for accommodating these devices becomes large, it takes several tens of minutes to evacuate the vacuum, and the cost, workability, and maintainability of the device are reduced. Leave room for improvement.

The object of the present invention is to eliminate the above-mentioned problems of the prior art,
An object is to provide an apparatus for evaluating a soft error of a semiconductor device.

[Means for solving problems]

In order to achieve the above object, air is used as a moderator for α-ray energy, in other words, a distance is secured between the α-ray source and the semiconductor device under test to create an energy distribution of α-rays to be irradiated. did.

Means for solving the problem include means for controlling the relative distance between the α-ray source and the semiconductor device under test to an arbitrary value and means for driving the control means so as to obtain a predetermined energy distribution. See.

Specifically, the distance l between the α-ray source and the semiconductor device under test is
It is controlled by the α-ray source drive control device. Then, the relationship between the distance l and the α-ray energy attenuation is measured in advance, and the α-ray energy distribution for irradiating the semiconductor device under test is calculated based on this result, and the α-ray source is intermittently and continuously moved to move the distance l. By controlling .alpha.-ray energy and generating a continuous and wide energy distribution.

[Action]

The relationship between the distance 1 and the α-ray energy attenuation is measured in advance, the α-ray energy distribution irradiated to the semiconductor device under test is calculated based on this result, and the α-ray source is intermittently / continuously controlled by the α-ray source movement control device. The α-ray energy and the energy distribution are adjusted by controlling the distance 1 by moving the magnetic field.

〔Example〕

 An embodiment of the present invention will be described below with reference to the drawings.

α-rays have a large ionizing action, and when α-rays penetrate into a certain substance, the α-rays attenuate energy by the ionizing action on the substance and finally disappear.

Let E be the energy that α rays initially have, and range R that is the distance that α rays disappear after passing through the substance.
There is a relationship of R∝E ^3/2 . In air, the relation between R and E is expressed by Geiger's equation R = 0.318 · E ^3/2 and is as shown in FIG. However, the α-ray source actually used for acceleration evaluation has a single energy distribution 11 having a peak at a certain energy value, as shown in FIG. The amount of attenuation cannot be calculated. Therefore, as shown in FIG. 4, the distance between the α-ray source and the detector for analyzing the α-ray energy distribution is set to 1, and the change in the energy distribution when 1 is used as a parameter is examined. As a result, the distance l becomes l ₀ , l ₁ , l ₂ , l ₃ , l ₄ (l ₀ <l ₁ <
As _{2 2} <l ₃ <l ₄ ) increases, the amount of α-ray energy attenuation in the air increases, and the energy distribution of α-rays incident on the detector is N _o (E), N ₁ (E), N ₂ (E), N
It was found that the energy values of ₃ (E) and N ₄ (E) shift to smaller values, and at the same time, the peak value of the number of α particles also decreases.

Therefore, the relationship between the distance 1 and the amount of α-ray energy attenuation is measured in advance, and the generation condition of the α-ray energy distribution to be irradiated (the time t for holding the distance 1 and the distance 1) is determined based on this data. The soft error is evaluated by irradiating the semiconductor element under test with α-rays while controlling the α-ray source by the α-ray source movement control device, using the result obtained by calculation (simulation).

That is, when it is desired to generate a continuous and wide α-ray energy distribution 12 as shown in FIG. 5, it can be expressed as the sum of this single distribution, and α-rays for generating each single distribution can be expressed. The relationship between the moving distance l of the source and the time t during which the α ray is held may be obtained, and the relationship is as shown in FIG. Regarding the selection of the time t, the holding time t is t ₀ ′ <t ₁ ′ <with the distance 1 kept constant as shown in FIG.
By increasing t ₂ ′ <t ₃ ′, it is possible to increase the number of α particles in the single distribution while keeping the α-ray energy E constant.

FIG. 1 is a configuration diagram of a test apparatus according to an embodiment of the present invention. In FIG. 1, 1 is an α-ray source that emits α-rays, and 2 is an α-ray source.
An α-ray source movement control device for moving the radiation source 1 up and down and holding it for a required time, 3 is a semiconductor device under test, 4 is a socket for extracting a signal from the semiconductor device under test 3, and 5 is a semiconductor device under test 3 A soft error detection tester that compares the data read from the normal data with the normal data to check whether a soft error has occurred, 6 is a detector for detecting the α-ray energy emitted from the α-ray source 1, and 7 is a detector A wave height analyzer for processing the signal from 6 and measuring the α-ray energy distribution emitted from the α-ray source 1, 8 simulates the α-ray energy distribution with which the semiconductor element 3 under test is irradiated, and determines the generation conditions. Calculator for setting, 9 is the semiconductor device under test 3
It is a dark box that shields light coming in from the outside.

In the illustrated example, the semiconductor device under test 3 (socket 4) is α
The detector 6 is located right below the radiation source 1 and is arranged beside it, but the respective arrangements of the semiconductor device under test 3 (socket 4) and the detector 6 can be appropriately exchanged. .

The operation of the apparatus described above will be described. First α source 1
Using the distance l between the detector and the detector 6 as a parameter, the α-ray energy distribution at each distance l ₀ , l ₁ , ... In as shown in FIG.
N ₀ (E), N ₁ (E), ... Nn (E) is measured for a certain period of time T ₀ by using the wave height analyzer 7, and this data is input to the computer 8 as basic data. The computer 8 simulates the α-ray energy distribution with which the semiconductor element 3 to be tested is irradiated using the above-mentioned basic data, and obtains the generation condition (distance 1 and holding time t of the distance 1). At this time, if the generated α-ray energy distribution is N (E), The generation distribution can be calculated from the relational expression of, and the relationship between l and t as shown in FIG. 6 can be obtained. next,
By using the obtained relationship between l and t, the α-ray source drive / control device 2 intermittently and continuously controls the movement of the α-ray source 1.
The irradiation α-ray energy can be varied and a continuous and wide α-ray energy distribution can be generated. Then, the semiconductor device under test 3 (socket 4) is arranged directly under the α-ray source 1, the semiconductor device under test 3 is irradiated with α-rays, and the data from the semiconductor device under test 3 is read by the soft error detection tester 5. Read and compare the data with normal data to check if a soft error has occurred.

As described above, the relationship between the α-ray energy attenuation in the distance between the α-ray source and the detector in air is obtained, and the desired α
The generation condition of the line energy distribution (distance 1 and holding time t of the distance 1) is obtained, and the α-ray source drive / control device controls the α-ray source based on this result, so that the incident α on the semiconductor device under test. It is possible to change the line energy and generate a continuous and wide α-ray energy distribution, and it is possible to evaluate the soft error of the semiconductor device under test more accurately than before.

〔The invention's effect〕

According to the device of the present invention, it is possible to generate a wide range continuous energy distribution having a low energy region in air from an α-ray source having a single energy distribution having a peak in energy by a method simpler than the conventional method. Therefore, the state close to the α-ray energy distribution emitted from the actual package material can be reproduced, and the soft error rate of the semiconductor device can be evaluated more accurately than before.

Also, by changing the α-ray energy applied to the semiconductor element,
By measuring the α-ray radiation resistance (α-ray energy dependence) of the semiconductor element, it is possible to perform an evaluation corresponding to the structure of the semiconductor element and to analyze the mechanism of the soft error failure.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a test apparatus according to an embodiment of the present invention, FIG. 2 is a distribution chart of α-ray energy emitted from a package material of a semiconductor device and an α-ray source, and FIG. 3 is an α-particle in air. A diagram showing the relationship between energy and range, Fig. 4 (a) is α
FIG. 4 (b) is a diagram showing the positional relationship between the radiation source and the detector, FIG. 4 (b) is an α-ray energy distribution diagram when the distance between the α-ray source and the α-particle energy detector in air is used as a parameter, and FIG. 5 is FIG. 6 is a diagram showing an example of an α-ray continuous energy distribution generated by the device of the present invention, FIG. 6 is a time chart showing the generation condition of the α-ray energy distribution generated by the device of the present invention, FIG.
The figure is an α-ray energy distribution diagram when the distance between the α-ray source and the detector is fixed and the α-ray irradiation time is used as a parameter. 1 ... α-ray source 2 ... α-ray source drive / control device 3 ... Semiconductor device under test

Claims (1)

[Claims] 1. A semiconductor device testing apparatus for irradiating a semiconductor device under test with α-rays to test malfunctions of the semiconductor device due to α-rays, wherein an α-ray source placed in air and the α-rays are used. Means for controlling the distance between the source and the semiconductor device under test, the distance and the time for holding the distance are appropriately set, and the α-ray having the required energy and energy distribution is applied to the semiconductor device under test. A semiconductor device testing device characterized by irradiation.