JPH11195684A - Semiconductor integrated circuit disconnection detecting equipment and method, and recording medium storing control program therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily and quickly detect a wiring failure in a semiconductor integrated circuit of CMOS structure. SOLUTION: A repetitive pulse potential is applied through a power supply wire and a grounding wire taking advantage of a feature of a circuit structure where either a power supply wire or a grounding wire is electrically connected to an inner signal wiring, when an input of data to a semiconductor integrated circuit of CMOS structure is kept in a steady state, whereby a potential change is propagated through an inner circuit. A non-defective unit is prepared as a specimen to detect a point of disconnection in a defective unit, an electronic beam tester mounted with an apparatus which is capable of replacing a semiconductor integrated circuit without decreasing a degree of vacuum is used to quickly obtain potential images formed by a potential change propagated through the inner wirings of both the non- defective and defective unit, one of the potential images obtained from a non-defective and a defective unit is subjected to picture image processing such as parallel displacement, deformation or inclination, so as to reduce a divergence of potential image between a non-defective and a defective unit to a minimum, then image processing where the image pictures are alternately displayed on the same picture plane is carried out, whereby only a point of disconnection is enhanced in visibility to easily detect a region as a point of disconnection where a potential change different from that in a normal wiring is indicated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus and method for detecting a disconnection fault in a semiconductor integrated circuit, and more particularly to an improvement in a method for detecting a disconnection fault in a semiconductor integrated circuit having a CMOS structure. .

[0002]

2. Description of the Related Art Conventional semiconductor integrated circuits (LSI)
In the electron beam tester used to identify the source of the failure signal, the change in the potential of the internal wiring due to the test vector supplied from the logic tester affects the surface potential when the electron beam is irradiated on the LSI surface. By measuring the change in the amount of secondary electron emission, a potential change pattern of the wiring of interest is obtained, and this is compared between non-defective and defective LSIs.

[0003] In addition, as a simpler detection method specializing only in a disconnection failure, electrons are injected into a gate directly connected to a disconnected wiring, so that a gap between a power supply line and a ground line is reduced. CIVA (Ch
arge Induced Voltage Alternation) method is described in Reference 1 (EI
Cole Jr. and REAnderson, “Rapid Localization of
Open Conductors Using Charge Induced Voltage Alter
nation ”Proceedingsof the 30th International Rel
iability Physics Symposium, 288-298 (1992)), which has been put to practical use.

However, the failure analysis using the electron beam tester has a disadvantage in that the analysis standard time requires two hours or more as compared with the conventional analysis technique specialized for single mode failure. 2. Description of the Related Art Conventionally, an electron beam tester requires a test vector for performing an operation test of an LSI in order to verify a potential change of a wiring of interest and a logic tester for providing the test vector to the LSI. It is necessary to take time to prepare a complicated wiring jig for connecting the LSI and the logic tester, and the increase in the number of pins due to the large scale of the LSI increases the preparation time that is indispensable in advance of analysis. There is.

Further, in the CIVA method, for example, in a three-input NAND gate, two inputs other than a broken line have a high level (hereinafter, referred to as H) and a low level (L).
As in the case where the output is fixed and the input of the gate having a disconnection is changed, the disconnection cannot be detected unless a change occurs in the gate output. In other words, there is a problem that it cannot be detected by the CIVA method depending on the location of the disconnection failure. Further, in addition to a normal scanning electron microscope (SEM) device, a separate arithmetic processing device is required for imaging, which has a disadvantage that the system configuration is complicated.

In order to solve the problems pointed out with respect to these conventionally proposed technologies, the present inventors disclosed in Japanese Patent Application No. 8-294149 a CMOS image sensor.
By applying pulse voltage repeatedly from the power line and the ground line of the semiconductor integrated circuit with the structure, most of the internal wiring of the semiconductor integrated circuit is always electrically connected to either the power line or the ground line in a static state Using the circuit structure to perform the above, the potential change obtained by propagating the potential change to the internal wiring without using the test vector is superimposed on the CAD layout image as a mask to improve the visibility of the disconnected portion and detect the disconnected portion. Application for technology.

Further, the inventors of the present invention have also disclosed in Japanese Patent Application No.
No. 80549 filed an application for a technique for detecting a broken portion only from a difference image between potential images of a good product and a defective product without using layout data in the above invention.

[0008]

SUMMARY OF THE INVENTION However, Japanese Patent Application No. Hei 8-2
The disconnection detection method described in U.S. Pat. No. 94149 requires a high-performance WS (workstation) for performing real-time image processing using layout data in addition to acquiring an electric potential image. Not a detection technology.

In the disconnection detection method described in Japanese Patent Application No. 9-080549, when comparing the potential images of good and defective products, an electron beam tester is used to quickly obtain the potential images of good and defective products. Each time two units are prepared or a good product and a defective product are exchanged, the degree of vacuum is reduced, so that it takes a long time to acquire an image, which is not a simple and high-speed detection technique.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a simple and high-speed method for detecting a disconnection fault in an LSI of a large-scale circuit.

[0011]

According to the present invention, a CMO
The semiconductor integrated circuit is irradiated with an electron beam while applying a constant voltage to the signal line of the S-structured semiconductor integrated circuit and applying a pulsed voltage to the power supply line and the ground line, respectively. A disconnection failure detection device for a semiconductor integrated circuit, which detects a signal to obtain a potential image of a wiring of the semiconductor integrated circuit and detects a disconnection failure from the potential image. A semiconductor integrated circuit, comprising: means for transferring and exchanging a non-defective product with the electron beam irradiation unit to acquire each of the potential images; and means for alternately displaying the potential images on the same screen. A disconnection failure detection device is obtained.

Further, according to the present invention, a constant voltage is applied to a signal line of a semiconductor integrated circuit having a CMOS structure, and a pulsed voltage is applied to a power supply line and a ground line, respectively. In the method of detecting a disconnection failure of a semiconductor integrated circuit, a beam is irradiated, a secondary electron signal due to the irradiation is detected, a potential image of the wiring of the semiconductor integrated circuit is obtained, and a disconnection failure is detected from the potential image. A step of transferring and exchanging a non-defective product and a defective product of the semiconductor integrated circuit to the irradiation part of the electron beam to acquire each of the potential images; and alternately displaying each of the potential images on the same screen. And a method for detecting a disconnection failure of the semiconductor integrated circuit.

The disconnection of the defective product is detected based on the brightness difference of each of the potential images, and the difference between the irradiated areas of the non-defective product and the defective product with the electron beam is minimized. The method further includes the step of adjusting one of the irradiated areas to acquire the potential image.

Further, the method further comprises the step of adjusting the brightness of one of the potential images so as to minimize the brightness difference between the potential images of the non-defective product and the defective product and acquiring the potential image.
Further, the method includes a step of generating a difference image between the potential images of the non-defective product and the defective product, and detects a broken portion of the defective product based on the presence or absence of an intermediate brightness area in the differential image.

Further, the method includes a step of generating a difference image between the potential images of the non-defective product and the defective product, and a step of changing the brightness of the difference image, and the disconnection of the defective product is performed by the brightness-changed difference image. It is characterized in that a portion is detected.

Further, according to the present invention, a constant voltage is applied to a signal line of a semiconductor integrated circuit having a CMOS structure, and a pulsed voltage is applied to a power supply line and a ground line, respectively. A method of detecting a disconnection failure of a semiconductor integrated circuit, which irradiates a beam, detects a secondary electron signal due to the irradiation, acquires a potential image of the wiring of the semiconductor integrated circuit, and detects a disconnection failure from the potential image. A recording medium recording a control program to be executed by a computer, wherein a non-defective product and a defective product of the semiconductor integrated circuit are moved to and exchanged with an irradiation unit of the electron beam to obtain each of the potential images. A step of alternately displaying each of the potential images on the same screen.

The operation of the present invention will be described. The potential images of the non-defective product and the defective product obtained by the above-described method correspond to the broken wiring and L
A region having no structure contributing to the operation of the SI and a region having a structure contributing to the operation of the LSI are images that can be identified by the difference in the brightness of the pixels. Thus, the disconnection can be quickly detected.

In addition, the electron beam irradiation for the non-defective product and the defective product can be moved and exchanged in a vacuum chamber.
Because it is possible to exchange the sample without lowering the degree of vacuum, the degree of vacuum is conventionally reduced at the time of sample replacement, it is possible to omit those that required a waiting time to raise the degree of vacuum to a predetermined value, The waiting time can be reduced from 10 minutes or more to 10 seconds or less.

Further, the potential images of the non-defective product and the defective product display the same position coordinates, and the other potential image displays the same position coordinates with reference to the center position coordinate of one of the potential images. After moving the display position and applying image processing such as distorting or tilting one image to minimize the deviation between the potential images, the potential images are acquired and displayed alternately on the same screen This makes it easier to detect a broken portion.

The image obtained as described above is subjected to image processing so that defective pixels and non-defective pixels having the same position coordinates overlap each other, and thus have a structure that contributes to the operation of the broken wiring and the semiconductor integrated circuit. This is an image in which a non-existent area and an area having a structure contributing to the operation of the semiconductor integrated circuit can be identified by the magnitude of the change in brightness, and the brightness differs at the disconnection point. In this way, the visibility of the broken portion is improved, so that the broken portion can be easily detected.

Further, image processing is performed so that the potential images of the defective and non-defective products completely overlap each other, and further, one of the potential images is used as a reference so that the difference in the brightness distribution of the potential images is minimized except at the disconnection point. The other potential image is subjected to image processing for shifting the lightness distribution, and alternately displayed on the same screen to make it easier to detect a broken portion.

The image obtained as described above is more effective in the operation of the broken wiring and the semiconductor integrated circuit than the image processed so that pixels having the same position coordinates as defective and non-defective products overlap on the screen. This is an image in which a region having no contributing structure and a region having a structure contributing to the operation of the semiconductor integrated circuit can be easily distinguished by the brightness of the pixel, and the image has a large brightness change only at a broken portion. In this way, the visibility of the broken portion is improved, so that the broken portion can be easily detected.

Further, in a region where the potential images of the defective product and the good product are other than the disconnection, a difference image is created from the potential image having the same brightness distribution and displayed on the screen, thereby making it easier to detect the disconnection portion.

The image obtained as described above is different from a potential image which has been subjected to image processing so that the same lightness distribution is obtained in regions other than the defective product and the non-defective product other than the disconnection. The image is an image that can be identified by the brightness of the pixel, and is different from the intermediate brightness only in the broken portion. In this way, the visibility of the broken portion is improved, so that the broken portion can be easily detected.

Further, by changing the brightness of the pixel in the difference image between the two potential images and the brightness specific to the broken portion to white or black, two images are displayed and alternately displayed on the same screen. , To make it easier to detect a broken portion.

The image obtained as described above is an image in which the visibility of a broken portion is improved as compared with an image in which a broken wire and another region can be distinguished by the brightness of a pixel. It will be easier.

[0027]

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

FIG. 1A is a block diagram showing an example of a failure analysis device for a semiconductor integrated circuit embodying the present invention. This analyzer is a scanning electron microscope (SE) including an electron gun 110.
M) 101, an apparatus 105 for exchanging non-defective products and defective products without lowering the degree of vacuum, and a workstation (WS) 111 for creating a potential image of the sample CMOS-LSI 100. An apparatus including the SEM 101 and the like is shown as an electron beam tester 102. SEM1
In the sample LSI 100 installed in the vacuum vessel No. 01,
The pulse generator 103 is connected to the power supply line terminal 106 and the ground line terminal 107 and is repeatedly applied with a pulse voltage.
8, a first-stage input voltage is applied by connecting a constant-voltage power supply 104.

The potential change of the internal wiring of the CMOS-LSI can be determined by irradiating the CMOS-LSI with the electron gun 110 while scanning the primary electron beam 109 in the vacuum vessel of the SEM 101, and irradiating the primary electron beam. Is measured by the amount of secondary electrons emitted according to the amount reaching the detector. The measurement result is subjected to image processing so that the emission amount of the secondary electrons is displayed in black and white gradation for each predetermined area divided by the WS 111 and displayed on the display 112.
The storage medium 118 is a memory device such as a ROM in which an operation control program of the WS 111 is recorded.

Next, FIG. 1B shows the movement exchange of the sample LSI 100 placed in the irradiation area of the electron beam 109.
This will be described with reference to FIG. FIG. 1B is a configuration diagram of an apparatus 105 for moving non-defective products and defective products to an electron beam irradiation area.
Either a non-defective product or a defective product is set in the electron beam irradiation area by moving in the horizontal direction 116, and the pulse generator 1 is further moved in the vertical direction 117.
03 and an electric signal from the constant voltage power supply 104 can be supplied.

Next, supply of an electric signal and a method of acquiring a potential image in the present invention will be described with reference to FIG. FIG.
(C) is a schematic diagram of the applied voltage. The potential difference between the high voltage of the repetitive pulse voltage 121 applied to the power line by the pulse generator 103 and the low voltage of the repetitive pulse voltage 123 applied to the ground line is applied to the maximum rated voltage 120 determined for each sample LSI and to the power line. The voltage of the constant voltage power supply 104 connected to the signal line is set so that the low voltage of the repetitive pulse voltage 121 is lower than the high voltage of the repetitive pulse voltage 123 applied to the ground line. Set the voltage value. afterwards,
In the secondary electron signal detection period 124 when the voltage is high or in the secondary electron signal detection period 125 when the voltage is low, the potential images of the non-defective product and the defective product are acquired and the two are compared to detect a broken portion. I do.

In FIG. 1C, the reason why two periods of the secondary electron signal detection periods 124 and 125 are provided alternately is that the detection (sampling) is performed in one period and the secondary period is detected in the other period. By performing the reset, the detected potential image is made clearer, and details thereof are described in the above-mentioned Japanese Patent Application No. 8-294149, which has been proposed by the present inventors.
Issue. Further, the constant voltage 12 applied to the signal line
The reason why 2 is a voltage value in the range where the pulse voltages 121 and 123 intersect is that a voltage in this range produces the best result.

Next, a method for detecting disconnection will be described with reference to FIG. FIG. 2 is a flowchart showing an algorithm of disconnection detection according to the first embodiment. First, in the non-defective and defective semiconductor integrated circuits, the defective LSI is fixed to the electron beam irradiation area (step 200).
A pulse voltage is repeatedly applied from a power supply line and a ground line, and a constant voltage is applied from a signal input line (step 201), and a potential image is obtained (step 202).

Next, the supply of the pulse voltage is stopped (step 203), and the defective and non-defective products are exchanged (step 204).
After restarting the supply of the pulse voltage (step 205),
A non-defective potential image is obtained (step 206). Next, the potential images of the defective product and the good product obtained in step 202 and step 206 are alternately displayed on the same screen (step 208).
It is confirmed whether or not there is a portion where the brightness of the non-defective product indicating the disconnection is different from that of the defective product (step 209).

In the potential image, if there is an area where pixels having different brightnesses are different between a good product and a bad product, it is determined that there is a wire to which a disconnection is connected in the image (step 224), and the process ends. I do. If there is no wiring connected to the disconnection, the irradiation position of the electron beam is changed (step 223), and the process returns to step 200 to repeat the above operation.

The determination criterion in step 209 can be based on the presence or absence of an area where pixels of different brightness at the same position in the potential image are present at a certain density or higher. The determination may be made by visual inspection, or may be performed by a known software technique such as setting a boundary value for the density of pixels of brightness determined to be a disconnection, and the method is not particularly limited.

Hereinafter, a method for detecting a disconnection by comparing images according to the first embodiment will be specifically described with reference to FIG. FIG. 3A is a schematic diagram of a potential image of a non-defective product in which a constant voltage is applied from a signal input line and a pulse voltage is repeatedly supplied from a power supply line and a ground line in FIG. (B) is a schematic diagram of a potential image which is defective but has no disconnection in the potential image. FIG. 3C is a schematic diagram in the case where there is a disconnection in the potential image of a defective product.

In each of the figures, 301 is a wiring through which a high potential propagates, 302 is a region where no wiring exists, 303 is a region through which a low potential propagates, 304 is a broken portion, and 305 is broken to change potential. The wiring which does not propagate is shown.

First, a pulse generator 103 shown in FIG. 1A applies a pulse voltage applied to a power supply line between 5 V and 2.5 V to a defective semiconductor integrated circuit having a disconnection therein. , And the pulse voltage applied to the ground line is set to swing between 2.6V and 0V. The pulse frequency is set to 2000 Hz, and the phase difference between the pulses applied to the power supply line and the ground line is set to 90 degrees.

A potential image of a non-defective product by detecting the number of arrivals of secondary electrons within 15 μsec after the rising or falling of the pulse by using the repetitive voltage pulse as a trigger, and further integrating the repetitive signal to improve the S / N ratio. 3
(A) is created. FIG. 3B is a potential image obtained by applying a voltage to the defective semiconductor integrated circuit under the same conditions.
Create In the image comparison between FIGS. 3A and 3B, it is determined that there is no disconnection, and the electron beam irradiation area is moved so as to display an adjacent area.

As shown in FIG. 3C, when there is a disconnection in the potential image, the disconnection is visually detected by alternately displaying FIGS. 3A and 3C. Alternatively, disconnection can be detected in the potential image by performing automatic determination by software.

Further, a second embodiment of the present invention will be described. The potential images of the non-defective and non-defective products are obtained by the method described in the first embodiment, and the potential images of the defective and non-defective products are displayed with the same position coordinates as the reference position coordinates at the center of one of the potential images. Image processing such that the display position is moved so that the other potential image displays the same position coordinates, and the position of one image is corrected or tilted so that the displacement between the potential images is minimized. , The potential image is acquired and alternately displayed on the same screen, thereby making it easier to detect only the change in the brightness at the broken portion.

In order to easily identify a broken portion based on the potential images of good and defective products, a change in brightness in the image of the broken portion caused by application of a voltage pulse is caused by a change in the brightness of the image in other regions. Create an image that is larger than the brightness change at.

Next, a method of detecting disconnection in the second embodiment will be described with reference to FIG. The hatching is different from FIG. 2, and FIG. 4 is a flowchart showing the disconnection detecting method according to the second embodiment.

First, in the case of defective and non-defective semiconductor integrated circuits, step 200 described in the first embodiment is performed.
To 206 are executed. Next, the potential images acquired in step 202 and step 206 are compared to detect a shift in the center coordinates of the potential image between the defective product and the non-defective product, which is usually caused by a chip position shift or a difference in inclination when packaging each chip. (Step 207), the coordinates of the potential images of non-defective products are matched by shifting the coordinates (Step 210), and the positional shift between the potential images is further adjusted so that the corresponding area of the image is displayed at the same position (Step 210). 211).

In step 211, the "parameter for deforming the image" is a parameter for adjusting the positional deviation, and is a deviation angle and a deviation amount of the central coordinates.

The potential images of non-defective products and defective products are alternately displayed on the same screen (step 208), and it is confirmed whether or not there is an area where pixels of different brightness are concentrated between non-defective products and defective products indicating disconnection (step 209). . If there is an area where pixels having different brightnesses are different between a non-defective product and a non-defective product in the potential image, it is determined that there is a wire connected to the disconnection in the image (step 224), and the process ends. If there is no wiring connected to the disconnection, the irradiation position of the electron beam is changed (step 223), and the process returns to step 200 to repeat the above operation.

An image processing method according to the second embodiment will be specifically described below with reference to FIGS. FIG. 3A is a schematic diagram of a potential image in a non-defective product in which a constant voltage is applied from a signal input line and a repetitive pulse voltage is supplied from a power supply line and a ground line in FIG. FIG. 3B is a schematic diagram of a place where a defective product is present but no disconnection exists in the potential image.

FIG. 5A shows an image processing in which the beam irradiation area is moved so that FIG. 3A is overlapped with the center coordinates of FIG. 3B, and furthermore, the distortion of the potential image is adjusted to that of FIG. 3B. It is a schematic diagram of the applied potential image. The following method is applied to the sample LSI including the disconnection shown as a schematic diagram in FIG.

First, a pulse generator 103 shown in FIG. 1A applies a pulse voltage applied to a power supply line between 5 V and 2.5 V to a defective semiconductor integrated circuit having a disconnection therein. , And the pulse voltage applied to the ground line is set to swing between 2.6V and 0V. The pulse frequency is set to 2000 Hz, and the phase difference between the pulses applied to the power supply line and the ground line is set to 90 degrees.

FIG. 5 is a diagram showing a potential image of a non-defective product by detecting the number of arrivals of secondary electrons within 15 μsec after the rise or fall of the pulse by using the repetitive voltage pulse as a trigger, and further improving the SN ratio by integrating repetitive signals. 3
(A) is created. By applying a voltage to the defective semiconductor integrated circuit under the same conditions to obtain a potential image and then performing image processing to display an image at the same position as the potential image of the defective product, as shown in FIG. 5A, which is a potential image at the same position coordinates as in FIG. In the image comparison between FIGS. 5A and 5B, it is determined that there is no disconnection, and the electron beam irradiation area is moved so as to display an adjacent area.

As shown in FIG. 5C, when there is a disconnection in the potential image, the disconnection is visually detected by alternately displaying FIGS. 5A and 5C. Alternatively, disconnection can be detected in the potential image by performing automatic determination by software.

Further, as a third embodiment of the present invention,
The potential images of non-defective and non-defective products are obtained by the method described in the first embodiment, and the potential coordinates of the non-defective and non-defective products are displayed on the same screen by the method described in the second embodiment. After performing image processing to display, the brightness distribution of the potential image of a good or defective product is shifted so that the difference in the brightness distribution of the pixels is minimized, and alternately displayed on the same screen to make the broken part more Make it easier to detect.

In order to easily identify a broken portion based on the potential images of good and defective products, the brightness change in the image of the broken portion caused by applying a voltage pulse is caused by the change in the brightness of the image in the other region. Create an image that is larger than the brightness change at.

Next, a method for detecting disconnection in the third embodiment will be described with reference to FIG. The hatched portions are different from those in FIG. 4, and FIG. 6 is a flowchart showing the disconnection detection method in the third embodiment.

First, in the case of defective and non-defective semiconductor integrated circuits, step 2 described in the second embodiment of the present invention is performed.
Execute from 00 to 211. Next, the brightness distribution (contrast) of the defective product is shifted so as to display a potential image in which the brightness difference between the pixels is the minimum except for the broken portion (step 21).
2). The potential images of non-defective products and defective products are alternately displayed on the same screen (step 208), and it is confirmed whether or not there is an area in which pixels having different brightnesses in non-defective products and defective products exhibiting disconnection are concentrated (step 209).

In the potential image, if there is an area where pixels having different brightnesses are different between the non-defective product and the non-defective product, it is determined that there is a wire connected to the disconnection in the image (step 224), and the processing is performed. finish. If there is no wiring connected to the disconnection, the irradiation position of the electron beam is changed (step 223), and the process returns to step 200 to repeat the above operation.

An image processing method according to the third embodiment will be specifically described below with reference to FIG. FIG. 7A is a schematic diagram of a potential image in a non-defective product in which a constant voltage is applied from a signal input line and a repetitive pulse voltage is supplied from a power supply line and a ground line in FIG. FIG. 7B is a schematic diagram of a potential image in a defective product. The following method is applied to a sample LSI including a disconnection schematically shown in FIG. 7B to detect a disconnection location.

First, the pulse voltage applied to the power supply line by the pulse generator 103 shown in FIG. The pulse voltage applied to the ground line is set so as to swing between 2.6V and 0V. The pulse frequency is set to 2000 Hz, and the phase difference between the pulses applied to the power supply line and the ground line is set to 90 degrees.

A repetitive voltage pulse is used as a trigger to detect the number of arrivals of secondary electrons within 15 μsec after the rise or fall of the pulse. Further, the repetition signal is integrated to improve the S / N ratio. FIG.
(B) is created. By applying a voltage to the non-defective semiconductor integrated circuit under the same conditions to obtain a potential image and performing image processing to display an image at the same position as the potential image of the defective product, FIG. 7A, which is a potential image at the same position coordinates as in FIG.

FIG. 7 shows a case where a disconnection exists in the potential image.
The disconnection was visually detected by alternately displaying (A) and (B). Alternatively, disconnection is detected in the potential image by performing automatic determination by software.

Hereinafter, a fourth embodiment of the present invention will be described with reference to the drawings. As in the first embodiment, the disconnection detecting device and method schematically shown in FIG. 1 are used. An image processing method for easily detecting a disconnection will be described with reference to FIG. The hatched portions are different from those in FIG. 6, and FIG. 8 is a flowchart showing the disconnection detecting method according to the fourth embodiment.

First, steps 200 to 212 described in the third embodiment are executed for defective and non-defective semiconductor integrated circuits. A difference image from the potential image is created (Step 600).

Next, in the difference image created in step 600, it is determined by software that visually or automatically determines whether there is a pixel whose brightness is not intermediate brightness (gray) at a certain density or higher (step 601). If there is no wiring connected to the disconnection, the irradiation position of the electron beam is changed (step 223), and the process returns to step 200 to repeat the above operation. The search is continued until a region where pixels having different brightnesses are concentrated in the potential image is found.

The features of the fourth embodiment will be described with reference to FIG. FIG. 9A is a schematic diagram of a difference image between a potential image of a defective product having no disconnection in the observation region and a potential image of a good product, and FIG. 9B is a defective product including a disconnection portion 304 in the observation region. And a potential image of a non-defective product.

In the disconnection detecting method according to the third embodiment,
As shown in FIGS. 7A and 7B, there are three or more lightness areas in the potential image, and it is easy to discriminate a broken part from another area. Visibility is not good because of brightness. In the image processing in this example, as shown in FIG. 9 (B), only the area on the wiring directly connected to the disconnection point is displayed with a different brightness from the other areas, so that it is necessary to specify the disconnection point. It can be specified in a shorter time than in the first embodiment.

In the disconnection detection method according to the present embodiment, whether there is a pixel having a brightness higher or lower than the intermediate brightness, which is the brightness specific to the disconnection, at a density equal to or higher than a certain value in the difference image is visually or automatically determined by software. By doing
It is possible to find a pixel having a brightness specific to the disconnection in the potential image.

The method for detecting a disconnection in a semiconductor integrated circuit according to the fifth embodiment of the present invention is the same as that of the difference image obtained by performing the image processing shown in the second to fourth embodiments except for the disconnection. By creating two images in which only the disconnection is displayed in black or white in brightness, and alternately displaying them on the same screen, it is possible to more easily detect the disconnection location.

Hereinafter, an image processing method for easily detecting disconnection will be described with reference to FIG. The hatched portions are different from those in FIG. 8, and FIG. 10 is a flowchart of the disconnection detecting method according to the fifth embodiment.

First, steps 200 to 600 described in the fourth embodiment are executed on the defective and good semiconductor integrated circuits, and the brightness different from the intermediate brightness specific to the broken portion is determined based on the difference image. Are created by displaying all the pixels having the color of black or white with pixels of brightness (step 800). In the two images, pixels whose brightness is not intermediate brightness (gray) exist at a certain density or higher. The presence or absence of the area is determined by software that visually or automatically determines (step 801).

When there is no wiring connected to the disconnection, the irradiation position of the electron beam is changed (Step 223), and Step 2 is performed.
Returning to 00, the above operation is repeated. In the potential image,
The search is continued until a region where pixels having different lightness exist at a certain density or higher is found.

The features of this embodiment will be described with reference to FIG. FIG. 11A shows a broken portion 30 in the potential image.
FIG. 7B is a schematic diagram of an image in which the brightness of a broken portion is minimized in the difference image between FIG. 7B which is a schematic diagram of a defective product including No. 4 and FIG. 7A which is a schematic diagram of a potential image of a non-defective product; FIG. 11 (B)
5B is a schematic view of a defective product including a broken portion 304 in a potential image, and FIG. 5A is a schematic diagram of a potential image of a non-defective product.
FIG. 5 is a schematic diagram of an image in which the brightness of a broken portion is maximized in the difference image of FIG.

First, by applying the method of acquiring a potential image described in the second embodiment to a semiconductor integrated circuit which is a defective product, the position coordinates at which the potential image is displayed for a good product and a defective product are determined. A potential image with the smallest deviation was obtained. Next, by applying the method for acquiring a potential image described in the third embodiment to a non-defective semiconductor integrated circuit, a potential image having a minimum difference in brightness distribution between a non-defective product and a defective product was obtained.

A difference is calculated for each pixel of the potential image between a good product and a defective product, and when the brightness is not the intermediate brightness, the brightness of the pixel is changed to the maximum brightness or the minimum brightness, thereby obtaining FIG.
And (B). By performing visual or software automatic discrimination by visual or software to determine whether pixels with lighter or darker brightness than the intermediate brightness, which is the specific brightness of the disconnection in the difference image, are present at a certain density or higher, the brightness specific to the disconnection in the potential image is obtained. Pixels can be found.

In the disconnection detecting method according to the fourth embodiment, as shown in FIG. 9B, the brightness distribution in the potential image is very localized at the intermediate brightness, and the broken portion exists in the lower wiring. In such a case, it is difficult to distinguish it from other areas. In the image processing in this example, as shown in FIGS.
Since only the wiring region 305 directly connected to the semiconductor element 4 and the semiconductor element is displayed with white and black lightness other than the intermediate lightness, the visibility of the disconnected portion is improved as compared with the fourth embodiment.

[0076]

According to the present invention, the same voltage can be supplied to each of the signal line, the power supply line, and the ground line, and a wiring jig for giving a potential change inside the LSI can be provided. Preparation time is less than 1 /.

Further, there is no need for an LSI tester for giving a potential change to the wiring inside the LSI, and there is no need for a high-performance WS for performing an image processing operation using layout data every time the screen is moved. In addition, since it is only necessary to have an ability to control the electron beam tester, the system can be simplified. Further, the time for separately acquiring the potential images of the non-defective product and the defective product can be shortened, and the broken portion can be specified in a short time.

Further, in comparison of potential images obtained by simply applying a repetitive pulse voltage, a broken portion which is displayed at intermediate brightness and has poor visibility because no potential change occurs,
Conversely, by performing image processing such that only the disconnection location has a significant change in brightness, the disconnection location can be easily detected.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an example of an LSI failure analysis apparatus according to the present invention (A), a partially enlarged view of an apparatus for moving and exchanging non-defective and defective products (B), and a schematic diagram of applied voltage (C). is there.

FIG. 2 is a flowchart illustrating an algorithm of a disconnection detection method according to the first embodiment.

3A is a schematic diagram showing an enlarged potential image in a non-defective LSI 113 shown in FIG. 1, FIG. 3B is a schematic diagram showing an enlarged potential image in a defective LSI 114, and FIG. 3B is a schematic diagram showing no broken line. C).

FIG. 4 is a flowchart illustrating an algorithm of a disconnection detection method according to the second embodiment.

FIGS. 5A and 5B are schematic diagrams of potential images obtained by performing image processing so as to indicate the same position coordinates as the potential image of a good product in the potential image of a defective product shown in FIGS. 3B and 3C; C).

FIG. 6 is a flowchart illustrating an algorithm of a disconnection detection method according to the third embodiment.

FIG. 7A is a schematic diagram of a potential image obtained by performing image processing on a potential image of a non-defective product so as to show the same brightness distribution as that of a defective product, and a potential image of a defective product in which a broken portion exists in the potential image; (B) of FIG.

FIG. 8 is a flowchart illustrating an algorithm used in the fourth embodiment.

9 is a difference image (A) between FIG. 5 and FIG. 7 (A), and a difference image (B) between FIG. 7 (A) and FIG. 7 (B).

FIG. 10 is a flowchart illustrating an algorithm of a disconnection detection method according to the fifth embodiment.

FIG. 11 is a schematic diagram (A) obtained by changing the brightness of FIG. 9 (B) based on FIG. 7 (A) to that of FIG. FIG. 7B is a schematic diagram (B) in which brightness is changed to be displayed in white.

[Explanation of symbols]

 REFERENCE SIGNS LIST 100 Sample LSI 101 SEM 102 Electron beam tester 103 Pulse generator 104 Constant voltage power supply 105 Exchanger between good and bad products 109 Primary electron beam 110 Electron gun 111 Workstation (WS) 112 Display showing potential image 113 Good LSI 114 Defective product LSI 118 storage medium

Claims (8)

[Claims] 1. A semiconductor integrated circuit having a CMOS structure is irradiated with an electron beam while applying a constant voltage to a signal line and applying a pulsed voltage to a power supply line and a ground line, respectively. A disconnection failure detection device for a semiconductor integrated circuit, wherein a secondary electron signal due to irradiation is detected, a potential image of a wiring of the semiconductor integrated circuit is obtained, and a disconnection failure is detected from the potential image. Means for transferring and exchanging a non-defective product and a defective product of the circuit to the irradiation part of the electron beam to obtain each of the potential images; and a means for alternately displaying the potential images on the same screen. Device for detecting a disconnection failure of a semiconductor integrated circuit. 2. A semiconductor integrated circuit having a CMOS structure is irradiated with an electron beam while applying a constant voltage to a signal line and applying a pulse voltage to a power supply line and a ground line, respectively. A method of detecting a disconnection failure of a semiconductor integrated circuit, wherein a secondary electron signal due to irradiation is detected, a potential image of a wiring of the semiconductor integrated circuit is obtained, and a disconnection failure is detected from the potential image. Moving and exchanging a non-defective product and a defective product of the circuit with the electron beam irradiation unit to obtain each of the potential images; and alternately displaying each of the potential images on the same screen. A method for detecting a disconnection failure of a semiconductor integrated circuit, comprising: 3. The method according to claim 2, wherein a disconnection point of the defective product is detected based on a brightness difference of each of the potential images. 4. The method according to claim 1, further comprising the step of adjusting one of the irradiated areas so as to minimize the difference between the irradiated areas of the non-defective product and the defective product with the electron beam and acquiring the potential image. 4. The method for detecting a disconnection fault of a semiconductor integrated circuit according to claim 2 or 3. 5. The method according to claim 1, further comprising the step of adjusting the brightness of one of the potential images so as to minimize the brightness difference between the potential images of the non-defective product and the defective product and acquiring the potential image. 4. The method for detecting a disconnection failure of a semiconductor integrated circuit according to 2 or 3. 6. The method according to claim 1, further comprising the step of generating a difference image between the potential images of the non-defective product and the defective product, wherein the disconnection portion of the defective product is detected based on the presence or absence of an intermediate brightness area in the differential image. A method for detecting a disconnection fault in a semiconductor integrated circuit according to claim 2. 7. A step of generating a difference image between the potential images of the non-defective product and the defective product, and a step of changing the brightness of the difference image, wherein the broken portion of the defective product is determined by the brightness-changed difference image. 6. The method according to claim 2, wherein the disconnection is detected. 8. A semiconductor integrated circuit having a CMOS structure is irradiated with an electron beam while applying a constant voltage to a signal line and applying a pulse voltage to a power supply line and a ground line, respectively. A method for detecting a disconnection failure of a semiconductor integrated circuit, in which a secondary electron signal due to irradiation is detected to obtain a potential image of a wiring of the semiconductor integrated circuit, and a disconnection failure is detected from the potential image by a computer. A recording medium on which a control program is recorded, wherein a non-defective product and a defective product of the semiconductor integrated circuit are moved to and exchanged with an electron beam irradiating unit to obtain each of the potential images; and Recording a program including alternately displaying on the same screen.