JP2917948B2 - LSI short-circuit failure detection method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an LSI inspection, and more particularly, to a non-contact electrical inspection method such as a short circuit of a wiring.

[0002]

2. Description of the Related Art Conventionally, LSI inspection is performed by a method in which a probe is mechanically brought into contact with a device to be inspected to conduct an electrical inspection such as continuity or electric potential, or an LSI pattern is optically inspected to detect a shape defect. There was a way to detect it.

However, if this conventional electrical inspection method is used during the LSI manufacturing process, there is a problem that the contact of the probe becomes difficult with the miniaturization of the LSI and the element is destroyed by the contact. In the optical inspection method,
Since the detected shape abnormality does not always lead to the LSI electrical failure, the normal electrical function of the LSI is not guaranteed even if the cause of the shape abnormality is removed. In addition, there is a limit to the resolution to be detected, and it is difficult to detect a defect caused by a three-dimensional structure because it is a planar inspection, and it is also difficult to detect a leak failure of a thin film.

In addition to these methods, there is also a method of knowing the wiring potential using an EB tester. However, in this case, since an electric signal must be supplied from the outside to the wiring, it cannot be applied to the inspection during the manufacturing.

[0005]

SUMMARY OF THE INVENTION In view of the foregoing, an object of the present invention is to provide a method for inspecting an LSI wiring failure that enables an electrical inspection to be performed in a non-contact manner even during a manufacturing process. To provide.

[0006]

It is known that when the amount of secondary electrons emitted by irradiation of a charged particle beam is observed, the observed value differs depending on the potential of the sample to be observed. That is, a portion with a high potential has a small number of observed secondary electrons, and a portion with a low potential has a large number of observed secondary electrons. According to the present invention, an electric charge is injected by a charged particle beam as a means for giving an electric potential, and a short circuit is given an electric potential by injection into a short circuit destination. In this case, different potentials are applied to a normal portion and a short-circuit failure portion of the element, and this is detected.

[0007] LSI short-circuit failure detection method of the present invention, charged
Continuous irradiation of the particle beam across multiple wires
Near the opposing edges of two adjacent wires
Two wirings whose difference in secondary electron emission amount is smaller than a certain value
It is determined that there is a short circuit between them.

[0008]

[0009] Two kinds of charged particle beams having different potentials given by charge injection by irradiation of the charged particle beam are successively irradiated successively across a plurality of wirings.
There is a short circuit between two wirings in which the difference in the amount of change caused by using two types of charged particle beams in the amount of secondary electron emission near the opposite edges of two wirings adjacent to each other is within a certain range. it is determined that.

A first charged particle beam is pulse-irradiated so as to traverse a plurality of wirings and to irradiate each wiring at least once, and to traverse the plurality of wirings at each time interval of the pulse irradiation. The secondary electron emission amount is observed while continuously irradiating the second charged particle beam, and the peak of the fluctuation of the secondary electron emission amount during the irradiation of the second charged particle beam performed before and after the time interval is directly or jumping over lacking portion of the wiring, it is determined that mutually has a short circuit between two wires observed in both of the adjacent wires.

[0011]

Embodiments of the present invention will now be described with reference to the drawings. (First Embodiment) FIG. 1 is an explanatory view of a first embodiment of the present invention, and shows a method of detecting a short circuit failure of a gate electrode wiring after forming a gate electrode wiring in an LSI manufacturing process. .

The gate electrode wiring has, for example, a thickness of 1
Microns, width 1 micron, spacing 2 microns.
The parasitic capacitance of one gate electrode wiring is 4 picofarads. As shown in FIG. 1A, the electron beam is scanned from left to right at right angles to the gate electrode wiring, and at this time, a beam current of 500 picoamps is applied at an acceleration voltage of 3 KV, for example, so that the electron beam does not penetrate. I do.

[0013] When the insulating film is scanned and irradiated, the charge injection amount of the electron beam is the same at any place, so that the potential becomes uniform at any place and the observed amount of secondary electrons becomes uniform. However, when scanning over the wiring, the injected charge is diffused throughout the wiring, and when scanning over the same wiring, the integrated injection amount of the electron beam increases from left to right in the wiring. The potential change increases. To obtain a potential change of 1 V, scanning is performed at a scanning speed at which scanning of one gate electrode wiring is completed in about 8 milliseconds. At this time, the amount of observed secondary electrons is as shown in FIG.

That is, when the electron beam is scanned from left to right and irradiated, the amount of injection of the electron beam is directly reflected on the potential on the insulating film and the observed amount of secondary electrons is uniform.
At the left end of the wiring, the injected electron beam is diffused and the fluctuation of the potential is small, and the initial potential of the wiring is reflected. Therefore, the observed amount of secondary electrons at this location decreases more rapidly than on the insulating film. On the same lines, the so integrated injection charge of the whole line is increased potential is gradually lowered secondary electrons view as scanning in the right direction proceeds surveying gradually increases. When the right end of the wiring is illuminated, the accumulated injected charge becomes maximum and the secondary electrons
Watched survey is maximized.

Even when the second wiring is irradiated, when the left end is irradiated, the initial potential of the wiring is reflected, so that the observed amount of secondary electrons decreases more sharply than on the insulating film. The difference in the observed amount of secondary electrons when irradiating the left end of the second wiring is larger than the amount of observed secondary electrons when irradiating the right end of the first wiring on the left side. However, the amount of secondary electrons observed when the left end of the third wiring is illuminated is 2
Since the first wiring and the third wiring are short-circuited, the potential when the right end of the second wiring is irradiated is reflected, and is substantially the same value as when the right end of the second wiring is irradiated. .

Utilizing this, the electron beam is scanned from left to right to illuminate the right end of the wiring and illuminate the secondary electron observation amount and the left end of the wiring adjacent to the right. When the absolute value of the difference is equal to or less than a certain value, it is determined that the wiring is short-circuited. FIG. 1 (c)
Figure 2 shows the difference in the amount of observed secondary electrons. Since the difference is smaller at the third wiring among the four wirings, it can be detected that there is a short circuit between the second and third wirings. (Second Embodiment) FIG. 2 is an explanatory diagram of a second embodiment of the present invention. This is also effective when the initial value of the potential of each wiring on the LSI is not always uniform.

As in the first embodiment, the electron beam is scanned from left to right at right angles to the gate electrode wiring. In the first scan, the secondary beam is scanned with a relatively small beam current of, for example, 100 picoamps. The amount of electron emission is observed (a). In the second scan, the amount of secondary electron emission is observed while scanning with a relatively large beam current of, for example, 500 picoamps (b). (C) shows the amount of change between the first observed secondary electron quantity and the second observed secondary electron quantity. The amount of change at the right end of the wiring and the amount of change at the left end of the wiring on the right are compared, and when the absolute value of the difference is equal to or less than a certain value, it is determined that the wiring is short-circuited. Since the difference is smaller at the third wiring among the four wirings,
It can be detected that there is a short circuit between the third and third wires.

In this embodiment, correction can be made even when the initial value of the wiring potential is not uniform. (Third Embodiment) FIGS. 3 and 4A and 4B are explanatory views of a third embodiment of the present invention. FIG. 3 shows an electron beam irradiation method, and FIGS. (b) is a flow explanatory diagram.

As shown in FIG. 3, a first charged particle beam is radiated to discrete points on the LSI chip on which the wiring is formed. The secondary electron emission amount is observed while irradiating the second charged particle beam in the time interval of these pulse irradiations. That is, first, a first charged particle beam is applied to one point on a virtual line on the chip, and then secondary electrons are observed while scanning the second charged particle beam linearly. Next, after irradiating the next point on the virtual line with the first charged particle beam, secondary electrons are observed while scanning the second charged particle beam linearly again. In this way, the pulsed irradiation at discrete points is performed while shifting the irradiation point of the first charged particle beam,
The secondary electron emission is observed while irradiating the charged particle beam. The irradiation time or the beam current is selected so that the wiring potential fluctuation caused by the first charged particle beam irradiation is sufficiently larger than the wiring potential fluctuation caused by the second charged particle beam irradiation. At this time, the pulse irradiation of each discrete point of the first beam, the flow of the second beam irradiation, and the second
Fig. 4 which plots the secondary electron quantity observed value by the beam
This will be described with reference to FIG.

First, the emitted secondary electrons are observed while scanning the second beam for the first time so as to cross over the sample in which the second and third wires of the four wires are short-circuited. The observed amount of secondary electrons is uniform, reflecting the initial potential. (Since the current of the second beam is selected to be sufficiently smaller than the current of the first beam, the potential change of the wiring due to the second beam is ignored.) Next, the first beam is placed on the first wiring. Pulse irradiation. Thus, the potential of the first wiring changes. Then 2
Secondary electrons are observed while scanning the second beam for the second time.
In this case, the amount of observed secondary electrons increases only in the first wiring. Next, pulse irradiation of the first beam is performed on the second wiring. Thus, the potential of the third wiring short-circuited with the second wiring changes. Since the wiring capacitance becomes twice as large as that of the first line, the potential change becomes half of that of the first line. Next, secondary electrons are observed while scanning the second beam for the third time. In this case, the amount of secondary electrons observed in the first wiring, the second wiring, and the third wiring becomes large. Next, pulse irradiation of the first beam is performed on the third wiring. As a result, the potentials of the third wire and the second wire short-circuited therewith further change. Next, secondary electrons were observed while scanning the second beam for the fourth time.
The observed amount of secondary electrons of the third and third wirings is large. Next, pulse irradiation of the first beam is performed on the fourth wiring. Thus, the potential of the fourth wiring changes. Next, when the secondary electrons are observed while scanning the second beam for the fifth time, the amount of secondary electrons observed for the first, second, third, and fourth wirings is large. FIG. 4B shows the amount of change in the secondary electron observation value due to each second beam scanning. 2
The first and second observations at the time of the second beam scanning
The fluctuation from the observed value during beam scanning has a peak only in the first wire. The fluctuation of the observed value from the second observed value at the time of the third second beam scanning has two peaks at the second and third wirings. The fluctuation of the observation value at the time of the fourth second beam scanning from the third observation value is also
Peaks appear at two locations of the second and third wires. The fluctuation of the observed value from the fourth observed value at the time of the fifth second beam scanning has a peak only in the fourth wiring. As described above, if there is a short-circuited wire, peaks appear in both of the short-circuited wires, but only one normal wire has a peak. In this example, both the second and third wires have peaks, so that it is possible to detect that the second and third wires are short-circuited. Further, in this example, the example in which the pulse irradiation of the first beam is performed at one place for one wiring is shown. However, it is not particularly necessary to provide one place for each wiring. For example, the pulse irradiation interval smaller than the wiring width is used. If you irradiate with
It is possible to detect a short circuit between all wirings and other wirings.

[0021]

As described above, according to the present invention, the LS can be realized without connecting a power supply or other electric signals from the outside to the LSI.
By irradiating the charged particle beam to the I surface and observing the secondary electrons emitted thereby, a short-circuited portion of the wiring in the LSI can be detected. There is an effect that a short circuit defect can be detected.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of a first embodiment of the present invention.

FIG. 2 is an explanatory diagram of a second embodiment of the present invention.

FIG. 3 is an explanatory view of an electron beam irradiation method according to a third embodiment of the present invention.

FIG. 4 is an explanatory diagram of a flow according to a third embodiment of the present invention.

Claims (3)

(57) [Claims] 1. A method of irradiating a charged particle beam onto an LSI surface and applying a charge by irradiating the charged particle beam.
In an LSI short-circuit failure detection method for detecting a short-circuit between wirings in an LSI by observing the amount of secondary electrons emitted from the LSI surface in accordance with the applied potential , a charged particle beam is caused to cross a plurality of wirings. Irradiate continuously
While near the opposing edges of two adjacent wires.
The difference in the amount of secondary electron emission in the two arrangements is smaller than a certain value.
An LSI short-circuit failure detection method, comprising determining that there is a short circuit between lines . 2. A method for irradiating a charged particle beam onto an LSI surface.
And given by charge injection by irradiation of a charged particle beam.
Secondary electrons emitted from the LSI surface according to the applied potential
By observing the amount of emission, the amount of wiring between LSI
In the LSI short-circuit failure detection method for detecting short-circuits, the short-circuit is given by charge injection by irradiation of a charged particle beam.
Two types of charged particle beams with different potentials
Irradiates continuously across a plurality of wirings, and observes near the opposing edges of two wirings adjacent to each other.
The two types of charged particle beams of secondary electron emission measured
The difference in the amount of change caused by using
That there is a short circuit between the two wirings
LSI short-circuit failure detection method. 3. A method of irradiating a charged particle beam onto an LSI surface.
And given by charge injection by irradiation of a charged particle beam.
Secondary electrons emitted from the LSI surface according to the applied potential
By observing the amount of emission, the amount of wiring between LSI
In an LSI short-circuit failure detection method for detecting a short circuit, one or more illuminations are performed across a plurality of wirings and on each wiring.
Pulse irradiation of the first charged particle beam so that irradiation is performed
And traverses the plurality of wirings at each time interval of the pulse irradiation.
As shown in the figure, the secondary electron beam
Observing the release amount, the second charged particle beam performed in the preceding and following time gaps
Of fluctuation of secondary electron emission during irradiation
It is a jump over lacking part of the direct or wiring, adjacent to each other
There is a short circuit between the two wires observed on both of the wires
A short circuit detection method for an LSI.