JPS60167341A - Detection for fine pattern - Google Patents

Abstract

PURPOSE:To enable to finely calculate the pattern width dimension, the pattern space dimension and so forth of a pattern, and at the same time, to prevent the complication of a pattern detecting device by a method wherein the second order differential signal of the pattern is calculated by differentiating the level electric signal found from the optical profiles of the pattern, and also, the pattern edges of the pattern are detected by finding the zero point positions of the two-story differential signal. CONSTITUTION:A pattern is irradiated by a pattern illuminating system 25 and the pattern is imaged on a CCD33 and a CCD34. As a result, an optical profile corresponding to the pattern configuration can be respectively detected in the CCDs 33 and 34, and the electric signal of voltage level is detected, being accompanied with this detection. This level electric signal S1 detected is differentiated in a first arithmetic circuit 38, and furthermore, that is differentiated in a second arithmetic circuit 39 and a two-story differential d<2>V/dl<2> signal S3 is found. When zero points in the case of d<2>V=0 of this two-story differential signal S3 are found by a third arithmetic circuit 40, the zero point positions Pa, Pb, Pc and Pd can be calculated. These zero point positions Pa, Pb, Pc and Pd are the maximum and minimum positions of the characteristics of the pattern shown in the diagram (D), that is, are each both end positions of the pattern edge parts 1a and 1a.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to a method for detecting a fine pattern, and particularly to a method for detecting a fine pattern that can detect the width of a fine pattern formed on a semiconductor wafer with high precision.

[Background technology]

In semiconductor manufacturing technology, it is necessary to control the manufacturing process and quality by measuring the pattern, especially the pattern width, formed on a photomask or semiconductor substrate. Therefore, it is conceivable to rely on the human eye using a microscope for this measurement, but there are problems with individual errors and reproducibility, and there is also the problem of causing worker fatigue. For this reason, measurement devices using ordinary light or laser light have been developed to solve the above-mentioned problems. For example, in the case of -, the device scans and irradiates a pattern with light and simultaneously measures the intensity of the reflected light, thereby detecting the position of the pattern edge from the unique reflection state at the edge of the pattern and determining the pattern width. One possible method is to calculate it. Another possible method is to illuminate a pattern, form a pattern image on an optical element such as a COD, and calculate the pattern edge portion and pattern width from changes in the output of the COD.

Incidentally, although there are differences in optical means among the above-mentioned methods, they are the same in that the detected electrical signals are processed in accordance with a predetermined method when detecting pattern edges. For example, when the signal shown in FIG. 1 is obtained, pattern edge 1.1 is detected (calculated) from this signal waveform.

Conventionally, various methods have been used for this signal processing, one of which is the M2 diagram (as in Al, the signal is processed by software to linearize the waveform, and the intersection point P1 of this and an appropriately set threshold value is There is a method of calculating the pattern edge from the average of ~P4.Also, after differentiating the signal as shown in the same figure (Bl), find the intersection P with the large and small thresholds, ~P, and calculate the intersection of this point. There is also a method that uses averages.

However, since each of these methods utilizes the intersection of a signal and a threshold value, if the threshold value is not set appropriately, there is a risk that detection accuracy may decrease or detection may become impossible. Further, when the threshold value changes, the detected value also changes, and the reproducibility of detection also deteriorates. For this reason,
Measures such as threshold setting and fluctuation prevention are also required, which poses the problem of complicating the measuring device.

[Purpose of the invention]

An object of the present invention is to detect pattern edges accurately and with high precision, thereby making it possible to precisely calculate pattern width dimensions, pattern space dimensions, etc., and to detect fine patterns without complicating the device. The purpose is to provide detection technology.

The above and other objects and novel features of the present invention include:
It will become clear from the description of this specification and the accompanying drawings.

[Summary of the invention]

A brief overview of typical inventions disclosed in this application is as follows.

That is, the second-order differential signal is calculated from the level signal determined from the optical profile of the pattern, and the pattern edge is detected by setting the zero point position of this second-order differential signal.Thus, the threshold value can be set. This eliminates all the above-mentioned problems related to threshold values.

〔Example〕

FIG. 3 shows a pattern detection apparatus for carrying out the method of the present invention. First, this apparatus will be explained, and then the method of the present invention will be explained together with its operation.

In the figure, a conductor wafer 10 having a pattern to be detected is placed on an XY table 12.2 tables 13. It is mounted on a movable table 11 consisting of a θ table 14, and each table drive unit (not shown) moves x, y, z, θ.
direction. A laser length measuring device 16 equipped with a mirror 15 is disposed on one side of the movable table 11 to measure each table position.

A pattern detection unit 17 is provided directly above the wafer 10 to obtain a level electrical signal based on the optical profile of the pattern on the wafer 10. This pattern detection section 17 has an objective lens 18 with a high NA (numerical aperture) placed at a position facing the wave/.
0. The half mirror 19 in which the half mirrors 21 and 22 are arranged in series is further provided with a half mirror 23, and a focusing illumination system 24 is provided laterally to the half mirror 23.
A pattern illumination system 25 is arranged above.

The focal illumination system 24 includes a lamp 26. Infrared transmission filter 2
7. The pattern illumination system 25 includes a slit 28 and a lamp 29 and an infrared cut filter 30. Further, a monitor TV camera 31 is arranged opposite to the half mirror 21.

Further, the half mirror 22 includes No.-7 mirrors 32 and 3.
A focus/dimension detection system 36 having C0Ds 33, 34, and 35 is provided. These three CCDs 33, 34,
35, the pattern on the wafer 10 is imaged on the surface by the objective lens 18 and the relay lens 20, and the CC
D33 and D34 detect the pattern edges in the X direction and Y direction, respectively, and the CCD 35 detects the focal point of the pattern.

These CCDs 33, 34, and 35 are connected to a control system 37. The control system 37 includes a first arithmetic circuit 38 that performs first-order differentiation, a second arithmetic circuit 39 that differentiates the output of this first arithmetic circuit 38 again, that is, performs second-order differentiation, and a second arithmetic circuit 39 that performs second-order differentiation. A third arithmetic circuit 40 is provided to find the zero point of a certain second-order differential value. The control system 37 also includes a main calculation circuit 41 that performs other calculations and a table drive control circuit 4.
It also has 2. The control system 37 is connected to the laser length measuring device 16 and a table drive unit (not shown).

According to the above configuration, the focusing illumination system 24 illuminates the surface of the wafer 10 prior to detecting the pattern, and detects the pattern by C.
After the image is formed on the CD 35, the two-table position is controlled by the control system 37 using the sharpness of the output of the CCD 35, and the focal position is set.

Next, the pattern is irradiated by the pattern illumination system 25, and the pattern is imaged onto the CCDs 33 and 34. This results in C0D
At 33 and 34, an optical profile corresponding to the pattern shape can be detected, and along with this, an electrical signal at a voltage level is detected. Now, if we look at CCD33 in the X direction,
Figure 4 (Plane and cross-sectional pattern I of (Al, (Bl)
If A is detected, the edge portion 1a of pattern IA,
In Ia, since the reflected light directly upward is reduced, it is possible to obtain an electrical signal with a voltage level having characteristics as shown in the figure (C1).

Therefore, this detected level blood/blood signal S.

By differentiating this in the first arithmetic circuit 38, we can obtain d V 7 , signal S2 as shown in the figure (DI), and further differentiating this in the second arithmetic circuit 39, we obtain 2V fa as shown in the figure. The second derivative /, 12 signals S3 can be set. Then, d”V of this second-order differential signal Ss
= If point o is determined by the third arithmetic circuit 40, the zero point position Pa
, Pb, Pc, and Pd can be calculated. These zero point positions Pa and Pb.

Pc and Pd are the maximum and minimum positions P a, P b', P c', and P d' of the characteristics shown in (1) l of the same figure, and furthermore, these are the change of the characteristics of (C) of the same figure. Curved point position Pa, Pb
, Pc.

Pd, that is, the positions at both ends of the pattern edge portions 1a and la. Therefore, in this example,
By measuring the distance between the zero point positions Pa and Pd using the positions of the corresponding picture elements on the CCD 33, the pattern I is created.
The width dimension of A can be adjusted. The same applies to the pattern space, and the same applies to the pattern in the Y-axis direction.

Therefore, there is no need to compare the detected signal with the threshold value, and there is no detection error caused by fluctuations in the threshold value. It is also possible to detect the U ht degree.

〔effect〕

(1.1 By second-order differentiating the level electrical signal obtained from the optical profile of the pattern to be inspected and calculating the zero point of this second-order differential, the edges of the pattern can be detected and the pattern width and pattern space width can be calculated. Comparison with a threshold value can be made unnecessary, various problems caused by the threshold value can be prevented, and accurate detection can be performed.

(2) Since the signal is obtained by second-order differentiation of the level electrical signal, even if the peak of the level electrical signal is gradual, it can have a steep peak characteristic, and pattern edge portions can be detected with high accuracy based on the zero point position. I can do it.

Although the invention made by the present inventor has been specifically explained above based on examples, it is to be understood that the present invention is not limited to the above-mentioned examples, and can be modified in various ways without departing from the gist thereof. not. For example, the optical element for detection may be a photo sensor other than a CCD, an image pickup tube, or may be configured to use laser light. Also,
If an arithmetic circuit that calculates the second-order differential at once is used, the first arithmetic circuit can be omitted.

[Application field]

In the above description, the invention made by the present inventor was mainly applied to the detection of patterns on semiconductor wafers, which is the background field of application of the invention, but the invention is not limited thereto.

It can be applied to anything that requires detection of alignment marks in mask aligners and other pattern detection.

[Brief explanation of drawings]

Figure 1 is a diagram showing an example of a level electrical signal, Figure 2 (A1
．． (Bl is an explanatory diagram of each different detection method, FIG. 3 is an overall perspective view of an apparatus for carrying out the method of the present invention, FIG. 4 is a diagram explaining the method of the present invention, (
A). (Bl is the plan view and cross-sectional view of the pattern, (C1~1 is (
It is a characteristic diagram made to correspond to a Bl diagram. IA...pattern, la, la...edge part, 10
...Wafer, 11...Movable table, 16...
Laser length measuring device, 17... pattern detection section, 18...
Objective lens, 24... Focusing illumination system, 25... Pattern illumination system, 33-35... CCD (chemical element), 3
6... Focus/dimension detection system, 37... Control system, 38.
...First arithmetic circuit (-differential), 39...Second arithmetic circuit (second-order differential), 40...Third arithmetic circuit (zero point), P
a”Pd...Zero point position. Agent Patent attorney Akio Takahashi Figure 1! Figure 2 Figure 3 Figure 7 Procedural amendment (method) Display of case 1982 Patent Application No. 21621 Amendment person 1i11 Patent applicant name 4510+ 1J Shikikai 1 1] Specification page 11, lines 7 to 9 [Figure 4 (5) to ■)]
It is... " is corrected to "FIG. 4 is a diagram explaining the method of the present invention."

Claims (1)

[Claims] 1. Determine the second-order differential signal from the level electrical signal obtained from the optical profile of the detected pattern, calculate the zero-point position of this second-order differential signal, and measure the pattern dimensions based on this zero-point position. A fine pattern detection method characterized by: 2. Image the pattern to be measured on an optical element such as COD,
2. The fine pattern detection method according to claim 1, wherein a level electrical signal is obtained by the output of the COD. 3. The fine pattern detection method according to claim 1 or 2, wherein the pattern width and pattern space dimension are determined based on the dimension between selected zero point positions of a plurality of second-order differential signal positions.