JPS62237303A - Measuring instrument for mask mismatching - Google Patents

Abstract

PURPOSE:To easily measure line intervals of a pattern to be inspected in a short time by providing a means which picks up the image of the body to be inspected, a means which decides the shift mask matching based on an edge signal. CONSTITUTION:A microscope 11 is provided over a wafer 10, and an ITV camera 12 is fitted and outputs the image signal of the wafer 10. This image signal is converted 13 into a digital image signal, which is stored in a video memory 14. Then, the image data is inputted to the spatial filter circuit 15 composed of the 1st spatial filter circuit 16 which performs linear differentiation and the 2nd spatial filter 17 which performs secondary differentiation. Here, the image data is read out in picture element units and differentiated successively to obtain differential data which has a peak value corresponding to the position of a resist edge part. The output data from this circuit 15 is inputted to an information processing part 18 to detect the position of each peak value corresponding to each line position of the differential data, thereby deciding whether there is the shift of mask matching or not on the basis of the linear interval calculation result.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a mask misalignment measuring device used for detecting mask misalignment in, for example, a semiconductor manufacturing process.

(Prior Art) In the semiconductor manufacturing process, there is a process of forming a circuit pattern on a surface, and this circuit pattern is inspected to see if it has been formed abnormally. Figure 13 shows part of the semiconductor manufacturing process, in which a wafer (1) is coated with silicon substrate (Sl) (2) as shown in Figure 13 (a).
An insulating film (3) made of io2 is formed.

First, as shown in FIG. 13(b), a resist (4) is applied to a wafer (1). After this, a mask (5) is applied, exposure is performed, and development is performed (first
Figure 3 (C)). When this development is completed, as shown in FIG. 13(d), a portion (6) of the resist (4) is irradiated with light.
will melt, and the masked area will remain intact. Thereafter, etching is performed as shown in FIG. 13(e).

Here, if the mask (5) is not precisely aligned with the predetermined position, the portion to be etched will be so small that, for example, the position and amount of the impurity to be bonded will be different, and the predetermined properties will not be exhibited. For this reason, after the development process,
A check is made to see if the masked position is accurate. By the way, this inspection was carried out on the edge part of the resist (4) (
This is done by measuring the positions of 7a) and (7b). Note that each line (7a) and (7b) is connected to the wafer (1
) is formed in a linear shape when viewed from the top, so the interval between each line is the tb edge portion (7a).

The position and spacing of (7b) will be measured. Conventionally, this measurement was carried out by an operator using an optical microscope, but recently, laser light has been used to measure the edge portions (7;]), (7b
), or by using an optical slit to measure the edge portions (7a) and (7b) of the resist (4) from interference light.
) is carried out by a method of measuring the position of

(Problems to be Solved by the Invention) However, in each of the above methods, the entire apparatus used is complicated and expensive. Sara: Te.

The above method cannot be applied to online measurement because it takes time to measure one edge portion (7a), (7h). Note that the operator needs to be skilled in order to perform measurements through an optical microscope.

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, it is an object of the present invention to provide a line spacing measuring device that can easily measure line spacing in a short time.

[Structure of the invention]

(Means for resolving interlayer points) The present invention provides an imaging means for imaging an object to be inspected for mask misalignment, and an edge signal indicating mask misalignment based on an image signal obtained by the imaging means. The present invention is intended to achieve the above object by providing an edge signal output means for outputting an edge signal, and a determination means for determining mask misalignment based on the edge signal output from the edge signal output means.

(Function) By providing each of the above-mentioned means, the present invention calculates the distance between a pair of lines indicating the mask misalignment, and determines the mask misalignment based on these line spacings.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram of a line spacing measuring device. (10) is Ueno I after the development process, and this Ueno OQ
There is a microscope office lady on the top! is provided. The microscope αB is equipped with an ITV camera (industrial television ITV camera 0z outputs an image signal of an enlarged image of the sea urchin αC. The ITV camera α2 is equipped with an A/D
Converter (analog/digital converter) 0 is connected,
The image signal is converted into a si-digital image signal by the FiA/D converter C13) and sent to the image memory 041.
! The data is stored as image data. Now, fl, 5) is a spatial filter circuit, which reads out the image data stored in the image memory αa pixel by pixel and sequentially differentiates it, and calculates the differential value with a peak value corresponding to the position of the resist edge. It is used to obtain data. The specific configuration includes a first spatial filter circuit (161) that performs first-order differentiation, and a second spatial filter circuit (17) that performs second-order differentiation. The configuration of the spatial filter circuits ae and ση is as shown in Fig. 2.In other words, the ITV camera (one of the
8-bit shift register 20+ corresponding to scanning
, f211.

(C) are connected in series so that the image data G is taken into the shift register Cυ one pixel at a time. Then, the first 3 bits of aIJ and t2z of each shift register are filled in and moved to the corresponding addresses of the differentiation area (width). Still, (24) is the weight for performing differentiation. A coefficient is stored in each address corresponding to the differential arear 3). Therefore, each time one pixel is read, each brightness level and weighting coefficient of the differentiation area (:) are integrated for each address by the product-sum operation section (to), and then these integrated values are added to form the differential data. It is designed to be sent out as a BD.The α second is an information processing section, which is a spatial filter circuit (receives the differential data output from I9, and processes each peak value corresponding to each line position of this differential data. It has the functions of an interval calculation means that detects the position and calculates the line interval, and a determination means that determines whether there is mask misalignment based on the result of this interval calculation.

Next, the operation of the apparatus configured as described above will be explained.

It is assumed that the wafer α1 is processed into a shape as shown in FIG. 3 based on the phenomenon image. The part (10a) is the part of the resist that was masked and melted during the development process! The portion (10b) is the resist portion that was not masked and melted. Therefore, the microscope αυ is
It is installed so that line B is within the field of view, and the image on line A-B is magnified. This enlarged image is imaged by the TV camera α2, and the image signal is sent to the A/D converter 03, where it is converted into a digital image signal and sent to the image memory α. Here, the image signal sent is one that shows an image on line A-B by scanning. By the way, this A-B
As shown in FIG. 4, the brightness level of the image signal on the line is lower at the edges of the resist than at other parts. Therefore, the image data of this image signal is stored in the image memo IJ (14), and this image data is sequentially read out pixel by pixel and sent to the first spatial filter circuit αe. This first spatial filter circuit (in 11, shift registers ■, I2υ, I27J for each pixel of image data)
The sum-of-products calculation unit (
Then, these integrated values are added together and sent out as differential data j3D. Therefore, the image signal shown in FIG. 4 is differentially processed by the first spatial filter circuit αe to become first-order differential data as shown in FIG. 5.

Although the actual differential processing is performed using image data, the first differential data shown in FIG. 5 is expressed in analog form -C in order to make it easier to distinguish the processing state. Then this one
When the second-order differential processing is completed, the first-order differential data shown in FIG.
It is differentiated to the next degree. As a result, second-order differential data as shown in FIG. 6 is obtained. Now, as shown in Figure 6, this second-order differential data has a peak value of 1) 1.1 corresponding to the position of each line.
)2. p3. p4 appears.

Therefore, the information processing unit αg receives this second-order differential data and receives each peak value 1)1.1)2. p3. The position of 4 is detected, and the intervals e and i between peak values p1 and p2 and p3 and p4 are detected.
'Calculate. Then, these intervals 1 and l' are compared, and if the difference is larger than a predetermined value, it is determined that the mask position is not accurately aligned, and an NG (No Good) signal is sent.

By the way, in the spatial filter circuit α9, the first and second order
The reason why the second order differentiation is performed is as follows.

In other words, for example, as shown in FIG.
As shown in the figure, the interval between the peak values is 12, which is different from the actual interval j1. Therefore, by performing second-order differentiation, second-order image data as shown in FIG. 9 is obtained. The peak value interval 13 of this second-order differential data is the interval 11
is equal to Therefore, first and second order differentiation is performed.

In this way, the green line spacing of the resist is measured at multiple locations other than on line A-B, and the average value of these measured line spacings is calculated to obtain the final line spacing. Note that singular values due to noise etc. are excluded.

Next, the mask misalignment determination in the information processing section (181) is performed as follows. That is, the first
Figure 0 (A), (B) and Figure 11 (A), CB) respectively show the case of "no mask misalignment" and the case of "mask misalignment", but it is difficult to distinguish between the two. On the silicon substrate (2), the pattern to be inspected (Pl)
The determination is made based on whether or not there is a pattern to be inspected (Pl) in the pattern. That is, first, in FIG. 12, the line spacing of the pattern to be inspected (Pl) is Ll, and the pattern to be inspected (Pl) is
), the difference between the two is △L, that is, ∆
Calculate L=L1-L2. In addition, Fig. 11 (A),
CB), the pattern to be inspected (Pl), (Pl
) can be observed because the pattern to be inspected (Pl) is transparent. Next, it is determined whether the line spacings J and l' satisfy the following conditions (formulas 2 and 2) at the same time.

l<△L ... ■ z'<△L ... ■ In other words, as shown in Fig. 11, in the case of "mask misalignment", at least one of the line spacings l and l' is It becomes larger than ΔL. In the case of "Mask misalignment", equations (■) and (2) cannot be satisfied at the same time. In this embodiment, this characteristic is utilized. Therefore, as shown in FIGS. 10 and 12, if equations (1) and (2) are satisfied at the same time, it is determined that there is no mask misalignment. In this way, in the above embodiment, the image R Song data is sent to the spatial filter circuit α9.
Second-order differentiation is performed to create differential data with peak values corresponding to the line positions, and from the interval between these peak values, the line interval I, j' on both sides of the pattern to be inspected (PI), (P2) is calculated.
is calculated, and the presence or absence of mask misalignment is determined based on the result of this line spacing calculation. Therefore, mask misalignment can be determined with a simple configuration and with high accuracy and high efficiency. In particular, since primary and secondary image processing is performed, higher accuracy is achieved. Since the line spacing is mainly measured by image processing, the time required for measurement is quick, and it can be used online for semiconductor fabrication process knowledge. Also 1
It is also inexpensive.

It should be noted that the present invention is not limited to the above-mentioned embodiment, but can be modified in various ways without departing from the spirit thereof. For example, an 8EM may be used instead of the microscope lens, or a normal camera lens may be used. Furthermore, a structure may be adopted in which a smoothing circuit is provided before the spatial filter circuit to remove noise components.

Also, the spatial filter circuit has a
Only the second derivative processing may be performed. This spatial filter circuit may also use an image processing processor, and in this case the cost will be low.

〔Effect of the invention〕

As described in detail above, according to the present invention, it is possible to provide a line spacing measuring device t that can easily measure mask misalignment in a short time.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing an embodiment of the line spacing measuring device according to the present invention, Fig. 2 is a specific block diagram of a spatial filter circuit in the device of the present invention, and Fig. 3 is an external appearance of a wafer to be measured. Figures 4 and 6 are diagrams for explaining the measurement action of the device of the present invention, and Figures 7 to 9 are diagrams for explaining the first and second differential effects of the device A of the present invention. Figure 1O (A), CB) and Figure 11 (A),
CB) is a plan view and z-z, 14-view cross-sectional view, respectively, to explain mask misalignment, Figure 12 (5), μ button is an explanatory diagram of mask misalignment, and Figure 13 is a diagram for semiconductor manufacturing. It is a diagram for explaining the royal power of a part of α[...Wafer, αV...Microscope, 0z...I
TV camera (imaging means), (15)... spatial filter circuit (edge signal output means), αQ... first spatial filter circuit, αD... second spatial filter circuit. (181...Information processing unit (judgment means).

Claims (3)

[Claims] (1) An imaging means for enlarging and imaging the resist pattern formed through the mask of the object to be inspected for mask misalignment; edge signal output means for outputting an edge signal corresponding to a pair of line spacings indicating mask misalignment; and calculating the pair of line spacings based on the edge signals and determining whether or not there is mask misalignment based on these line spacings. 1. A mask misalignment measuring device, comprising: determining means for determining. (2) The mask according to claim 1, wherein the determining means determines that there is no mask alignment shift when both of the pair of line spacings are smaller than a preset reference value. Misalignment measuring device. (3) The mask misalignment measuring device according to claim 1, wherein the resist pattern is a pattern to be inspected for measuring mask misalignment.