JP3288280B2 - Exposure method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a so-called scan for synchronously scanning a reticle and a semiconductor substrate opposed to each other via a projection optical system and sequentially transferring a circuit pattern image passing through a slit-shaped exposure area onto the semiconductor substrate. BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exposure method using a mold exposure apparatus, and particularly to a method of focusing on a semiconductor substrate.

[0002]

2. Description of the Related Art High integration of semiconductor integrated circuits is continually progressing, and the degree of integration is continually increasing. The most effective method for promoting high integration is miniaturization. In optical lithography, g-line (436 nm), i-line (365 nm), K
An rF excimer laser (248 nm) and a shorter wavelength are being developed. However, especially DRAM (Dynami)
In order to increase the capacity of a highly integrated memory device typified by c Random Access Memory), miniaturization alone is not enough, and chip area is also being increased. To cope with this increase in the size of chips, the exposure field of a stepper, which is a typical projection exposure apparatus, is 15 mm square, 17.5 mm square, and 20 mm square.
It is enlarged to □ and 22 mm □ is now common.

In a 64M DRAM, a plurality of chips can be accommodated in a 22 mm square exposure area.
The 6MDRAM can store only one chip. Furthermore, 1
If it is equal to or larger than the GDRAM, no single chip can fit in the 22 mm square exposure area. As a countermeasure against this, it is conceivable to further increase the exposure area of the stepper. However, from the technical difficulties and economic viewpoints, a scan-type reduction projection exposure apparatus has been proposed and is being used for device production. A scanning type reduction projection exposure apparatus arranges a reticle and a semiconductor substrate to face each other via a projection optical system, synchronously scans the two in directions opposite to each other, and sequentially converts a circuit pattern on the reticle passing under a slit-shaped illumination area onto the substrate. It is to be transcribed. For example, a typical lens scan type exposure apparatus, NSR-S201A manufactured by Nikon Corporation, has an exposure area of 25 × 33 mm ² by scanning a semiconductor substrate under a slit exposure area of 25 × 8 mm ^2. are doing.

A focusing mechanism of a scanning type exposure apparatus (referred to as a focusing mechanism) is often different from a focusing mechanism of a conventional stepper. The reason is that while the stepper only needs to focus while the substrate is stationary, the scanning method scans the substrate at high speed, so the focus is detected by detecting the height of the substrate in the exposure area. In this case, a control delay occurs (see FIG. 3). Therefore, Japanese Patent Application Laid-Open No. 7-861
As disclosed in Japanese Patent Application Publication No. 35-35, it is necessary to take measures such that the height of an unexposed area of the semiconductor substrate in front of the slit-shaped exposed area is detected in advance by a focus position detection device so that control delay does not occur. Become. Japanese Patent Application Laid-Open No. Hei 6-260391 discloses a method for correcting the difference between the height and the inclination of the best imaging surface and the holder surface holding the semiconductor substrate by utilizing the mechanism for pre-reading the height of the semiconductor substrate. Proposed.

[0005]

However, there remains a problem which cannot be solved by the above-mentioned focusing mechanism or the proposal relating thereto. A first problem is that a dimensional change occurs due to a focus shift due to a step on a semiconductor substrate. This will be described with reference to FIG. Here, it is assumed that there is a step having a height h on the Si substrate 209 as illustrated. The horizontal distance between the upper end and the lower end of this step is generally at most about several μm, and the exposure slit width W
Therefore, when the scale is adjusted to the exposure slit width W, a vertical step is formed as shown in FIG. When focusing on a semiconductor substrate having such a step, even if the semiconductor substrate is moved up and down in a very short time at the step, defocus occurs near the step.

The reason for this is that in a scanning type exposure apparatus, an image of one point on a semiconductor substrate is formed by scanning this one point across the exposure slit. Therefore, an image forming position at a certain point near the step is a position where the focal positions above and below the step are weighted by the stay time in the exposure slit. Since a defocus is inevitable, in practice, the height of the substrate is generally gradually changed as shown in FIG. 9 in consideration of ease of control. That is, the focus position 211 is
The width is gradually changed with respect to the width W of 0. By the way, in the isolated line pattern and the hole pattern, L & S
(Line and space) pattern, under certain conditions, the dependence of pattern size on defocus, that is, CD (critical)
The dimension) -Defocus curve is a curve that is convex upward as shown in FIG. Therefore, when focusing is performed on a semiconductor substrate having a step structure as shown in FIG. 8 as indicated by a dotted line in the figure, for example, the dimension of the isolated line pattern of 0.25 μm is CD-Defocus of FIG.
It changes in the order of →→→→→→ on the curve. As a result, in the example of FIG. 10, the pattern size fluctuates by about 0.05 μm, which is a level that has a significant effect on device characteristics and yield.

The second problem is that when performing exposure, there is a difference between the substrate height detected by the focus position detecting device and the optimum substrate height viewed from the resist image after exposure, and it is common to input an offset. It comes from something. As a cause of the difference, as shown in FIG. 11, the state of reflection of the focus beam 212 (usually using a He-Ne laser or a halogen lamp) for detecting the substrate height changes depending on the underlying film structure and the like. [A case where the aluminum under the resist 213 is (a) and a case where the under the resist 213 is the Si oxide film 215 (b), etc.] and the position where the focus beam 212 is irradiated and the strictest dimension as shown in FIG. Are different in the positions where the patterns are arranged (see FIG. 12).
In order to clarify the density of the pattern, a latent image 216 to be formed on the resist 213 in the future is shown). As a result, in the conventional stepper method, as shown in FIG. 13, an appropriate focus offset value is experimentally obtained for each product and process, and a constant is input to the exposure apparatus.

In a scanning exposure apparatus, a beam for detecting the height of a substrate also scans the semiconductor substrate. As shown in FIG. 14, on the Si substrate 209, there is a region having a clearly different film structure in the scanning exposure region, such as a case where a region of polycrystalline Si 217 and a region of Si ₃ N ₄ 218 are mixed. Even when the width of the region in the scanning direction is larger than the width of the illumination slit and it is effective to set a different focus offset value, the conventional lens scan type exposure apparatus performs exposure with a constant focus offset value. This causes a focus shift based on the difference in the focus offset value, and further causes a dimensional change due to the focus shift. An object of the present invention is to solve the above-described problems and improve the focusing accuracy in a scanning exposure apparatus,
It is intended to improve device characteristics and yield by improving dimensional uniformity.

[0009]

An exposure method according to the present invention is an exposure method using an exposure apparatus of a projection optical system,
A first step of detecting a step one exposure area on the semiconductor substrate prior to exposure of the semiconductor substrate at the focus position detection detecting device of the exposure apparatus, the height of the highest region of the step on the semiconductor substrate and most a second step of calculating an average value of the height of the lower region, the exposure apparatus 3 of the imaging position of the projection optical system performs exposure while meeting the height of the average value on the semiconductor substrate And a process.

Preferably, in the third step, the imaging position of the projection optical system is set to a height obtained by adding an offset value preset for each region to the height of the average value on the semiconductor substrate. To match. At this time, the offset value is set as a function of the distance from the exposure start point. Preferably, in the second step, the average value is obtained by setting a value obtained by adding a preset offset value to the height of the semiconductor substrate detected by the focal position detection device as the substrate height. In the first step, detection is performed for a plurality of exposure areas, and in the second step, the average of the height of the highest area and the average of the height of the lowest area are calculated. Calculate the average value.

[Operation] In the exposure method of the present invention, exposure is performed by setting the image forming position of the projection optical system to the average height of the highest step area and the lowest step area. Therefore, the amount of defocus due to the substrate step in the exposure area is substantially the same in the vertical direction of the imaging position. Since the distance in the horizontal (scanning) direction of the step portion is sufficiently short with respect to the width of the exposure slit in the scanning direction, it can be treated as a vertical step in focusing. Generally, since the CD-Defocus curves of various patterns are symmetrical with respect to a line having a zero defocus amount, no dimensional change occurs when an equal amount of defocus occurs above and below the image plane. In addition, when there is a region having a different underlying structure in one exposure region, the focus offset value is set as a function from the exposure start point, and the substrate height is controlled based on the function, thereby being most suitable for each underlying structure. Exposure can be performed with the offset value. As a result, it is possible to suppress a change in the pattern dimension and prevent the pattern shape from deteriorating.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a sectional view of a semiconductor substrate taken along a scanning direction for explaining a first embodiment of the present invention. The Si substrate 101 has a portion raised by h at the center of the illustrated exposure region. FIG. 2 shows a plan view of the Si substrate including a plurality of exposure regions shown in FIG.
The exposure region including the step shown in FIG. 1 is shown as a chip 102 in FIG. First, a step on the chip 103 is detected before exposure is started. A semiconductor substrate 207 (Si substrate 10) is placed on a wafer stage 208 of the scan type exposure apparatus shown in FIG.
1) is mounted, the stage 208 is moved in the horizontal direction (scanned), and the surface height is detected by the focus position detection devices 205 and 206 having a light emitting unit and a light receiving unit. On wafer stage 208, reticle 202 held by reticle holding unit 203 via projection optical system 204 is arranged. Further, an illumination optical system (not shown) is provided on the reticle 202, and a slit is opened in the illumination optical system so that an image is formed on the reticle. Is not scanned). The detected height information (step information) is stored in the memory of the exposure apparatus. The number of chips for detecting height information is 1
Although a chip may be used, the sampling chip 1 (1
03), it is preferable to perform measurement on a plurality of chips such as the sampling chip 2 (104) while changing the scanning direction. When measurement is performed for a plurality of chips, the height at each point of the chip is assumed to be the average value of the height information of the corresponding points of the plurality of chips.

Subsequently, the focal position at the time of exposure is determined based on the detected focal position information. In the example of FIG. 1, since a step having a height h is detected, the focal position is
01 is determined to be h / 2 above the low place (,), in other words, a position lower by h / 2 than the high place () of the Si substrate 101. Next, the focus position of the projection optical system is h / 2 higher than the low point on the substrate (the position indicated by the dotted line in FIG. 1).
To start exposure. That is, in FIG. 3, exposure is performed by scanning the reticle 202 in, for example, the right direction and the wafer stage 208 in the left direction while irradiating the illumination light 201. During the exposure, the height of the Si substrate is kept constant. Therefore, h / 2 below the step in the direction away from the projection optical system.
However, conversely, h / in the direction approaching the projection optical system above the step
By two, defocus will occur.

The dimensional change due to the defocus is caused by the CD-Defo of the 0.25 μm isolated line pattern shown in FIG.
This will be described with reference to a cus curve. It should be noted that, other than the isolated line pattern, the hole pattern and the slit pattern always have the same CD-Defocus curve, and the L & S pattern shows the same characteristics under certain conditions. When the exposure is performed such that the focal plane of the projection optical system is located at the position indicated by the dotted line in FIG. 1, the size of the lower part of the step is -h / 2 defocused in FIG. 4, which is about 0.20 [mu] m. Si
When the substrate is scanned and the top of the step comes under the exposure slit,
The pattern size is also +0/2 in FIG. 4, but the pattern size is also 0.20 μm. When the area shown in FIG. 1 comes below the exposure slit, the pattern moves from (on) to (on) the CD-Defocus curve, but the pattern dimension does not change. That is, it is possible to suppress the change in the pattern size due to the defocus shown in FIG.

Next, a second embodiment of the present invention will be described with reference to the drawings. As shown in FIG. 5, a case is considered in which a region having a distinctly different underlying structure exists, such as a case where a region of the Si oxide film 105 and a region of the aluminum 106 coexist in one exposure region. A step is often present between such two regions, but FIG. 5 shows a flat structure for simplicity of explanation. As described above, if the underlying film structure changes, the appropriate focus offset value also changes. In the substrate having the underlayer structure shown in FIG. 5, it is assumed that the optimum focus offset value in the Si oxide film 105 is −0.1 μm, and that in the aluminum 106 is 0 μm. In the case of a conventional constant focus offset value input, when the focus offset value is 0 μm, the Si oxide film portion has a focus offset value of −0.1.
At μm, the aluminum portions were out of focus. On the other hand, as shown in FIG. 6, by setting the focus offset value as a function from the chip start point and changing the substrate height accordingly, it is possible to set the optimum focus offset value for each area. However, when the scanning direction in the scanning exposure is reversed, the chip start point and the chip end point in FIG. 6 are switched.
In the present embodiment, the correction of the focus offset value is performed, so that the boundary of the change in the underlying structure is
Of the pattern size in accordance with the CD-Defocus curve of FIG. However, when the focus offset value is 0.1
When the thickness is about μm, the change is small compared to a dimensional change due to a step. When the focus offset value is corrected for a substrate having a step as in the first embodiment, first, as in the first embodiment,
After the height of each part is detected and the focus position is determined based on this, the focus offset value is corrected for the focus position.

Next, a third embodiment of the present invention will be described with reference to the drawings. This embodiment includes two materials and regions having different heights, and, like the first embodiment, is for dealing with a case where dimensional variation near the step must be considered. As shown in FIG. 7, it is assumed that polycrystalline Si 107 is formed on Si oxide film 105 with a step of h, and that the difference in focus offset value is a. That is, a = (focus offset value at the upper part of the step)-(focus offset value at the lower part of the step) unit: μm In this case, if the step is extremely gentle, the method described in the second embodiment is suitable. However, since the step is generally steeper than the width of the exposure slit, the image forming position of the projection optical system is determined with reference to the lower part of the step in consideration of the dimensional change near the step described in the first embodiment. (H + a) / 2. The subsequent exposure method is the same as in the first embodiment. According to the exposure method of the present embodiment, it is possible to suppress the dimensional fluctuation of various fine patterns.

[0017]

As described above, since the exposure method of the present invention performs exposure while adjusting the focal position to the average height on the semiconductor substrate and fixing the height of the substrate, the exposure method of the present invention And the amount of defocus at the lower part can be made equal. Therefore, according to the present invention, it is possible to suppress a variation in pattern dimension near the step. Further, according to the embodiment in which the focus offset value is set as a function of the distance from the scanning exposure start position, and the substrate height is changed according to the focus offset value, for a substrate including a different base structure in one exposure region, In addition, it is possible to always perform exposure with an optimum focus offset value, and it is possible to suppress a change in pattern dimension due to a change in the underlying structure.

[Brief description of the drawings]

FIG. 1 is a cross-sectional view of one exposure region for explaining a first embodiment of the present invention.

FIG. 2 is a plan view of the Si substrate for explaining the first embodiment of the present invention.

FIG. 3 is a schematic diagram of a scanning exposure apparatus for explaining a first embodiment of the present invention.

FIG. 4 is a characteristic diagram showing a relationship between a pattern dimension and a defocus amount for explaining an effect of the first embodiment of the present invention.

FIG. 5 is a sectional view of a substrate for explaining a second embodiment of the present invention.

FIG. 6 is a diagram illustrating a relationship between a substrate position and a focus offset value for explaining a second embodiment of the present invention.

FIG. 7 is a cross-sectional view of a substrate for explaining a third embodiment of the present invention.

FIG. 8 is a cross-sectional view of one exposure area for explaining a conventional example.

FIG. 9 is a sectional view of a substrate showing a conventional focusing method.

FIG. 10 is a characteristic diagram illustrating a relationship between a pattern dimension and a defocus amount according to a conventional focusing method.

FIG. 11 is a cross-sectional view of a substrate showing a base structure for explaining a problem of a conventional example.

FIG. 12 is a cross-sectional view of a substrate showing a pattern above and below a step for explaining a problem of the conventional example.

FIG. 13 is a diagram showing a relationship between a process for explaining a problem of the conventional example and a focus offset value.

FIG. 14 is a cross-sectional view of a substrate showing a difference between regions of a base structure for explaining a problem of a conventional example.

[Explanation of symbols]

Reference Signs List 101 Si substrate 102 Chip 103 Sampling chip 1 104 Sampling chip 2 105 Si oxide film 106 Aluminum 107 Polycrystalline Si 201 Illumination light 202 Reticle 203 Reticle holding unit 204 Projection optical system 205 Focus position detecting device (issuing unit) 206 Focus position detecting device (Light receiving unit) 207 Semiconductor substrate 208 Wafer stage 209 Si substrate 210 Exposure slit 211 Focus position 212 Focus beam 213 Resist 214 Aluminum 215 Si oxide film 216 Latent image 217 Polycrystalline Si 218 Si ₃ N ₄

Claims (5)

(57) [Claims] 1. An exposure method using an exposure apparatus of a projection optical system, wherein a step in an exposure area on the semiconductor substrate is detected by a focal position detection / detection apparatus of the exposure apparatus prior to exposure of the semiconductor substrate. and one of the processes, a second step of calculating an average value of the height of the highest area level with the lowest area of the step on the semiconductor substrate, the imaging position of the projection optical system of the exposure device the A third step of performing exposure while matching the height of the average value on the semiconductor substrate. 2. A third process, matching the imaging position of the projection optical system, the height plus the average of the heights preset offset value for each area on the semiconductor substrate 2. The exposure method according to claim 1, wherein the exposure is performed. 3. The exposure method according to claim 2, wherein the offset value is set as a function of a distance from an exposure start point. 4. A second step, characterized in that determining the average value of the value obtained by adding a preset offset value to the height of the semiconductor substrate which is detected by the focus position detecting device as substrate height The exposure method according to claim 1, wherein 5. In the first step, detection is performed for a plurality of exposure areas, and in the second step, the average height and the lowest height of the highest step area on the semiconductor substrate are detected. 2. The exposure method according to claim 1, wherein an average value with the average height of the region is calculated.