JP3465710B2 - Projection exposure apparatus, projection exposure method, and integrated circuit manufacturing 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 apparatus, such as a projection exposure apparatus, which requires focus position adjustment with respect to a substrate (photosensitive substrate) to be processed, and more particularly to a focus when the processing region of the substrate is uneven. The present invention relates to a technique for performing good alignment.

[0002]

2. Description of the Related Art Conventionally, in a projection exposure apparatus for exposing a pattern of an integrated circuit onto the surface of a semiconductor wafer or the like, in order to accurately align the wafer surface with a predetermined exposure surface, an optical focus position shift is required. Some adopt a detection method. As one method of this optical focus position shift detection, a light beam having a slender cross section is obliquely projected from above the wafer, which is the surface to be detected, and the reflected light is condensed through a vibrating slit. There has been considered a method of detecting a focus position by an output that is electrically processed after photoelectrically converting the generated optical signal. However, since the detection light beam used for this focus position deviation detection is always irradiated to a fixed position on the stopped wafer, it is possible to perform local focusing of the plane at the irradiation position. If there is a surface in the area (field of view) that has a height different from the height of the plane at the irradiation position of the detected light beam, that surface does not match the focal plane.

In this case, the design rule of the integrated circuit pattern on the surface having a height different from the height of the plane at the irradiation position of the detected light flux is the strictest in the projection exposure area,
If it is necessary to match the focus at that position, inconvenience will occur. In recent years, as the line width of the integrated circuit pattern has been reduced, the NA of the projection exposure lens has increased, and the margin of the depth of focus has decreased. Therefore, when the height of the pattern of each circuit in the projection exposure area is not uniform, it is preferable to perform the focus position adjustment in accordance with the circuit pattern that requires the smallest focus position deviation among them.

[0004]

For this reason, in the prior art, the height of the plane at the irradiation position of the detected light beam used for detecting the focus position shift and the plane at the circuit pattern position where it is necessary to minimize the focus position shift. The difference in depth is specified in advance as the focus offset amount,
In practice, a method of correcting a focus offset amount designated in advance by electrical processing when the focus is made to coincide with the plane at the irradiation position of the detected light flux has been used. In the case of this method, it was necessary to measure the focus offset amount in advance.

Further, after moving the irradiation position of the light beam used for the focus position shift detection to the circuit pattern position where the focus position shift needs to be minimized in advance and focusing is performed, the focus is maintained, A method of moving to the projection exposure position and performing exposure (focus lock) is also conceivable, but there is a drawback that the processing time increases with each exposure processing.

On the other hand, even when the alignment sensor is performing the alignment operation with respect to the alignment mark on the substrate to be processed, the irradiation position of the detected light beam used for detecting the focus position shift is generally the alignment mark position. Therefore, when the height of the plane at the irradiation position of the detected light flux is different from the height of the plane at the alignment mark position, the same problem occurs.

In view of the above, the present invention provides a focus alignment for a substrate to be processed in a projection exposure apparatus such as a step-and-repeat system or a step-scan system when the unevenness of the processing region of the substrate is not uniform. It aims to be efficient.

[0008]

According to the present invention,
The technical point is to automatically measure the difference between the height of the plane at the focus detection position in the projection exposure processing area and the height of the plane at the optimum focus required position in the projection exposure processing area.

In the projection exposure apparatus according to the present invention, when a plurality of processing areas formed on the substrate are sequentially exposed in accordance with the focus position of the projection optical system, the projection optical system of the projection optical system is located at the focus detection position within the processing area. A projection exposure apparatus for detecting a position in the optical axis direction, and before exposing a plurality of processing areas, a portion in the processing area which is to be aligned with the focal position of the projection optical system most strictly and the focus detection position thereof. And a measuring means for measuring step information corresponding to the deviation in the height direction of the substrate surface, and when exposing the plurality of processing regions, align the focus position of the projection optical system most strictly in the processing regions. as want part fits the focal position of the projection optical system, it possesses a correction means for correcting the surface position of the substrate in accordance with the level information, and the measuring means, its the measured light
Irradiate the processing area of the
While relatively moving the measurement light of
It is characterized by measuring step information .

Further, in the projection exposure method of the present invention, when a plurality of processing areas formed on the substrate are sequentially exposed in accordance with the focus position of the projection optical system, the projection optical method is performed at the focus detection position within the processing area. A projection exposure method for detecting a position in the optical axis direction of a system, which irradiates a processing area with measurement light before exposing the plurality of processing areas ,
Detecting the reflected light, the level information corresponding to the height direction of the deviation of the substrate surface of the most closely fit section to be the focal point of the projection optical system in the processing region and its focus detection position, the Move the measuring light and its processing area relatively
While measuring, when exposing the plurality of processing regions, as most closely portion to fit the focal position of the projection optical system in the processing region matches the focal position of the projection optical system, on the level information The surface position of the substrate is corrected accordingly.

[0011]

According to the present invention, before exposing a plurality of processing areas, a portion in the processing area which is to be most closely aligned with the focal position of the projection optical system and a position in the optical axis direction of the projection optical system at the time of exposure are defined. Since the step information between the detected position and the detected position is measured, even if the undulations in the processing area are not uniform, the portion of the projection optical system where the focus position of the projection optical system is desired to be most accurately adjusted is accurately projected. The focus position of the system can be adjusted, and the focus position can be adjusted efficiently.

[0012]

FIG. 1 shows the overall structure of a projection exposure apparatus (stepper) to which the present invention is applied. The pattern area formed on a reticle (mask) R is illumination light from a condenser lens CL of an exposure illumination system. Irradiated uniformly. The light transmitted through the pattern area reaches the wafer W through the projection lens PL that is both-side telecentric or image-side telecentric, and the image of the pattern area is projected onto a predetermined shot area on the wafer W. Reticle alignment microscope RA1,
RA2 detects a mark formed around the reticle R, aligns the reticle R with the optical axis AX of the projection lens PL, or detects a relative mark between the reticle R and the mark on the wafer W. The amount of misalignment is detected and the reticle R and the wafer W are aligned.
Further, a TTL (through the lens) alignment system WA for detecting a mark on the wafer W only through the projection lens PL.
Has a mark detection area within the projection field of the projection lens PL and outside the projection range of the pattern area of the reticle R. Two sets of the same TTL alignment mark system WA are provided so that the mark detection areas are arranged at two positions in the projection visual field.

On the other hand, the wafer W is held on the Z stage 20, and the Z stage 20 is held on the XY stage 21. The Z stage 20 is driven by a drive system such as a motor 19 to move the X stage.
It is configured to be finely movable in the optical axis AX direction (Z direction) with respect to the Y stage 21. Focusing is thereby performed so that the surface on the wafer W matches the best image plane of the projection lens PL. The position of the surface of the wafer W in the optical axis AX direction is
The projector AF is detected by the oblique incident light type position detection system.
The non-exposure light from P is obliquely projected onto the wafer W, and the reflected light is received by the light receiver AFR. The detection signal from the light receiver AFR (a signal whose state changes according to the magnitude of defocus) is sent to the focus control system AFU.
The control system AFU sequentially detects the deviation amount of the surface of the wafer W from the best image plane based on the detection signal, and drives the motor 19 for the Z stage 20 until the deviation amount is driven within a predetermined allowable value. .

The wafer W is exposed by the step-and-repeat method, and its stepping position is detected as a two-dimensional coordinate value by the laser interferometer IFM.
This laser interferometer IFM is used to move the wafer W in a moving plane (xy).
The coordinate position in the plane is detected with a resolution of about 0.01 μm, and it is also used for mark position measurement during alignment.
The motor 40 moves the XY stage 21 (that is, the wafer W) in the x direction and the y direction, and is controlled by the stage control system STU. When the data of the movement target position of the wafer W is sent from the main control system MC, the control system STU calculates the difference from the current position measured by the laser interferometer IFM, and the motor at the optimum speed according to the difference. 40 is driven, and the motor 40 is servo-controlled so that the count value of the laser interferometer IMF is within ± 5 counts (± 0.05 μm) with respect to the target position.

The main control system MC inputs the measurement results by the reticle alignment microscopes RA1 and RA2, the mark position measurement results by the TTL alignment system WA, and bidirectionally sends and receives data and commands with the focus control system AFU. Interact. Particularly in this embodiment, the main control system MC sends the focus offset data to the focus control system A.
It is related to the present invention that the part is sent to the FU, and the focus control system AFU sets the focus offset to the photoreceiver AFR based on it.

Now, FIG. 2 shows the focus control system AFU in FIG.
It is the figure which showed each structure of the light projector AFP and the light receiver AFR in detail. The broadband illumination light IL having a band in the red or infrared region illuminates the slit 1. Slit 1
The light from is obliquely projected onto the surface of the wafer W via the lens system, the mirror 3, the aperture stop 4, the objective lens 5, and the mirror 6. At this time, the image of the slit 1 is formed on the wafer W. The reflected light of the slit image is reflected by the mirror 7, the objective lens 8, the lens system 9, the vibrating mirror 10, and the parallel flat plate glass whose angle is variable (hereinafter referred to as plane parallel).
It is re-imaged on the detection slit 14 via 12. The photomultiplier 15 photoelectrically detects the light flux of the slit image transmitted through the slit 14, and outputs the photoelectric signal to the synchronous detection circuit (PSD) 17. Vibration mirror 1
0 is an oscillator (O) via a drive circuit (M-DRV) 11.
SC. ) 16 oscillates in a constant angular range in response to a sinusoidal constant frequency signal. As a result, the image of the slit 1 re-formed on the detection slit 14 slightly vibrates in the direction orthogonal to the longitudinal direction of the slit, and the photoelectric signal of the photomultiplier 15 is OSC. It will be modulated corresponding to 16 frequencies. PSD17 is OS
C. The photoelectric signal from the photomultiplier 15 is phase-detected with reference to the original signal from 16 and the detected signal SZ
Is output to the processing circuit (CPX) 30 and the drive circuit (Z-DRV) 18 of the motor 19 for the Z stage 20. The detection signal SZ becomes zero level when the surface of the wafer W coincides with the best image formation (reference plane).
When is deviated upward along the optical axis AX, the level is positive, and when it is deviated in the opposite direction, the level is negative.
Therefore, if the Z-DRV 18 is a servo amplifier whose reference is the zero level, the Z-DRV 18 drives the motor 19 so that the detection signal SZ becomes the zero level. This achieves automatic focusing of the wafer W.

On the other hand, the processing circuit (CPX) 30 reads the level of the detection signal SZ and outputs the drive signal DS to the drive unit (H-DRV) 13 that adjusts the inclination of the plane parallel 12 with respect to the optical axis. The H-DRV 13 includes a drive motor and an encoder that monitors the tilt amount of the plane parallel 12, and the up / down pulse output ES from the encoder is read into the CPX 30.

FIG. 3 shows the detailed structure of the CPX 30.
The level of the detection signal SZ is the analog-digital converter (A
It is converted into a digital value by DC) 301 and stored in the storage unit 302. The data held in the storage unit 302 is processed by the correction data calculation unit 303 and the drive data creation unit 304, and the tilt position of the plane parallel 12 corresponding to the amount to be focus offset is output to the subtractor 305. The counter 312 counts up or down the encoder output ES, and outputs the current value of the tilt angle of the plane parallel 12 to the subtractor 305. The subtractor 305 is the creation unit 30.
The difference from the output of the counter 312 is obtained using the data from 4 as a reference, and the difference value is output to the digital-analog exchange (DAC) 308. The DAC 308 applies the converted analog level to the servo amplifier 310, and the amplifier 310 outputs the drive signal DS of the plane parallel 12.

In the configuration of FIG. 3 described above, the storage unit 302, the correction data calculation unit 306, the drive data creation unit 304, and the subtractor 305 are realized by a computer program, and are therefore limited to the above-mentioned functions. Not only does it work
It may also perform some related actions. The specific operation of the correction data calculation unit 306 will be described in detail later, but the main function is to calculate the focus offset amount in connection with the present invention.

Now, FIG. 4 shows the projection field P of the projection lens PL.
The surface structure of the wafer W appearing in the LF is shown in FIG.
(A) shows a state in which the pattern of the reticle R is superposed on the shot area on the wafer W and exposed. Wafer W
In the upper shot area, two chip areas CP1,
CP2 is across the street (scribe) line,
It is assumed that two chip patterns are also formed in the pattern area on the reticle R with the street line portion sandwiched therebetween. At the center of the projection visual field PLF (the position where the optical axis AX passes), the imaging light beam FB from the projector AFP of the focus position detection system is the X axis in the XY plane,
The light enters from a direction inclined by 45 ° with respect to each of the Y axes, and an image SB of the light transmitting slit 1 is formed on the wafer W. This slit image SB is set to a size (several hundred μm to several mm) which is considerably shorter than the entire area (diameter) of the projection visual field PLF.
In the state of (A), it is projected on the street line between the chip regions CP1 and CP2. Further, in one shot area on the wafer W, a wafer mark WMx for the X direction and a wafer mark WMy for the Y direction detected by the wafer alignment system WA are formed. Further, a detection area WAX of the wafer alignment system WA for the X direction and a detection area WAY of the wafer alignment system WA for the Y direction are set at two peripheral positions in the projection visual field PLF. In the detection areas WAX and WAY, the wafer alignment system WA spots the laser beam (several μm to 1 m).
In the method of projecting onto the wafer with m), it corresponds to the projection area of the spot, and in the method of detecting the mark image with the television camera (CCD), it corresponds to the imaging area by the camera.

As shown in FIG. 4A, in the exposed state, the wafer mark WMx and the detection area WAX are displaced in the Y direction, and the wafer mark WMy and the detection area WAY are displaced in the X direction. This is because a part of the optical system of the wafer alignment system WA (mirror or the like) is arranged as close to the projection field PLF as possible so as not to enter the projection area of the reticle pattern. When the reticle alignment systems RA1 and RA2 shown in FIG. 1 are made to be able to detect a wafer mark by the die-by-die method, the detection area (WA) in the exposure state of FIG.
The positions of the objective lenses of the reticle alignment systems RA1 and RA2 may be set such that (X, WAY) and the marks WMx and WMy match.

FIG. 4B shows a state in which the wafer mark WMx is detected by the detection area WAX of the wafer alignment system for the X direction, and the XY stage 21 is shown in FIG.
The state is shifted from the state of (A) in the Y direction by a certain amount. In this case, the slit image S by the focus position detection system
B is projected on the street line between the chip areas CP1 and CP2.

FIG. 4C shows a state in which the wafer mark WMy is detected by the detection area WAY of the wafer alignment system for the Y direction, and the XY stage 21 is shown in FIG.
The state is shifted from the state of (A) in the X direction by a certain amount. In this case, the slit image SB of the focus position detection system
Are located in the approximate center of the chip region CP2. Normal,
When arranging a plurality of chips in one shot, all the chips are made to have the same pattern structure. Therefore, the fact that the slit image SB is located at the center of the chip area CP2 is
This means that the focus position detection system (AFP, AFR) can typically detect the height position of the surface in all the same chip regions formed on the wafer W.

FIG. 4D shows a state in which the XY stage 21 is positioned so that the wafer mark WMy is located at the center of the projection visual field PLF (the position of the slit image SB). Since the wafer marks WMy and WMx are formed in the street lines around the chip area, FIG.
In the case of, the height position of the wafer surface detected by the focus detection system is almost the same as that in the case of FIG. This is almost the same even if the wafer mark WMx is arranged at the position of the slit image SB.

Generally, the surface of the chip region CP formed on the wafer W becomes higher than the street line as the process progresses. In addition, there are portions with large steps depending on the pattern structure in the chip.
Therefore, if focus detection (projection of the slit image SB) is always performed on one shot area on the wafer W at the same portion, accurate focusing may not be performed due to the influence of the step. .

Therefore, in the first embodiment of the present invention,
Prior to exposing the wafer by the step-and-repeat method, information about flatness within one shot area on the wafer W is measured by a focus position detection system, and optimum focusing is performed with the flatness reflected. I chose Figure 5
3 is a flow chart schematically showing the sequence according to the first embodiment. This sequence is executed jointly by the main control system MC shown in FIG. 1 and the CPX 30 (FIG. 3) shown in FIG.

First, the main control system MC selects some typical shot areas from the plurality of shot areas based on the array data of the shot areas formed on the wafer W during step and repeat, and N Individual designated shot areas are set (step 100). This shot designation may be executed by the operator. Further, in step 100, a parameter M of a point to be focus-detected other than the center within one shot area is set.
This parameter M gives the position and the number of measurement points in the shot area other than the central portion in one shot area to be measured. However, the position of the measurement point may be defined as a shift amount in the X and Y directions from the center of the shot area, and it is not necessary to set it individually for each of the N designated shot areas.

FIG. 6 shows such a situation. When the focus position is detected at the exposure position as in FIG. 4A (most shot areas are applicable), the parameter M is M0.
(0, 0) is set. Also in FIG. 6, the two chip areas CP1 and CP2 are treated as one shot area SA. As shown in FIG. 6, one shot area SA
When measuring at 6 points other than the center, the parameters M are M1 (x1, y1), M2 (x2, y2),
M3 (x3, y3) ... M6 (x6, y6) is set. Coordinate values of each parameter M0 to M6 (xm, ym)
Is set as the amount of deviation from the central measurement point M0 (0, 0). Also, in Fig. 6, 45 ° diagonally passing through each measurement point
The line of represents the slit image SB shown in FIG.

In the case of FIG. 6, one shot area SA has two chip areas CP having the same pattern structure.
Since 1 and CP2 are built in, it may be sufficient to select the measurement point for only one of the chip regions. The operator appropriately determines how to arrange the measurement points. As described above, the parameter N,
When M is determined, in step 101, the XY stage 21 is stepped to the exposure position for the designated shot area. At this time, the designated shot area is the projection visual field P.
It is positioned in the LF as shown in FIG. Next, in step 102, the focus position detection system (1
~ 17) and Z-DRV 18, motor 19, Z stage 2
By 0, the wafer W is focused so that the detection signal SZ becomes zero level. Then, when the detection signal SZ is at the zero level, the servo control by the Z-DRV 18 is stopped and the motor 19 is locked so as not to be driven in response to the signal SZ. As a result, the center point M0 (0, 0 in FIG.
Focusing is achieved for 0).

Next, in step 103, XY is set based on the parameter M so that the measurement point m in the shot area SA coincides with the slit image SB of the focus position detection system.
Position the stage 21. At this time, Z stage 2
Since 0 does not move in the optical axis AX direction, the first measurement point M
When 1 (x1, y1) comes to the position of the slit image SB, the detection signal SZ of the PSD 17 shown in FIG. 2 has a level corresponding to the difference in the height direction between the center point M0 and the measurement point M1. Therefore, the CPX30 sets the level to ADC3 in FIG.
It is read as ΔZnm by 01 and stored in the storage unit 302. When the deviation amount ΔZnm for the first measurement point M1 is stored, it is determined in step 104 whether or not there is a point to be measured in the shot (m = M?). Repeat the operation of.

By the above operation, the amount of deviation in the height direction with respect to the center point M0, that is, the flatness is obtained for each of the M measurement points in the designated shot area. Then, in step 105, the focus lock is released, and in step 106, it is determined whether or not all N designated shot areas have been processed (n = N?). If there are remaining designated shot areas, step 101 is performed. Repeat the sequence from ~ 105.

As described above, the flatness within a shot can be obtained at multiple points within the designated shot area, but if only one specific spot within the shot area needs to be accurately focused, the step 100 is executed. The parameter M to be set is set only in one place. If zero is set as the parameter M, the sequence after step 101 is not executed. That is, as in the conventional case, the exposure is performed only by focusing on the center point of the shot area.

Next, the CPX 30 calculates the stored deviation amount data ΔZnm by the correction data calculation unit 306 of FIG. 3, but before that, in step 107, it is selected whether each data ΔZnm is within a predetermined allowable value. . What is this tolerance?
For example, it is set to a value that is at least twice the degree of the unevenness of the surface assumed within one shot. Of course, this allowable value also needs to be changed depending on the layer structure formed on the wafer W, and also depending on how the chips are taken in the shot area (multi-die or single-die) as shown in FIGS. Need to change. It is also possible to select an allowable value that does not depend on the layer structure or the way of taking chips. It is a method of obtaining the variance (σ value or 3σ value) of each measured data ΔZnm and rejecting only the deviations from the range. Alternatively, there is also a method of obtaining a simple average value of the data ΔZnm, defining a certain range (determined from the measurement error of the focus detection system, etc.) with respect to the average value, and rejecting those outside the range. Data Δ
Those of Znm that are out of the allowable range are rejected,
The remaining data ΔZnm is averaged in the next step 108.

Step 108 (correction data calculation unit 30)
The average value ΔZ calculated in 6) is used to set the focus offset in the next step 109. Average value ΔZ
Is originally obtained as the level of the detection signal SZ, so that when the focus offset is added on the electric signal, the offset voltage equal to the average value ΔZ is
The detection signal SZ which is given to the amplifier provided in the output stage of SD17 and is output is level-shifted or Z-D
The offset voltage may be given to the amplifier provided at the input stage in the RV 18 and added with the detection signal SZ from the PSD 17.

Further, as shown in FIG. 2, when optically offsetting by the plane parallel 12, the CPX
Reference numeral 30 is a drive data creation unit 304 of FIG.
It outputs to 05 and controls H-DRV13. When the plane parallel 12 correction is completed, the DAC in the CPX30
The output of 308 may be forcibly held at zero, or the input of the servo amplifier 310 may be forcibly held at zero.

When the sequence of FIG. 5 is completed as described above, the main control system MC starts the exposure operation of the step-and-repeat system. At this time, the focus position detection system detects the height of the central portion in the shot area. The best focus position is set on a surface different from the position, and it works so as to match the best image surface with a portion in one shot area that is most carefully focused.

Next, a second embodiment of the present invention will be described. The basic sequence is the same as that of FIG. 5, but the method of determining the measuring point M in the designated shot area is the first embodiment. Different from. Normally, when overlay exposure is performed on the wafer W, it is necessary to detect the mark position on the wafer W and perform relative alignment with the reticle R before that. The wafer alignment system WA shown in FIG. 1 is used to detect the mark position, but at the time of detecting the mark, each mark is detected in a focused state in the detection areas WAX and WAY in the field PLF of the projection lens PL. Is desirable.

Therefore, first, as shown in FIG. 4D, XY is set so that the mark portion of the shot area to be aligned (mark position detection) on the wafer W comes to the position of the slit image SB of the focus position detection system. The stage 21 is positioned, and the Z stage 20 is servo-controlled here to focus and servo lock as in step 102 of FIG. As a result, the portion of the street line where the wafer marks WMx and WMy are formed is focused on the position of the best image plane of the projection lens PL.

Next, the XY stage 21 is moved so that the marks WMx and WMy on the wafer W are moved to the positions shown in FIG.
Sequentially set to the state of (C), mark WMx,
Each position of WMy is detected based on the mark detection signal from the wafer alignment system WA and the measurement value of the laser interferometer IFM. At this time, especially in the state of FIG.
The X30 reads the level of the detection signal SZ via the ADC 301. In the case of FIG. 4C, since the slit image SB of the focus position detection system is projected on the substantially central portion of the chip area CP2, the level of the detection signal SZ in the state of FIG. , The Z-direction positional deviation ΔZn of the surface of the central portion of the chip region with reference to the surface of the street line portion where the marks WMx and WMy are formed.

Generally, at the time of global alignment for aligning the wafer W as a whole, it is necessary to detect the wafer mark associated with each of at least two shot areas on the wafer W.
The state as shown in (C) occurs at least twice in different shot areas. So at least 2
It is sufficient to determine the focus offset amount by averaging the position deviations ΔZn obtained at the points.

However, when similarly detecting the position deviation ΔZn for two or more shot areas, the XY stage 21 should be positioned in the arrangement as shown in FIG. 4C without executing the mark detection. Good. Further, when the portion to be strictly focused in the chip area is different from the state shown in FIG. 4C, the slit image SB is formed in the chip area while the wafer W is being moved to the state shown in FIG. 4C. The movement route of the XY stage 21 may be set so that it is projected onto a desired portion of the.

When the height position of the desired portion in the shot area is detected with reference to the street line portion on which the wafer mark is formed as in this embodiment, as shown in FIG.
As in (A), since it is not necessary to perform focus lock one by one at the center (exposure state) in the shot area, the number of movements of the XY stage 21 is reduced, and the focus is adjusted to the mark at the time of alignment. There is an advantage that it can be done. Further, if the position deviation ΔZn is detected almost at the same time when the mark is detected as in the present embodiment, all the shot areas to be exposed are exposed in the step-and-repeat exposure by die-by-die (each shot) alignment. Individual position deviation Δ
Zn will be required. Therefore, in the second and subsequent shots after the exposure of the wafer W is started, the positional deviation ΔZn detected in the previous shot exposure and the positional deviation ΔZn + detected in the die-by-die alignment of the current shot. It is possible to more accurately calculate the average value based on 1 and set the offset amount.

Next, a third embodiment of the present invention will be described with reference to FIG. 7, but this embodiment is basically the same as the first embodiment described above.
Alternatively, it is the same as the second embodiment. The different point is that the global alignment of the wafer W is combined with the EGA (enhancement global alignment) method. Regarding the EGA method, for example, JP-A-61-4
Since it is disclosed in detail in Japanese Patent No. 4429, etc., the explanation of its principle is omitted here, and only the sequence closely related to the present invention will be explained.

FIG. 7 shows an example of the arrangement of sample alignment shots according to the EGA method. In FIG. 7, the hatched shot areas are sample shots SA1 to SA8 designated by the EGA. In the case of the EGA method, the wafer alignment system WA uses sample shots SA1 to SA8.
Wafer marks WMx and WMy attached to each of the above are sequentially detected in the order of the arrows in FIG. 7, and the position of each mark is measured. Then, based on the measured plurality of mark positions (coordinate positions of sample shots), the coordinate positions of all shot areas on the wafer W are determined by a statistical method. Then, based on the determined coordinate position, stepping and exposure of the XY stage 21 are repeated.

At this time, each sample shot SA1 to S
At the time of detecting the mark position of A8, as in the first or second embodiment, the focus is strictly set with reference to the center point of the shot area onto which the slit image SB of the focus position detection system is projected at the time of exposure. The position deviation ΔZn of the portion to be aligned is sequentially detected. Similarly, after performing selection based on the allowable value, the data of the remaining position deviation ΔZn is averaged to obtain the focus offset amount, and the offset amount is added to the focus position detection system before executing the step-and-repeat exposure. Keep it.

By doing so, the deviation of the portion to be truly focused in the shot area is detected almost simultaneously during sample alignment in the EGA method, so that the focus offset can be determined without extremely lowering the throughput. There are advantages such as. As described above, in each of the embodiments of the present invention, the slit image SB of the focus position detection system, that is, the level of the detection signal SZ is read while the detector is stationary at a desired position in the shot area, but this is not always necessary. Alternatively, if the electrical response is sufficiently high, the level of the detection signal SZ may be digitally sampled at the moment when the detector crosses the desired position while the wafer W is moving. Alternatively, within a local range including the desired position, the laser interferometer IFM
It is also possible to sample the level of the detection signal SZ several times in response to the up / down pulse from and to use the average value as the data of one position deviation ΔZn.

When a plurality of wafers (lots) in the same process are continuously processed, instead of using the focus position detection system to determine the focus offset amount for each wafer, the first plurality (2 to Even if the offset amount is measured for 3 wafers, the average value of the focus offset amounts measured and determined for each of the first plurality of wafers is applied to the subsequent wafers. Good.
In this way, the wafers that are subsequently processed after the processing of the first plurality of wafers are not required to perform the operation at the time of measuring the focus offset, so that the overall processing speed can be improved.

Further, in each of the above embodiments, the position deviation ΔZ
Although the level of the detection signal SZ of the PSD 17 is measured as n, the value of the counter 312 corresponding to the inclination value of the plane parallel 12 can be used. In this case, when the wafer W is moved from the position where the focus is locked (the level of the detection signal SZ is zero) to a desired position in the shot area, the level of the detection signal SZ is held at zero in the plane parallel 12 If the inclination of is corrected and the correction amount is read from the counter 312, it corresponds to the position deviation ΔZn. When the position deviation ΔZn is detected by correcting the inclination amount of the plane parallel 12 as described above, a sufficiently high accuracy (resolution) can be obtained, but there is also a problem that it takes some time. for that reason,
When it is necessary to measure the unevenness in the shot area with high accuracy, the deviation ΔZ from the tilt amount of the plane parallel 12 is used.
It is desirable to measure n.

As described above, according to this embodiment, even when the optimum focus position cannot be adjusted at the projection exposure processing position, the optimum focus position is adjusted by the correction value calculated in advance. Is possible. Moreover, since this correction value can be automatically measured and calculated, it is possible to improve work efficiency.

Even when the optimum focus position cannot be adjusted at the alignment position, the optimum focus position can be adjusted by the correction value calculated in advance. Moreover, since this correction value can be automatically measured and calculated, it is possible to improve work efficiency. Further, according to the present embodiment, since the measurement for determining the best focus position according to the flatness in the shot area to be projected and exposed can be performed overlapping with the alignment operation, the throughput may be reduced. To be prevented.

[0051]

According to the present invention, even when the undulations in the processing area are not uniform, the portion that is most closely aligned with the focal point of the projection optical system can be precisely aligned with the focal point of the projection optical system. Therefore, it becomes possible to efficiently perform the focus position adjustment.

[Brief description of drawings]

FIG. 1 is a diagram showing an overall configuration of a projection exposure apparatus to which the present invention is applied.

FIG. 2 is a diagram showing a configuration of an oblique incident light type focus position detection system and its control system provided in the apparatus of FIG.

3 is a block diagram showing the configuration of a processing system 30 in FIG.

FIG. 4 is a diagram showing various states of a shot area on a wafer appearing in a projection visual field of a projection lens.

FIG. 5 is a flow chart for explaining the operation according to the first embodiment.

FIG. 6 is a diagram showing an arrangement example of a large number of focus measurement points in a shot area.

FIG. 7 is a view showing an arrangement example of sample shots on the wafer by the EGA method.

[Explanation of symbols]

R reticle W wafer PL projection lens PLF projection field of view AFP floodlight AFR receiver SB slit image (detector) 1 Slit for light transmission 12 plane parallel 14 Light receiving slit 17 Synchronous detection circuit (PSD) 19 motor 20 Z stage 30 Processing circuit (CPX)

Claims (16)

(57) [Claims] 1. When sequentially exposing a plurality of processing regions formed on a substrate in accordance with the focal position of a projection optical system, the position in the optical axis direction of the projection optical system at the focus detection position in the processing region is set. A projection exposure apparatus for detecting, prior to exposing the plurality of processing regions, of the substrate surface of the focus detection position and a portion that is to be most closely aligned with the focus position of the projection optical system in the processing region. Measuring means for measuring step information corresponding to a deviation in the height direction; and, when exposing the plurality of processing regions, a portion of the processing region that is desired to be most closely aligned with a focus position of the projection optical system is the projection optical system. Correction means for correcting the surface position of the substrate according to the step information so as to match the focal position of the system.
Then, the measuring means irradiates the processing area with measuring light.
Detects the reflected light, and the measurement light and the processing area
A projection exposure apparatus, characterized in that the step information is measured while moving relatively . 2. The apparatus further comprises setting means for allowing an operator to set a portion in the processing area that is desired to be aligned with the focus position of the projection optical system most strictly.
The projection exposure apparatus according to. 3. The height direction of the surface of the substrate at the focus detection position in the processing region and a plurality of points in a portion that is to be most strictly aligned with the focus position of the projection optical system. 3. The projection exposure apparatus according to claim 1, wherein the step exposure information is obtained based on the measurement result of the position at the position. 4. An interferometer for detecting the position of the substrate in a plane perpendicular to the optical axis, and the focus detection position in the processing region corresponding to the position of the substrate detected by the interferometer. 4. The storage device according to claim 3 , further comprising: and a storage unit that stores positions in the height direction of a plurality of points in a portion that is to be aligned with the focus position of the projection optical system most strictly. Projection exposure device. Wherein said measuring means is a projection exposure apparatus according to any one of claims 1 to 4, characterized in that measuring the level information at the designated processing region selected from the plurality of processing areas . 6. A first exposure for exposing a portion of the processing region that is desired to be most closely aligned with the focal position of the projection optical system by measuring the step difference information so as to be aligned with the focal position of the projection optical system. And a control unit capable of selecting a process and a second exposure process of exposing the focus detection position in the processing region to the focus position of the projection optical system without measuring the step information. Claim 1 characterized by
The projection exposure apparatus according to claim 5 . Wherein said projection exposure apparatus is a projection exposure apparatus according to any one of claims 1 to 6, characterized in that a projection exposure apparatus of step-and-scan method. 8. The integrated circuit manufacturing method characterized by the production of integrated circuits by using a projection exposure apparatus according to any one of claims 1-7. 9. When sequentially exposing a plurality of processing regions formed on a substrate in accordance with the focal position of the projection optical system, the position in the optical axis direction of the projection optical system at the focus detection position in the processing region is set. A projection exposure method for detecting, wherein measuring light is applied to the processing area before exposing the plurality of processing areas.
Along with irradiating the area, the reflected light is detected , and it corresponds to the deviation in the height direction of the substrate surface between the focus detection position and the portion to be most closely aligned with the focus position of the projection optical system in the processing area. The step information for the measurement light and the processing
Measurement is performed while relatively moving the area, and when the plurality of processing areas are exposed, the portion of the processing area that is most closely aligned with the focus position of the projection optical system is at the focus position of the projection optical system. A projection exposure method, which corrects the surface position of the substrate according to the step information so as to match. 10. The projection exposure method according to claim 9 , wherein a portion of the processing region that is desired to be most closely aligned with the focus position of the projection optical system is set by an operator. 11. The position in the height direction of the substrate surface is measured at the focus detection position in the processing area and at a plurality of points in a portion where the focus position of the projection optical system is to be most strictly adjusted. , claim 9 or 1, wherein the obtaining the level information on the basis of these measurement results
0. The projection exposure method according to 0 . 12. A focus detection position in the processing area corresponding to a position of the substrate detected by an interferometer, the position of the substrate in a plane perpendicular to the optical axis being detected. 12. The projection exposure method according to claim 11 , further comprising storing the positions in the height direction of the plurality of points in the portion that is to be aligned with the focus position of the projection optical system most strictly. 13. The projection exposure method according to any one of claims 9 to 12, characterized in that measuring the level information at the designated processing region selected from the plurality of processing areas. 14. By measuring the step information,
A first exposure process for exposing a portion of the processing region that is desired to be most closely aligned with the focus position of the projection optical system to the focus position of the projection optical system; and the process without measuring the step information. The projection exposure according to any one of claims 9 to 14 , wherein a second exposure process of exposing the focus detection position in the area in accordance with the focus position of the projection optical system can be selected. Method. 15. The projection exposure method, projection exposure method according to any one of claims 9-14, which is a projection exposure method of the step-and-scan type. 16. Integrated circuit fabrication method characterized by the production of integrated circuits using the projection exposure method according to any one of claims 9-15.