JP2671338B2 - Exposure method and substrate attitude control method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial applications]   The present invention also relates to an exposure method and a substrate attitude control method.
Therefore, it is necessary to control the posture of the substrate such as the semiconductor wafer and expose it.
Exposure method for executing light and posture of substrate such as semiconductor wafer
It relates to a control method. [Conventional technology]   A conventional device of this type is disclosed, for example, in JP-A-58-103136.
As previously disclosed in the report.
Measure the height position of three or more points on the surface of c and measure each value.
The approximate plane of the substrate surface is calculated based on
A chuck that holds the substrate so that it is parallel to the specified horizontal plane.
He was leveling (tilting). [Problems to be solved by the invention]   As described above, in the conventional technique, three or more points on the substrate
The position of
To calculate the drive amount of the drive shaft to drive
Method, it is difficult to improve the accuracy of the position of the substrate after driving.
There was a point.   That is, the measurement system and the drive system are independent of each other.
Therefore, if the driving accuracy is poor, the accuracy of the measurement system will be much better.
Even if the correct positioning accuracy is not obtained, and the driving accuracy
The accuracy of the measurement system is worse than the accuracy of the drive
This is because good accuracy cannot be obtained.   The present invention has been made in view of such conventional problems.
Then, it aims at improving the precision of the control position of the substrate. [Means to solve the problem]   To achieve the above object, the present invention has the following configuration.
Was. First, a mechanism for measuring the height of many points on the substrate.
This mechanism can also be preset to any height.
For example, when a point on the board reaches the set height, it is detected.
It has a function that can be put out.   Next, a mechanism for controlling the posture of the board is provided,
By supporting the above axes and moving at least two axes up and down
It is configured to control the posture of the substrate. Also, this figure
The force control mechanism is based on the detection signal from the measurement mechanism
Drive the axis so that the point on the top is the position of the detection signal.
You can drive in.   Using the measurement mechanism and attitude control mechanism described above,
Use the following method to make the substrate the target posture.   First, using the measurement mechanism, at three or more points on the board
The height is measured, and the n-th order equation of the substrate is calculated from the obtained measurement results.
Is calculated.   Next, set the desired target posture (flat
Calculate how much to move the drive axis when the
I do.   Then at least one monitor per drive axis
Point is set, the other drive shaft is fixed, and one drive shaft is targeted.
Calculate the height change of the monitor point when driven to the position
The measurement result so that the signal is output at that height.
I will put it.   Move the measuring device over the monitor point and move one drive axis
Drive so that the height of the monitor point becomes the calculated position
Drive the axis. This is repeated for the number of drive axes.   Highly accurate leveling control is achieved by the above method.
You. (Operation)   In the present invention, a certain point on the substrate was actually calculated.
While checking the height with a measuring device, drive
Since the driving device is driven, the posture of the substrate can be accurately controlled.
You. 〔Example〕   FIG. 1 shows an attitude control device according to an embodiment of the present invention.
Of step-and-repeat type projection exposure equipment
The configuration is shown. Reticle R with circuit pattern is a reticle
It is held by the holder 2 and the illumination term IL of uniform illuminance
Illuminated. The pattern of the reticle R is on the projection lens PL
Therefore, the image is projected onto the wafer W for semiconductor device fabrication.
You. The wafer W is placed on the leveling stage 9,
Inclined in any direction by the drive system 10, 11, 12
You. In FIG. 1, the wafer W has a taper.
Corrected surface of wafer W and best image plane of projection lens PL
To match (parallel) with (the conjugate plane with reticle R)
The state in which the leveling stage 9 is tilted is exaggerated and displayed.
I forgot. These leveling stage 9 and drive system 1
0, 11, and 12 are xy stations that translate in a two-dimensional manner in the horizontal plane.
The xy stage 8 is installed on the page 8, and the xy stage 8 is a motor, etc.
Driven by the stage drive unit 30 including
The position is sequentially measured by the stage interferometer 32. Also above
Of the drive systems 10, 11 and 12 for the leveling stage 9, the book
In the embodiment, for example, the drive system 10 is fixed and the drive systems 11 and 12 are
Up and down independently of each other (in the direction of the optical axis of the projection lens PL)
Movement). The drive systems 11 and 12 are
It moves up and down in response to the drive amount command from the drive unit 34.   The control unit 36 is based on the measurement coordinate value from the stage interferometer 32.
Then, when a predetermined drive command is output to the stage drive unit 30,
Both, the xy stage 8 (that is,
The wafer W) is positioned.   Now, the best image plane of the projection lens PL and the station on the wafer W are
Oblique incidence to match the local shot area surface
An optical focus sensor is provided. This sensor
Multicolor LED or halogen lamp with broad wavelength distribution
Light source 3 such as pump, condenser lens 4, slit 5, slit
Image projection objective 7, slit from wafer W surface
A light-receiving objective lens 13 for entering image-reflected light, a vibrating mirror 14,
Harving glass (parallel plate glass) 6, light receiving slit
15 and a photoelectric element 16 and the like, for details, see JP-A-60-
It is equivalent to that disclosed in Japanese Patent No. 168112. Therefore shake
By the action of the moving mirror 14, the slit image is reflected on the wafer W.
The reflected image vibrates so that it crosses the slit 15,
When the center of vibration coincides with the center of slit 15, the wafer
Set to match (focus) the surface of W and the best image plane
Have been. Although not shown in FIG.
These photoelectric signals (AC) are based on the vibration frequency of the vibrating mirror 14.
Synchronous detection is performed based on the
Ana that changes voltage between positive and negative depending on the front pin and rear pin
Log signal (so-called S-curve signal) as focus signal
can get. So that this focus signal is always OV,
Supports a V stage (not shown) (included in xy stage 8)
Each shot can be moved up and down by robot control.
The pattern image of the reticle R is focused and exposed for each area.
You. In addition, the harving glass 6 has a light receiving slit 15 and a slit.
This is to relatively displace the vibration center of the reflection image,
By tilting this harping glass 6,
The surface height of the wafer W detected as
Shift in the optical axis direction (apply focus offset)
Can). The inclination of the harving glass 6 is harvey
And the amount of tilt (foreground
Use the rotary encoder 17
Detected.   Here, the control unit 36 controls the harving drive unit 38 to operate in two modes.
Command to run one or the other. 1
The two modes measure height at multiple points on the surface of the wafer W.
In order to prevent the
Is a height measurement mode that is fixed to a preset point)
Mode, the best image plane of the projection lens PL itself is the environment
Fluctuations such as temperature, humidity, atmospheric pressure or accumulation of exposure energy
Harving glass 6 following the change in the optical axis direction
This is a follow-up mode in which the slope of is gradually changed.   This tracking mode tilts the harping glass 6
For the detailed operation to be performed, see, for example, JP-A-61-183928.
It is disclosed in the official gazette.   The above grazing incidence light type focus sensor is a synchronous detection type
A method other than the above, such as a reflection slit from the wafer surface, may be used.
Image in one-dimensional photo array (CCD etc.) or one-dimensional position
Sensor (PSD, etc.)
The same effect can be obtained by detecting the defocus from the change.
Also, the slit image of the slit 5 on the wafer surface is exposed.
Actually because it is projected almost in the center of the shot area that should be
Focusing of the exposed area is performed in real time.
You.   Now, FIG. 2 is a diagram showing the structure of the leveling stage 9.
Yes, wafer W is attached to wafer holder (chuck) WH
The wafer holder WH is mounted on the leveling stage 9
It is provided integrally. Drive systems 10, 11 and 12 at three locations are
C Center of W (or inside wafer W mounting surface of holder WH)
3 virtual lines extending radially from the heart_a, L_b, L_cOn top
Place the line l_a, L_b, L_cIs open at an angle of almost 120 °.
In addition, from the wafer center to the action point of each drive system 10, 11, 12
Both distances are almost equal. Hereafter the theory
The drive system 10 that is the fixed point is called the C-axis for the sake of simplicity.
And the two drive systems 11 and 12 that move up and down respectively,
I will call it the axis. The drive systems 11 and 12 have A and B axes.
Potential for detecting the height position from the predetermined origin of
A meter is provided. From this potentiometer
The measured value of is used in the open control described in detail later.
It is. In addition, the leveling stage having the above-mentioned configuration is, for example, a special feature.
It is disclosed in Japanese Unexamined Patent Publication No. 62-274201.   Next, the operation of this embodiment will be described.
Is the same as the conventional method, and the wafer surface measured in advance is
Be parallel (or coincident) with the plane (here the best image plane)
In order to measure the tilt of the wafer surface, drive system 1
Open control method that uniquely drives 1 and 12 by a predetermined amount
Formula (open drive) and the most characteristic moni
Alternative to terring control method (close drive)
Is configured to run.   Fig. 3 shows the above open control and monitoring control.
Either one of them is executed, and the whole surface of the wafer W and the best image plane are formed.
The basics of global leveling to match and in parallel
It is a flowchart figure showing a sequence.   The steps 100 to 128 in FIG. 3 will be described below. [Step 100]   Here, the harbing drive unit 38 has two modes.
When the tilting control of the harving glass 6 is performed in the following mode.
If this happens, stop following the harving glass 6
Return the point to and move to the height measurement mode. Harbi
The tilt of the glass 6 is controlled by the rotary encoder 17.
Since it is detected, returning to the origin is extremely easy.   Next, the control unit 36 moves the xy stage 8 to move the wafer.
Position so that the center of W is at the optical axis position of the projection lens PL.
Place it and get it from the focus sensor at that position
Adjust the Z signal so that the focus signal (S curve signal) becomes OV.
Move the tage up and down. Vertical movement of Z stage stopped
Then, keep the height position and execute the next step 102.
You. At this time, the surface of the central part of the wafer is
The focus is set for the sensor, but in follow-up mode
Therefore, it is not always possible to focus on the best image plane.
No. (Step 102)   Next, a plurality of measurement points on the wafer W designated in advance
P_i(I = 1, 2, 3, ...) One of the focus sensors
Position the xy stage 8 so that the slit image of
Make a decision. At this time, vertical movement of the Z stage is prohibited.
You. And the measurement point P_iObtained from the focus sensor at
Tilting of the harving glass 6 based on the focus signal
Servo-control the slope and measure point P_iAgainst the focus sensor
And make it the same as when focused. By this
When the dust signal (S curve signal) becomes OV, the control unit 36
Changes from encoder 17 via harving drive 38
Read the amount. The amount of this change depends on the wafer center point and the measurement point P._i
It corresponds to the difference in the height direction with.   The above operation is performed at multiple measurement points P_iRepeat for all
Return, each measurement point P_iThe height position of the
Measure.   Generally, one plane can be specified by determining 3 points,
For example, at least three measurement points P shown in FIG._a,
P_b, P_cShall be selected. Of course more measurements
You may choose a point. Measuring point P_a, P_b, P_cIs the line l_a,
l_b, L_cShall be located on the measurement point P_aBetween axis A
Distance, measurement point P_bAnd B-axis distance, and measurement point P_cAnd C-axis distance
Shall be predetermined. (Step 104)   Next, the control unit 36 controls the measurement points P_iHeight position and each measurement
Fixed point P_iPosition on the xy coordinate system, that is, measurement point P_i3D
Based on the coordinate values, the nth-order plane equation of the wafer surface (currently
Of the plane formula) is calculated. In this embodiment, the least squares method is used.
Then, the first-order plane equation is calculated. Least squares
The application is described in detail in JP-A-58-103136.
Therefore, the description is omitted here. [Step 106]   Next, a target corresponding to the best image plane of the projection lens PL is set.
Plane equation (assuming it is stored in advance) and
Compare the current first-order plane equation obtained in
Calculate how much the wafer surface is tilted
You. [Step 108]   Next, is the calculated relative inclination (deviation amount) within the allowable range?
Determine whether or not. This tolerance range is set by the projection lens PL
Depth of focus, size of shot area, size of wafer W
Etc. are taken into consideration. Here is within tolerance
If so, the control unit 36 jumps to step 120.
The height measurement mode from the height measurement mode.
And the leveling operation is completed. Out of tolerance
If this is the case, step 110 is executed. (Step 110)   Here, open according to the instructions from the operator
Select either drive or monitoring drive
You. Therefore, first, the case where the open drive is selected is explained.
Next, the case of monitoring drive will be described. (Step 112)   First, the current surface of the wafer obtained in step 104
Based on the plane equation, the A axis which is the leveling drive axis and
Height position T of the intersection with each current plane with the B axis_a, T_bTo
calculate. The position of this intersection is the drive system 10, 11, 12
C It is known in advance by the distance to the center. further
The control unit 36 controls when the wafer surface is aligned with the target plane.
Each height position T of A axis and B axis_a′, T_b′ Is calculated similarly and
Difference from current height S_a, S_bAsk for. By this calculation
The difference between the axes A and B S_a= T_a′ -T_a, S_b= T_b′ -T_bIt is. (Step 114)   Next, the control unit 36 controls the potentiometer provided in the drive system 11.
Meter is S from current value_aServo the A axis to change only
It moves up and down by control. (Step 116)   The controller 36 is provided in the drive system 12 continuously or simultaneously.
The current value of the potentiometer_bOnly changes
Move the B axis up and down by servo control.   By the above step 116, the open control of the wafer W is performed.
Global leveling by you is over. (Step 118)   Here, the wafer surface after leveling is the best image plane.
Judge whether to check that they are parallel and check
When you click, repeat again from step 102.
You. Here, when there is no need to check,
120 is executed and a series of open control
The global leveling operation ends.   By the way, in step 110, the monitoring control
Unique when selected, unlike open control
Execute the sequence of Ie focus sensor
To drive the drive system 11 and 12 after (or during) operation.
(C) More accurate surfacing while monitoring the height position of the surface
(Posture control).   The sequence is shown in FIGS. 4 (a), (b),
The description will be made with reference to FIG. (Step 122)   First, at this point, the surface (current plane) of the wafer W is shown in FIG.
Point A in (a)₁, B₁, C₁Which is on the surface defined by the three points
And Point A₁, B₁, C₁Is a point on each axis A, B, C
is there. Similarly, the target plane is point C.₁, A_Three, B_ThreeThe three points
It should be on the surface to be defined. Also surface A₁B₁C₁Set on
3 points P_a1, P_b1, P_c1Each of which is shown in FIG.
3 measuring points P_a, P_b, P_cIt corresponds to.
This measurement point P_a, P_b, P_cAs always shown in Fig. 2
It is not necessary to specify the
It does not matter if the position is measured in. However, in reality
The measurement point monitored during monitoring control is P_a, P_b2
Only points. Incidentally, surface C₁A_ThreeB_ThreeThree points P set above
_a3, P_b3, P_c3Is each measurement point P_a, P_b, P_cOn the target plane
Represents a place.   Now, of the wafer surface obtained up to the previous step 108
Based on the first-order plane equation of the current plane, the control unit 36 controls the A-axis.
Height position (point A₁) Only the height position of the target plane (point A_Three)
Find the plane equation of the first stage when you have it. You
That is, surface C shown in FIG. 4 (b)₁A_ThreeB₁Of the first plane
Find the formula.   Next, the measurement point P closest to the A axis_aPlane C of the first stage of₁A_Three
B₁Point P above_a2And find this point P_a2Height H_aIs calculated.
This height H_aIs the amount of inclination of the herring glass 6 from the origin,
That is, is it the focus origin position set in step 100?
Equivalent to the focus offset amount. (Step 124)   Measuring point P_aPlane C₁A_ThreeB₁Upper point P_a2Height H_aIs obtained
And the control unit 36 has a height H_aFour minutes from the origin
The dust offset is given to the focus sensor
And controls the harving drive unit 38. Harving drive 38
Is based on the reading from encoder 17
Servo control of tilt drive of the space 6. Give this offset
Control unit 36, the focus sensor_aTo
The height of the Z stage remains fixed so that it can be detected.
Position the xy stage 8.   The controller 36 then measures the measurement obtained from the focus sensor.
Point P_aA so that the focus signal becomes OV according to the height of
Only the shaft (drive system 11) is driven vertically. This is
By supplying the focus signal to the servo circuit of the drive system 11,
You can drive in very accurately. As a result, the wafer surface
The first-order plane equation is the plane C in Fig. 4 (b).₁A_ThreeB₁Matches [Step 126]   Next, the control unit 36 drives only the B axis to move the wafer surface to the first position.
Surface C of Figure 4 (b)₁A_ThreeB₁To eyes as shown in Fig. 4 (c)
Level C₁A_ThreeB_ThreeIf you have it, the measurement point closest to the B axis
P_bHeight H_bIs calculated. In addition, in FIG. 4 (c), points
P_b2Is the measurement point P_bFace C₁A_ThreeB₁The position above. This point P
_b2The height of can be also calculated in advance. (Step 128)   Next, the control unit 36 detects that the focus sensor is at the measurement point P._bDetect
The height of the Z stage is fixed so that the
Position the tage 8. And even more focus
Sir is height H from the origin_bOnly offset will be set
Then, the tilt amount of the harving glass 6 is adjusted.   The controller 36 then measures the measurement obtained from the focus sensor.
Point P_bB so that the focus signal according to the height of
Only the axis (drive system 12) is servo-driven vertically. this
As a result, the first-order plane equation of the wafer surface is the plane of FIG. 4 (c).
C₁A_ThreeB_ThreeMatches   Through the above steps 122 to 128, the wafer monitoring
Global leveling by the control method
Completed and following steps 118 to similar behavior as described above
Is executed.   Next, regarding the other sequence of this embodiment, the flow of FIG.
-It will be described with reference to a chart.   In this sequence, all steps 100 to 108
Similarly, the flow after execution is different and its characteristic
Steps between steps 108 and 118
200 to 212. However, step 208 is the previous step 1
This is a combination of 12, 114, and 116, and step 21
0 is a combination of steps 122, 124, 126, and 128.
It is.   The sequence shown in Fig. 5 is especially useful when the leveling amount is large.
It is effective in the case of open control and monitoring system.
Automatically select the control and the sequence of both control methods
It is a combination of char.   The steps 200 to 212 will be described below. [Step 200]   At step 108, the current and target planes of the wafer surface are
After it is judged that the deviation amount of the
Here, check whether the amount of deviation is more than a certain value.
You. If the amount of deviation is relatively large, open control
Select it and proceed to step 202.
Select the tailing control and proceed to step 204. (Step 202)   Here, set the sequence flag to the open side.
You. (Step 204)   Here, set the sequence flag to the monitoring side.
To (Step 206)   Check the flag here, and if it is open,
Execute step 208 (steps 112-116) to global
Perform leveling. When monitoring, the step
210 (steps 122 to 128) to execute the global level
Make a ring. (Step 212)   Check the flag again here and check the level on the open side.
When the ring operation is performed, return to step 102 again.
And repeat the same operation. On the contrary, the monitoring side
If you have performed
Performs the same operation as Ming.   According to the above sequence, it is determined in step 108.
Even if the amount of deviation is extremely large, it is fast and accurate.
The wafer W can be leveled.   In step 120, the harping drive unit 38 is set to follow mode.
Wafer surface and projection lens after leveling until
Although it is exactly parallel to the best image plane of PL,
The sushi is not always in agreement. Therefore step 120
After returning to the follow-up mode with
Z focus based on the focus signal detected by the scan sensor.
By servo-controlling the
Instead, the in-focus state can always be maintained.   As described above, according to this embodiment, the target plane A_ThreeB_ThreeC₁Is
Projection lens P
The best image plane of L is tilted with respect to the moving plane of the xy stage 8.
Even if there is an image plane tilt,
It is only necessary to set the target plane according to this image plane inclination.
Because of this, the part that has poor resolution within one shot to be exposed
There is an advantage that (the four corners of the shot, etc.) disappear.   Further, in this embodiment, the height positions of three points on the wafer are measured.
I decided to specify the plane, but of course this is a monitor
Wafer set on the first stage plane during ring control
The same can be done when re-measuring the surface. For example,
Point P in Figures 4 (b) and (c)_a2, P_b2Etc. are remeasured and this point
P_a2To point P_a3Recalculate the height of point P_a2Point P against
_a3Further focus offset by the amount of shift in the height direction of
Is given to drive the A axis for monitoring, and similarly, point P_b2To
Point P_b3Focus is further increased by the amount of
It is also possible to give a facet and drive the B axis for monitoring.
Wear.   In addition, the points on the wafer W to be measured in step 102 (accurate
Is the area detected by the focus sensor) and the monitor
In this embodiment, the point of monitoring during the
Although it has been duplicated, this is a non-overlapping task.
It doesn't matter at all.   Furthermore, it uses a grazing incidence type focus sensor.
Therefore, the wafer surface height measurement is extremely accurate (for example, 0.
The surface of the wafer has the advantage of being able to achieve a resolution of about 2 μm.
The accuracy of identifying the approximate plane by measuring many points above is high.
You. Therefore, for example, as shown in FIG.
To P₁~ P_FifteenSet the measurement points (measurement shots) up to
Included on the surface of the required area or the surface of the important area among the measurement points.
By applying a predetermined weight to the height measurement value of the point
The best fit to a part of the wafer surface
The plane (first-order plane equation) can be determined. Sixth
Measurement point (shot) P in the figure₁Is located almost at the center of the wafer W
Place, point (shot) P_Two, P_Three, P_Four, P_FiveIs point P₁From almost equal
It is defined to be located in four directions by a distance, and
G) P₆~ P_FifteenShould be located near the perimeter of the wafer surface
Determined. Here, due to the influence of the wafer process,
Consider the case where Wha W is curved as shown in FIG.
FIG. 7 shows the wafer W of FIG. 6 as a point (shot) P.₁₃, P_Four, P
₉Represents a cross section when cut by a line including. Wafer process
Point P₁₃A warpage occurs in the vicinity (on the left side of the wafer),
Point P₉If a warp occurs in the vicinity (on the right side of the wafer),
Each measurement point P by the measurement in step 102₁~ P_FifteenFrom the height of
Current plane A of the wafer surface obtained by the small square method₁B₁C₁Is the
As shown in Fig. 7, in the flat part near the center of the wafer surface
On the other hand, it will be slightly inclined. There u
Measurement point around the warped area, such as P₇,
P₈, P₉, P₁₃, P₁₄Least squares for height measurements at etc.
The weighting is reduced when calculating the modulus, and the point P near the center is₁~ P_Five
Weighting is increased for the height measurement value at.
In this way, the wafer peripheral area with poor yield will be
The center of the wafer, which has a good yield and is not affected by
The leveling will be focused.   Also, depending on the wafer, as shown in FIG.
In some cases, the central part is bulging and leaning. FIG. 8 is the sixth
Point P in the figure₁, P_Two, P_Four, P₆, P₁₁Cut the wafer at the line containing
Represents a cross section. For such wafers, the least squares method
The wafer surface that is approximated by the first-order plane equation is
A₁, B₁, C₁Partly parallel to the wafer surface
However, in many parts it is not always parallel
Absent. Therefore, in order to deal with such a situation, glow
Block leveling in addition to val leveling
It is good to have a can. What is block leveling?
Divide the wafer surface into several blocks, each block
Leveling based on the first-order plane equation approximated to
It's something that gets rid of. For example, in FIG. 6, point P₁,
P_Two, P_Three, P₆, P₇, P₈The first block containing P and the point P₁,
P_Three, P_Four, P₉, P_Ten, P₁₁2nd block containing point P₁,
P_Four, P_Five, P₁₁, P₁₂, P₁₃A third block containing
P₁, P_Two, P_Five, P₆, P₁₄, P_Fifteen4 with a fourth block containing
Divide into blocks. And first of all the measurement points in each block
After measuring the height in order in step 102,
Find the first-order plane equation for each square. As a result, for example, the first
For blocks, the primary plane RP as shown in Fig. 8₁But
Determined, for the second block, the primary plane RP_TwoIs decided
Is determined.   After the primary plane for each block is determined, the step
When each shot is exposed by the repeat method,
Find out which block the yotto belongs to, and
So that the primary plane of is parallel to the best image plane (target plane)
For example, if you perform leveling with open control method,
No. This block leveling is an alignment before the exposure operation.
Associated with multiple shot areas on the wafer
A sample error that detects the position of the alignment mark
Element should be executed in the same way
New   In addition, by measuring 3 points on the wafer and using open control.
After performing global leveling with
Measures a fixed point for exposure or sample alignment
Block leveling may be performed.   In addition, the throughput decreases slightly, but
In order to perform the belling with higher accuracy, for example, as shown in FIG.
You may provide a sequence as shown in the flow chart.
No. Figure 9 shows monitoring control during block leveling.
Efficient system precision and open control system high speed
Is a schematic representation of the combined sequence.
You.   First, in step 220, the first-order plane equation for each block is calculated.
Therefore, in step 222, for one block, A axis, B
The axis is monitored and controlled, and the current
Accurately match the face with the target plane. Focus at this time
The points to be monitored by the sensor are included in the block.
It is preferable to use at least two points that are rare. That
Leveling of blocks by monitoring control
When completed, in step 224 the A-axis and B-axis (drive
Read the height position of the dynamic system 11, 12) from the potentiometer
Remember This operation for all blocks on the wafer
Execute (step 226), leveling amount for each block
(Potentiometer value) is stored in advance. This level
Ring amount is determined by monitoring control
It is extremely precise. Next, sample alignment, or
If you move to the exposure operation, write it as in step 228.
Wafer for each block based on the stored leveling amount
It can be tilted by open control.   In addition, when performing block leveling,
If you increase the number of clicks,
Leveling, so-called each leveling (or Die by Die level
An operation corresponding to "belling" is also possible.   In this embodiment, the approximate plane equation of the wafer surface is 2
Since the following equations can also be used, for example, the waveform shown in Fig. 10 can be used.
If the wafer surface is wavy like C, many
Measurement point P (4 points or more)_iHigher plane based on the height value of
A₁B₁C₁′ May be determined. In this case, face A₁B₁C₁′
Although it takes some time for the calculation process to be set, surface A₁B₁C₁′
Once the
Since the most approximated inclination amount can be found immediately,
It is possible to perform leveling for each otto. this
The same applies to the wafers shown in FIGS. 7 and 8.
This is true.   By the way, the position of the xy stage 8 is measured by the interferometer 32.
However, the lateral shift amount of the wafer W due to the leveling operation is
The structure that cannot be detected by the interferometer 32 is one
General. It receives the laser beam from interferometer 32.
No moving mirror is installed on the leveling stage 9.
to cause.   Therefore, in the xy direction of the wafer W before and after the leveling operation,
Correct overlay exposure when the position is out of order
Can not be done. Of course global leveling or
After performing lock leveling, align on the wafer.
If you try to measure the position of the
No such problem arises. However, depending on the sequence
Perform sample alignment and
When the exposure position is determined and when the shot is actually exposed
The leveling amount may differ depending on the time. So
In this case, the difference in leveling amount
Calculate the amount of lateral deviation or calculate the lateral deviation amount
Move the sensor on leveling stage 9 and xy stage 8.
It may be provided between the mirror and the mirror.   In the global leveling described above, one wafer
Although I described the operation for Ha,
When processing wafers continuously, only the first one in the lot
Performs global leveling (or block leveling)
No, fix the A and B axes for subsequent wafers
It can also be used as is.   For the top two to three wafers in the lot,
Perform a ring operation to process and then verify each wafer.
And average the leveling amount
It is also possible to process subsequent wafers in quantity.   Operation of steps 102 and 104 in the embodiment described above
When there are 3 measurement points on the wafer, the 3 points are representative.
Present plane A₁B₁C₁When four or more measurement points are specified
Is automatically the current plane A by the method of least squares.₁B₁C₁Identify
You may do so.   Also, when the measurement point on the wafer is determined by the control unit 36,
Normally, the focus sensor is located at the center of the shot area.
Since it works to detect, the control unit 36 should be set first.
The central part of the shot area that has been placed protrudes from the wafer outline.
Check whether or not it is, and if it sticks out, it should be the measurement point.
Auto addressing for resetting the wafer area inside the wafer
It also has a ringing function.   Further, in the present invention, the best image plane to be the target plane
The method of determining the
4 in a shot area
Examine the resolution at the corners and maximize the resolution (best focus)
The height position in the Z direction (focus direction) is
You can use the method of asking about each person. For example in Figure 11
Shot by piecing for test chart as shown
The best focus position (height) at the four corners is Z₁, Z_Two, Z_Three, Z_Four
Suppose that you are asked for. In this case, the target plane A_ThreeB_ThreeC₁
The first-order plane equation that determines Z = ax + by + γ (only
, And a, b and γ are constants, the constants a and b are Is represented by Where L is in the x and y directions of the shot area
Equivalent to the length.   Of course, by trial baking, the shot area
To find the best focus position for multiple points,
You can determine the n-th order plane equation of the target plane by multiplication etc.
No. 〔The invention's effect〕   As described above, according to the present invention, the height position of the substrate is monitored.
However, if the leveling is performed in two steps, the substrate surface will be accurate.
Can approach the target plane. Therefore, during exposure,
It is possible to obtain the theoretical resolution of the device.   According to the invention described in the dependent claims, the projection level
In the exposure equipment with a lens, N.A.
If you have a lens with a shallow depth of focus,
By specifying the target surface according to the image plane tilt of the lens,
You can effectively use the shallow depth of focus.   In addition, the surface of the required area on the wafer or the important area
You can bring the surface closer to the target surface.   Also, taper of substrates such as individual wafers (or masks)
The effects that were being caused by
You.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram showing a configuration of a projection exposure apparatus incorporating a substrate attitude control apparatus according to an embodiment of the present invention, and FIG. 2 is a plan view showing a planar configuration of a leveling stage section. 3 is a flow chart for explaining the basic operation of this embodiment, and FIG.
Figures (a), (b), and (c) are perspective views showing a model of how the wafer leveling is executed by the monitoring control method, and FIG. 5 shows a sequence in which open control and monitoring control are combined. Flowchart diagram, 6th
The drawing is a plan view showing the arrangement of measurement points on the wafer, FIG. 7,
FIG. 8 is a diagram showing an example of a wafer cross-sectional shape, FIG. 9 is a flowchart of a sequence suitable for block leveling, FIG. 10 is a diagram showing an example of a wafer cross-sectional shape, and 11
The figure is a plan view showing a method for determining a target plane. [Explanation of Signs of Main Parts] PL ... Projection lens, W ... Wafer, 6 ... Harving glass, 7 ... Projection objective lens, 8 ... xy stage, 9 ... Leveling stage, 10, 11, 12 …… Leveling drive system, 13 …… Receiving objective lens, 16 …… Focus sensor photoelectric element, 34 …… Leveling drive section, 38 …… Herving drive section.

Continuation of front page    (56) References JP-A-60-109226 (JP, A)                 JP 60-221759 (JP, A)                 JP 62-58624 (JP, A)                 JP-A-62-208629 (JP, A)                 JP 59-47731 (JP, A)                 62-169353 (JP, U)

Claims (1)

(57) [Claims] At least three height positions of the first height position adjusting mechanism, the second height position adjusting mechanism, and the third height position adjusting mechanism are detected by detecting the position of the substrate in the direction of the gap between the predetermined reference plane and the substrate. An exposure method for adjusting a tilt of the substrate by an adjusting mechanism to form a mask pattern on the substrate, detecting positions in the interval direction at a plurality of measurement points on the substrate, and obtaining an approximate plane of the substrate. And a plane of the approximate plane of the substrate in a state where the intersection of the approximate plane of the substrate and the drive axis of the driving height position mechanism is driven by the first height position adjusting mechanism to coincide with the target plane. A step of obtaining an expression; a first measurement point of the first height position adjusting mechanism is arranged so that a first measurement point on the substrate corresponding to a drive axis of the first height position adjustment mechanism coincides with a plane defined by the approximate expression. The above while detecting the position Driving the height position adjusting mechanism of the second height position adjusting mechanism; and detecting the height position of the second measurement point on the substrate corresponding to the drive axis of the second height position adjusting mechanism, And a step of driving the second height position adjusting mechanism so as to coincide with the target plane. 2. The pattern of the mask is formed on the substrate via a projection optical system, and the target plane is set according to a state of an image forming surface of the projection optical system. The exposure method described in. 3. When the pattern of the mask is formed in a plurality of shot areas on the substrate, a plurality of height positions in the shot area are detected in the step of obtaining the first approximate surface. The exposure method according to item 1. 4. The first approximation surface is obtained by weighting detection values at the plurality of height positions.
And the exposure method according to item 3. 5. The first measurement point is located between the center position of the area surrounded by the drive shafts of the three height position adjusting mechanisms and the drive shaft of the first height position adjusting mechanism, and the center position and the Second
The exposure method according to claim 1 ', wherein the second measurement point is located between the height position adjusting mechanism and the drive shaft. 6. At least three height positions of the first height position adjusting mechanism, the second height position adjusting mechanism, and the third height position adjusting mechanism are detected by detecting the position of the substrate in the direction of the gap between the predetermined reference plane and the substrate. In a substrate attitude control method for adjusting the inclination of the substrate by an adjusting mechanism, a step of detecting positions in the interval direction at a plurality of measurement points on the substrate to obtain an approximate plane of the substrate; Driving one of the position adjusting mechanisms to obtain a plane formula of the approximate plane of the substrate in a state where the intersection of the approximate plane of the substrate and the drive axis of the driven height position mechanism is coincident with the target plane. And; detecting the position of the first measurement point so that the first measurement point on the substrate corresponding to the drive axis of the first height position adjustment mechanism coincides with the plane defined by the plane formula. While the first height position adjusting mechanism A step of driving; so that the second measurement point coincides with the target plane while detecting the height position of the second measurement point on the substrate corresponding to the drive axis of the second height position adjustment mechanism. And a step of driving the second height position adjusting mechanism.