JPH0547625A - Projection exposure apparatus - Google Patents

Abstract

PURPOSE:To provide a sufficent effect of enlargement of focal depth. CONSTITUTION:A driving unit 15 controls a Z stage 14 in a given working characteristic so that distribution (existence probability) of optimum imaging faces in an optical axis has approximately maximum values in at least two points when a projecting lens PL is used with a shutter 3 during the time from an opening start point t0 to a closing finish point t5. Then, a lamp control unit 23 changes the power supplied for a mercury lamp 1 according to the distribution (existence probability) of optimum imagery faces during the time (tfu-tfd) from the opening start point t0 to the closing finish point t5 so that a reticle R is irradiated with a reduced or increased light.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention relates to a semiconductor circuit pattern,
The present invention relates to an apparatus for projecting and exposing a liquid crystal display element pattern or the like onto a photosensitive substrate.

[0002]

2. Description of the Related Art Conventionally, in this type of projection exposure apparatus, the surface of a photosensitive substrate (semiconductor wafer or glass plate coated with a resist layer) is arranged on the best image plane of the projection optical system (the plane conjugate with the reticle). The reticle pattern was exposed in this state. However, the area (shot area) exposed on the wafer at one time is 15 mm square to 20 mm.
If there is a corner and the wafer surface is slightly curved in that area, or if there is unevenness of several μm due to the surface structure, a portion exceeding the depth of focus of the projection optical system appears in the shot area. This is because the depth of focus of the projection optical system is only about ± 1 μm on the image side (wafer side).

Therefore, a method has been proposed in, for example, Japanese Patent Laid-Open No. 63-42122, which allows an exposure apparatus having a projection optical system having a small depth of focus to effectively perform exposure with a wide depth of focus. In the method disclosed in this publication, the wafer is moved to two or three points in the optical axis direction of the projection optical system, and the same reticle pattern is multiple-exposed at each point. In this method, by setting two points apart in the optical axis direction to a width of the depth of focus ± ΔZ of the projection optical system,
The effective depth is increased by 1.5 to 3 times. In addition to the method of positioning the wafer at each of the multiple points in the optical axis direction and repeating the exposure as in the above publication, the wafer is continuously moved in the optical axis direction during one exposure operation for one shot area. A method of moving (or vibrating) is also proposed.

[0004]

A semiconductor integrated circuit is manufactured by repeating steps such as film formation, pattern transfer, and etching several times to more than ten times. Therefore, on the surface of the wafer in the process of forming the integrated circuit, a portion where films of several layers are stacked or a portion where no films are stacked at all are mixed in some cases. The thickness of one layer of the film is 0.1 μm to 1 μm
The level difference on the wafer surface (within one shot area) may be several μm at maximum.

On the other hand, the depth of focus of the projection optical system is generally ± λ / (2 · NA ² ) where λ is the wavelength of the illumination light for exposure and NA is the numerical aperture on the image plane side of the projection optical system. Is expressed as In recent projection optical systems, λ = 0.365 μm
(I-line of mercury lamp) with NA ≈ 0.5 is made, and the depth of focus Δ ± Z in this case is about ± 0.73 μm.

Therefore, when the wafer is fixedly placed on the best image plane of the projection optical system and exposure is performed as in the conventional case,
Upper (top) and lower (bottom) steps on the wafer
Are both ± with respect to the best image plane (best focus plane)
At a focal depth of 0.73 μm or more, the focal point is separated in the optical axis direction, which makes image formation impossible. Here, the state of exposure by a multiplex (here, three times) exposure method in which a conventional wafer is sequentially positioned at multiple discrete points in the optical axis direction and exposed a plurality of times will be briefly described with reference to FIG. FIG. 7 (A),
(B) and (C) are focus position -Z ₁ ,
It shows a state of multiple exposure at Z ₀ and + Z ₁ , where the lower part of the pattern step on the wafer is represented by (A), the middle part is represented by (B), and the higher part is represented by (C). The multiple exposure pattern is a hole pattern formed on the reticle (a reticle pattern of a minute transmissive part on the base of the light shielding part). The intensity distribution of the projected image of this hole pattern on the wafer is shown by Fmn, and m = 1 is the focus position −Z.
₁ and m = 2 represent the focus position Z ₀ , m = 3 represents the focus position + Z ₁ , and n = 1, 2 and 3 respectively represent the number of exposures (order).

First, when the best image plane of the projection lens is focused on the lower portion (position -Z ₁ ) of the pattern step on the wafer and the first exposure is performed, a sharp intensity is formed in the lower portion. The distribution F ₁₁ is imaged, but as shown in FIGS. 7B and 7C, the intensity distribution F ₁₁ ′ rapidly deteriorates (decreases the peak value and increases the width) toward the middle part and the high part. To do. Next, when the best image plane is focused on the center of the pattern step (position Z ₀ ) and the second exposure is performed, a sharp intensity distribution F ₂₂ is imaged in the middle and low parts, but in the low part and the high part. A deteriorated intensity distribution F ₂₂ 'appears in each of the parts. Similarly, in the third exposure,
Since the best image plane is focused on the high part (position + Z ₁ ) of the pattern step, a sharp intensity distribution F ₃₃ is obtained at the high part, and the intensity distribution F ₃₃ 'deteriorated toward the middle part and the low part.
Is obtained. When the exposure is repeated three times in this way, sharp image distributions F ₁₁ , F ₂₂ , and F ₃₃ can be obtained once in any of the low portion, the middle portion, and the high portion of the pattern step.

As shown in FIG. 7 (A), the intensity distributions F ₁₁ , F ₂₂ 'and F ₃₃ ' are given to the lower resist, and the integrated light amount distribution is shown in FIG. 7 (D). become that way. In the figure, a level Eth indicated by a broken line is an exposure amount required to remove the positive resist (form a hole pattern). Similarly, for the resist in the middle of the pattern step, the integrated light quantities of the intensity distributions F ₂₂ , F ₁₁ 'and F ₃₃ ' are shown in FIG.
As shown in FIG. 7 (E), the integrated quantity of light of the intensity distributions F ₃₃ , F ₁₁ ′ and F ₂₂ ′ is given to the resist at the high portion of the pattern step as shown in FIG. 7 (F). Therefore, in any of these three step portions, the resist gives a good image intensity distribution of the hole pattern, and as a result, over the width of 2Z ₁ from the high portion to the low portion of the step,
The apparent depth of focus will be expanded.

However, according to the multiple exposure method as described above, it is possible to cope with a step difference of several μm on the wafer, but a shutter system is used to stop and restart the exposure at multiple points in the optical axis direction. Drive the stage (Z stage) in the optical axis direction (Z direction) of the wafer holder,
There is a problem that the wafer processing capacity per unit time is significantly reduced due to the influence of the positioning operation characteristic and the shutter opening / closing operation characteristic.

Therefore, in addition to the method of performing multiple exposure while discretely changing the focus position as described above, a method of performing exposure while continuously moving the wafer in the optical axis direction is conceivable. A magnifying effect is obtained. However, as a result of experiments and the like, it was found that even if the wafer is randomly moved (or vibrated) in the optical axis direction during wafer exposure (during shutter opening), the desired expansion effect cannot be obtained.

Assuming that the Z stage holding the wafer is driven at a constant speed as a generally considered moving method, the distribution of the best focus image in the optical axis direction, in other words, the best focus in the optical axis direction. The existence time per unit position of the image, that is, the existence probability is as shown in FIG. The vertical axis in FIG. 6 represents the focus position (position in the optical axis direction of the wafer), and the horizontal axis represents the existence probability (probability that the best focus image stays). Therefore, in multiple (for example, three times) exposure in which the wafer is positioned at multiple points separated in the optical axis direction and exposed a plurality of times, focus position + Z ₁ ,
The existence probability takes a value other than zero only at three points of Z ₀ and -Z ₁ , but ± Z ₁ when the Z stage moves at a constant speed.
There is a problem in that the existence probability becomes constant over the entire range, and in addition to the effect of increasing the depth of focus, the disadvantage that the contrast of the exposed resist pattern deteriorates appears remarkably. In the method of vibrating the wafer in the optical axis direction during wafer exposure, the effect of increasing the depth of focus may not be sufficiently expected due to the amplitude, frequency, vibration waveform of the vibration, exposure time, and the like.

[0012]

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and provides a projection exposure apparatus capable of substantially increasing the depth of focus and at the same time reducing the decrease in throughput. With the goal. In order to achieve this object, the present invention is configured as follows.

That is, the existence time per unit position (presence probability) in the optical axis direction of the best imaging plane of the projection optical system from the opening start time to the closing end time of the shutter is at least at two points in the optical axis direction. Stage control means for controlling the movable stage (Z stage) so that the maximum values are substantially equal; only a predetermined time between the opening start time and the closing end time of the shutter according to the existence probability of the best imaging plane ,
An illumination light intensity adjusting unit that attenuates or amplifies the illumination light that illuminates the mask (reticle) is provided.

As the illumination light intensity adjusting means, a method of reducing the power supplied to the light source for generating the illumination light, a method of inserting a high-speed variable dimmer into the illumination light path, or the like can be considered. In particular, when a mercury discharge lamp is used as a light source for illumination light, the responsiveness of changes in emission brightness to changes in supplied power is 1
It is extremely fast at about msec, which is sufficiently shorter than the shutter operation time. Therefore, even if the electric power supplied to the discharge lamp is largely changed while the shutter is open, the exposure amount can be controlled with sufficient accuracy.

[0015]

In the present invention, a system (hereinafter referred to as a progressive focus exposure method) in which the photosensitive substrate and the best image plane are relatively continuously moved in the optical axis direction while the shutter is open is adopted. Has a special relationship between the timing of changing the intensity of the illumination light during the opening / closing operation of the shutter and the movement control characteristic in the optical axis direction.

The manner of exposure by the progressive focus exposure method adopted in the present invention will now be described with reference to FIG. Here, it is assumed that the pattern to be exposed is a hole pattern (a pattern in which a minute transmissive portion is formed on the base of the light shielding portion). FIG. 5A shows a Z position change pattern Z (t).
FIG. 5 (B) schematically shows the existence probability (presence time per unit position of the best imaging plane) in the Z direction (optical axis direction) when exposure is performed with the characteristics shown in FIG. 5 (A). FIG. 5C is a schematic diagram, and FIG. 5C shows the exposure amount at or near the center position of the hole pattern in the resist layer on the wafer (for example, in the exposure amount distributions shown in FIGS. 7D to 7F). 3 schematically shows a distribution of the exposure amount at the peak position) in the optical axis direction (thickness direction of the resist layer) (hereinafter, simply referred to as a light amount distribution in the resist layer).

[0017] Now, between the opening start time t _a of the shutter to close the end time t _b, at a rate characteristic of the two types indicated by the solid line and one-dot chain line in FIG. 5 (A), the Z stage position Za If it is assumed that the object moves from Z to Ze, the existence probability as shown in FIG. 5B is obtained. The existence probabilities actually shown in FIG. 5 (B) are the positions Za to Zb and the positions Zd to Z.
Since there is a peak (maximum value) in the vicinity of e, it is possible to increase the depth of focus. Here, since the Z stage is continuously moved, the existence probability cannot be zero at the position between peaks, for example, Zb to Zd.

The amount of movement of the Z stage during exposure, that is, the distance between the positions Za and Ze, is the resist layer when the existence probability shown by the solid line in FIG. 5C is obtained. It is assumed that the light intensity distribution at is almost uniform. Further, in FIG. 5C, the position RZ ₁
Indicates the position of the boundary surface between the resist layer and the wafer, and the position R
Z ₂ indicates the position of the surface of the resist layer.

On the other hand, since the existence probability shown by the alternate long and short dash line in FIG. 5B also has two peaks, the depth of focus can be similarly expanded, but the existence between the two peaks is larger than that of the solid line. The probability is high. Therefore, in such a case, the distribution of the light amount of the resist layer is not uniform as shown by the alternate long and short dash line in FIG. 5C, and the effect of increasing the depth of focus is not so large as compared with the case of the solid line. I understand. This means that depending on the Z stage, only the characteristics shown by the alternate long and short dash line in FIG. 5 (A) may be obtained, and in such a case, it is not possible to expect a significant increase in the depth of focus. ing.

In order to enhance the effect of expanding the depth of focus, the amount of movement of the Z stage during exposure is determined so that the light amount distribution in the resist layer becomes substantially uniform as shown by the solid line in FIG. Is desirable. However, as the level difference (unevenness) of the wafer (resist layer) increases,
As a matter of course, the movement amount of the Z stage must be made larger than the appropriate value. On the contrary, when the step is small and it is desired to increase the throughput even a little, the movement amount of the Z stage may be made smaller than the appropriate value. In the former case, the exposure amount (exposure amount) in the thickness direction of the resist layer becomes small near its center, and the light amount distribution in the resist layer is not substantially uniform. On the other hand, in the latter case, the exposure amount (exposure amount) in the vicinity of the center of the resist layer is larger than that in other portions, and the light amount distribution in the resist layer is convex as in the one-dot chain line in FIG. 5C. Become a state. Therefore, in either case, the contrast of the resist pattern may be deteriorated, and the effect of increasing the depth of focus may not be sufficiently expected.

Therefore, in the present invention, the focus position + Z ₁
And -Z ₁ at two points (for example, at two points separated by the width of the depth of focus of the projection optical system) are maximized to be approximately equal, and in the middle thereof, the contrast of the exposed resist pattern is not deteriorated. In order to suppress the existence probability to a low degree, while controlling the continuous movement of the photosensitive substrate and the best imaging plane in the optical axis direction while the shutter is open, at least depending on the existence probability of the best imaging plane, The intensity of illumination light to the mask is controlled to be attenuated or amplified only for a predetermined time while the existence probability is kept low. In order to realize this, in the following embodiments, the intensity of the illumination light to the mask is controlled to be switched in at least two stages. Here, particularly when exposing a hole pattern (a pattern in which a minute transmissive portion is formed on the base of a light shielding portion), the illumination light intensity adjusting means determines the center position of the hole pattern or its vicinity according to the existence probability of the best image plane. Exposure amount (sensitivity amount)
The intensity of the illumination light to the mask is controlled to be attenuated or amplified for a predetermined time so that the distribution in the direction of the optical axis of (1) (light amount distribution in the resist layer) becomes substantially uniform.
As a result, the movement characteristics of the Z stage and the positions + Z ₁ and −
Regardless of the distance from Z ₁ and the like, the light amount distribution in the resist layer can always be made substantially uniform, and the effect of expanding the depth of focus in the progressive exposure method can be sufficiently obtained.

[0022]

1 is a diagram showing the construction of a projection exposure apparatus according to an embodiment of the present invention. In FIG. 1, the illumination light from the mercury discharge lamp 1 is focused on the second focus by the elliptical mirror 2. The second
A rotary shutter 3 having a plurality of blades is rotatably arranged by a motor 4 near the focal point. The illumination light condensed at the second focus and transmitted through the shutter 3 enters an optical integrator 6 including an input lens 5, a fly-eye lens and the like. A plurality of secondary light source images are formed on the exit surface side of the optical integrator 6 to form a surface light source. The illumination light from each of the secondary light source images is
Mirror 7 having a slight transmittance (for example, about 10%)
After being reflected by, the light enters the first relay lens system 8, is integrated on the surface of the reticle blind 9, and illuminates the reticle blind 9 with a substantially uniform illuminance distribution. The reticle blind 9 is for limiting the illumination range on the reticle R, and the aperture image of the blind 9 is the second relay lens system 1
0, through the mirror 11 and the condenser lens 12,
An image is formed on the pattern surface of the reticle R.

The illumination light transmitted through the transparent portion of the pattern of the reticle R reaches the wafer W through the projection lens PL, and a projected image of the reticle pattern is formed near the surface of the wafer W. The wafer W is vacuum-sucked on a holder (not shown) of the Z stage 14 arranged on the XY stage 13. The Z stage 14 is moved in the optical axis direction of the projection lens PL by a drive unit (motor and tachogenerator) 15 on the XY stage 13. Normally, the Z stage 14 includes a projector 16
And an optical receiver 17 for oblique incidence light type focus detection system (A
Under the control of the AF unit 18 which receives the focus signal from the F sensor, the driving unit 15 is servo-operated to move in the optical axis direction. This is an autofocus operation performed during normal single exposure. Further, the photoelectric sensor 1 that detects the intensity of the illumination light (the pattern image of the reticle R or the opening image of the reticle blind 9) that has passed through the projection lens PL on the XY stage 13.
9 is provided.

Here, the drive unit 15 in this embodiment is
When the Z stage 14 is moved for progressive focus exposure, for example, the existence probability as shown by the solid line in FIG. 5B (existence time per unit position in the optical axis direction of the best imaging plane).
It is assumed that the Z stage 14 is driven so that Therefore, the drive unit 15 corresponds to the stage control means in the present invention. The amount of movement of the Z stage 14 during exposure is determined according to, for example, a step (unevenness) in one shot area (resist layer) on the wafer W,
In this embodiment, the AF unit 18 controls the drive unit 15 based on the focus signal from the light receiver 17 to accurately set the amount of movement.

By the way, behind the mirror 7, there is arranged a photoelectric sensor 20 which receives a part of the illumination light which has passed through the shutter 3 and detects the light intensity. It is input to the integrator circuit 21 that is performed on a regular basis. The integrator circuit 21 outputs a signal IL obtained by amplifying the photoelectric signal by a predetermined amount to the main control system 100. The signal IL is proportional to the illuminance on the reticle R or the illuminance on the wafer W. Further, the integrator circuit 21 includes a relatively circuit that outputs the closing signal Sc of the shutter 3 when the value of the proper exposure amount preset by the main control system 100 and the value obtained by integrating the previous light amount match. I'm out. This closing signal Sc is also sent to the shutter driver 22, whereby the motor 4 is rotated by a fixed amount, and the shutter 3 is closed. Further, the driver 22 exchanges an opening command for the shutter 3 and a shutter status signal with the main control system 100. In the case of normal single exposure, the photoelectric sensor 20, the integrator circuit 21, the shutter driver 22, and the motor 4 are used.
The automatic exposure / control is performed by the loop.

On the other hand, the mercury discharge lamp (hereinafter, simply referred to as a lamp) 1 has its power supply controlled by a lamp control unit 23. In recent years, the stepper has adopted a flash exposure method in which the electric power supplied to the lamp 1 is increased to about twice the rated electric power only while the shutter 3 is open in order to increase the exposure throughput. The lamp control unit 23 is provided with a command for switching between the normal exposure method and the flash exposure method from the main control system 100, and increases the power supplied to the lamp 1 in conjunction with the opening start of the shutter 3 during the flash exposure. In this embodiment, the lamp control unit 23 functions as the illumination light intensity adjusting means of the present invention, and the lamp 1 is opened in the shutter open state.
Control the emission brightness of.

The XY stage 13 in FIG.
Each of the plurality of shot areas above is stepped so as to move directly below the projection lens PL one after another, but this is not directly related to the present invention, so here, the drive unit of the XY stage 13 and the control system are illustrated. Omitted. Next, the functional blocks of the main control system 100 relating to the main part of the present invention will be described with reference to FIG. The function of each block in FIG. 2 is achieved by either hardware of an electric circuit or software such as a microcomputer.

In FIG. 2, an analog-to-digital converter (ADC) 110 inputs the amplified photoelectric signal IL from the photoelectric sensor 20 and converts the level of the signal IL into a digital value at every constant sampling time. The converted digital value is stored in the memory (RAM) 111 in the order of addresses. The address value of RAM111 is counter 1
The counter 112 counts the number of pulses generated by the clock generator 113 only while the clock pulse CKP generated by the clock generator 113 passes through the gate 114. Opening and closing of the gate 114 are switched by a signal CS ₁ from the microprocessor (μP) 150.

This μP 150 is connected to the bus lines DB ₁ and D
The AF unit 18 and the shutter driver 22 shown in FIG. 1 are also mutually connected via B ₂ and exchange data required for control. A data input unit (or man-machine interface) 120 inputs commands and data from an operator, and main commands are a switching command between progressive focus exposure and normal exposure and an exposure start command. The data is the proper (target) exposure amount per shot and the movement width of the Z stage 14 during progressive focus exposure. The initial characteristic data section 130 of the Z stage is mainly composed of Z
Data of the moving speed characteristic of the stage 14 (maximum speed value, acceleration value, etc.) is stored. This speed characteristic data is
The output value from the tacho-generator provided in the drive unit 15 of the Z stage 14 is converted into an A / D value at regular sampling intervals.
It is converted to a digital value by a converter and read to the memory inside the μP150 via the bus line DB ₁ , then μP15
It can be easily obtained by analysis of 0. μP15
The data obtained by 0 is stored in the initial characteristic data section 130. However, since the Z stage 14 may have different speed characteristics when moved from the bottom to the top and when moved from the top to the bottom, both of them may be used. It is preferable to obtain and store the speed characteristic in the case of.

Further, the μP 150 controls the output stop of the signal CS ₁ to the gate 114 based on the input of the opening signal Sc of the shutter 3. Furthermore, μP150 is bus line D
Various warnings are output to the operator via B ₃ . One of the warnings is that it is difficult to perform the progressive focus exposure under the exposure condition input to the data input unit 120, and therefore some condition needs to be corrected. Then, the μP 150 outputs to the lamp control unit 23 an instruction to change the supplied power and information on the power value to be changed.

Next, the basic operation of this embodiment will be described.
In the present embodiment, the description will be made assuming that the light amount distribution in the resist layer is not uniform and has a convex distribution as shown by the one-dot chain line in FIG. 5 (C). To In addition, here, the pattern to be exposed is described as a hole pattern (white small rectangle). First, it is assumed that the shutter 3 in FIG. 1 is closed, the lamp control unit 23 is set to the flash exposure method, and the integrated value of the light amount in the integrator circuit 21 is reset to zero. ..

An example of the flash exposure method executed with the shutter open will be described with reference to FIG. FIG. 3A is an example of shutter opening / closing characteristics,
At time t _SO , an opening command is output to the shutter driver 22, and at the same time, the signal CS ₁ is output from the μP 150. At time t ₀ , the shutter 3 starts to open, the illuminance on the reticle R rises from zero, and the shutter 3 is fully opened after a fixed time. At this time, as shown in FIG. 3 (B), the lamp control unit 2
3 is immediately before time t _{SO when} the shutter open command is output,
The power supplied to the lamp 1 is increased from the rated power Pm to the flash power Ph (Pn <Ph <2Pn) to increase the light emission intensity of the lamp 1. After that, the power supplied to the lamp 1 is reduced to the rated power Pn from the time t _fd to the time t _fu while the shutter 3 is fully opened. Then, from the time t _fu to the time t ₅ when the closing of the shutter 3 is completed, the lamp supply power is increased to the flash power Ph again, and after the time t ₅ , the supply power of the lamp 1 is restored to the original value. Return to the rated power Pn. The closing signal (SC ₁ ) of the shutter 3 is 2
Time t after time t _fu when the flash is flashed up for the second time
Output to _sc .

In this embodiment, the amount of movement of the Z stage 14 during exposure (corresponding to the distance between the positions Z ₀ and Z ₅ described later) is determined according to the step on the wafer W. However, it is assumed here that this is solved by switching the rated power Pn and the flash power Ph as described above. In other words, the value of the power Ph for flash near the position where the existence probability has a peak and the value of the rated power Pn at the intermediate position are determined so that the light amount distribution in the resist layer becomes substantially uniform. Has been.

Now, with such a control system, the time variation characteristic of the illuminance given on the wafer W is as shown in FIG. 3 (C). The total exposure amount given to one shot area on the wafer W is the integration of the illuminance from time t ₀ to t ₅ in the characteristic of FIG. Therefore, it suffices to control the exposure amount given by the characteristics of FIG. 3C so as to match the appropriate exposure amount (or a value set by the operator) for the wafer W, whereby the exposure amount can be easily controlled. It is possible to keep the accuracy within a desired allowable range.

In FIG. 3, the time t ₀ is accompanied by a certain delay with respect to t _so . This is due to the mechanical response delay of the shutter 3 and the electrical response delay in the shutter driver 22. Because it exists. Further, in the rotary shutter, the opening operation (rising) characteristic and the closing operation (falling) characteristic of the shutter are almost symmetrical, but they do not always match.

In the progressive focus exposure of this embodiment, the Z stage 14 is driven so that the existence probability reaches a peak at each of two points separated in the optical axis direction during exposure. It is set between time t ₀ and t _fd and between time t _fu and t ₅ . Therefore, when the shutter 3 is in the open state (time t _{0 to} t ₅ ), the Z stage 1
4 is moved at a time t _fu between time t _{0 and} t _fd.
It is late between ~ t ₅ and early between times t _{fd and} t _fu . That is, while the shutter is open, the Z stage 14 is not moved at a constant speed, but is accelerated and decelerated.

First, prior to the actual exposure operation, as shown in FIG.
It is confirmed by opening and closing the shutter whether or not the proper exposure amount can be obtained with the characteristic as shown in (C). In this case, as one condition example, time (t _fd −t ₀ ) and time (t ₅ −t _fu ).
It should be set so that and are almost equal. However, in order to stabilize the control accuracy, both the flash down time t _fd and the flash up time t _fu are set so as to occur while the shutter is fully open. Then, with the wafer W not under the projection lens PL, the shutter 3
Is driven in the control mode as shown in FIG. 3, and the integrator circuit 21 detects the integrated exposure amount from time t _{0 to} t ₅ . The integrated exposure amount is read by the μP 150 and compared with the appropriate exposure amount. If there is an excess or deficiency more than the allowable value with respect to the proper exposure amount, the shutter time (t _sc
-T _so ) is readjusted and the flashdown time (t _fu -t _fd ) is adjusted. Then, dummy exposure is performed again to confirm whether or not an appropriate exposure amount can be obtained. These confirmation and adjustment operations are performed a plurality of times, and when stable, the characteristics of FIG. 3C are stored in the RAM 111 in FIG.

Next, based on the illuminance characteristic of FIG.
The driving condition of the Z stage 14 is calculated. In this embodiment, in a period from time t ₀ ~t _fd in FIG. 3 (C), the and between time t _fu ~t _5, the movement of the Z stage 14 to a low speed, during the time t _fd ~t _fu Then, the speed control pattern is set so as to move the Z stage 14 at high speed. FIG. 4 shows an example of the relationship between the velocity change pattern of the Z stage, the Z direction position change pattern, and the illuminance change characteristic. Figure 4
FIG. 4A is a graph of the same illuminance characteristic P (t) as FIG. 3A, FIG. 4B is a graph of speed change pattern V (t), and FIG. 4C is a Z position change pattern Z ( It is a graph of t). In the present embodiment, the Z stage is moved at a low speed v ₁ at a constant speed until the flash down time t _fd, and is moved at a high speed v ₂ during the flash down (t _{fd to} t _fu ).
And from the flash-up time t _{fu, the} speed is low again v
Move at a constant speed with ₃ .

By driving the Z stage 14 with the above speed characteristics, as is clear from the characteristics of FIG. 4, the best focus in the optical axis direction (Z direction) accompanying the movement of the Z stage 14 in progressive focus exposure. The image distribution (probability of existence) is calculated between the positions Z _{0 to} Z _{1 in} the Z direction and the position Z.
_The maximum value is taken between _{4 and} Z ₅ (Fig. 5 (B)).
Solid line inside). Further, in the present embodiment, the lamp is provided only in the section in which the existence probability is small according to the existence probability of the best imaging plane (for example, between the positions Zb and Zd on the solid line in FIG. 5B). Since the illumination light intensity is attenuated by setting the power of 1 to the rated power Pm, the light amount distribution in the resist layer, that is, the central position of the hole pattern, or the exposure amount (photosensitive amount) in the vicinity thereof in the optical axis direction. The distribution of is almost uniform, and the effect of expanding the depth of focus is sufficiently obtained. In this embodiment, the velocity characteristic V (t), the position characteristic Z (t), and the illuminance characteristic P (t) at the time of opening / closing the shutter are strictly managed in the present embodiment. The two maximum values can be made substantially equal to each other in the solid line (probability of existence) shown, and the effect of expanding the depth of focus can be maximized.

By the way, the Z position of the Z stage during exposure can be detected by the AF unit 18 shown in FIG. 1, but if the movement range of the Z stage during exposure is large, A
It may exceed the detection range of the F sensor. in this case,
Since the focus signal cannot be obtained in the vicinity of the positions Z ₁ and Z ₄ shown in FIG. 4C, the position information may be obtained from a potentiometer or a position sensor that monitors the drive amount of the Z stage. In this connection
For example, while monitoring the height position of the surface of the wafer W by the AF sensor, the time at which the focus signal at the best focus position ZC indicates the focus (for example, the zero point) is checked, and these times are times t ₂ and t _3. It is also possible to roughly know whether or not the exposure operation for one shot has been successfully performed by determining whether or not the exposure operation is performed at approximately the midpoint.

As described above, in the present embodiment, the electric power supplied to the lamp 1 is switched in two stages (Ph, Pn). Generally, mercury discharge lamps have rated power (voltage, current)
Although it is desirable to turn on the light as described above, it is possible to reduce the power to the rated power or less for a short time. Therefore, by making it possible to switch the power supply to the lamp 1 in three or more steps, it is possible to more accurately make the light amount distribution in the resist layer uniform. Further, in the above-described embodiment, the illuminance is reduced in the section where the existence probability is low.
Needless to say, in the case where the movement width of 4 during exposure is large and the light amount distribution in the resist layer may be concave, the illuminance is increased in the above section.

Although the mercury lamp is used as the light source in the above embodiment, laser light (excimer laser or the like) may be used. To adjust the illumination light intensity, for example, the applied voltage (or charging voltage) to the light source,
Alternatively, the laser output can be finely adjusted by changing the number of repetitive pulses. Also,
Although the Z stage 14 is used in the above embodiment as a means for continuously moving the best imaging plane and the wafer in the optical axis direction during exposure, instead of moving the reticle R in the optical axis direction, or by using a projection lens. The wavelength of the exposure illumination light may be shifted using the axial chromatic aberration of PL to move the best image formation plane relative to the wafer W in the optical axis direction. Further, in the above embodiment, the case where the hole pattern is exposed has been described, but the present invention is applied to a pattern such as a line and space pattern by using a pattern other than this, for example, a contrast enhanced layer (CEL). can do.

[0043]

As described above, according to the present invention, the best image plane of the projection optical system and the photosensitive substrate are preliminarily arranged in the optical axis direction during one exposure time (one shutter opening operation). Since the movement is relatively performed with the controlled speed characteristic, the processing capacity (throughput) of the photosensitive substrate per unit time is not significantly reduced. Moreover, the intensity of light (exposure amount) in the thickness direction on the photosensitive substrate (resist layer on the semiconductor wafer or glass plate) is always made almost uniform by adjusting the intensity of the illumination light to the mask. It is possible to obtain a sufficient effect of expanding the depth of focus in the exposure.

[Brief description of drawings]

FIG. 1 is a diagram showing an example of the configuration of a projection exposure apparatus according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a specific configuration of a main control system in FIG.

FIG. 3 is a diagram for explaining a basic operation of the projection exposure apparatus according to the embodiment of the present invention.

FIG. 4 is a graph showing an example of an illuminance change characteristic due to opening and closing of a shutter and a drive characteristic of a Z stage.

FIG. 5 is a diagram for explaining the manner of exposure by a progressive focus exposure method.

FIG. 6 is a diagram showing the existence probability when performing exposure while moving the Z stage at a constant speed.

FIG. 7 is a diagram illustrating a state of exposure by a conventional multiple exposure method.

[Explanation of symbols]

 1 Mercury Discharge Lamp 3 Shutter 14 Z Stage 15 Drive Unit (Motor, Tacho Generator) 18 AF Unit 23 Lamp Control Unit 100 Main Control System R Reticle PL Projection Lens W Wafer

 ─────────────────────────────────────────────────── ─── Continued Front Page (72) Inventor Naomasa Shiraishi 1-6-3 Nishioi, Shinagawa-ku, Tokyo Inside Nikon Oi Manufacturing Co., Ltd.

Claims (1)

[Claims] 1. A light source for generating illumination light for illuminating a mask on which a predetermined pattern is formed, a shutter for switching between irradiation and blocking of the illumination light on the mask, and projection optics for projecting a pattern image of the mask. System, a movable stage that holds the photosensitive substrate in the vicinity of the best image plane of the projection optical system, and is movable in the optical axis direction while the shutter is open, and illumination to the mask. In a projection exposure apparatus having a photoelectric detector that detects a value corresponding to the amount of light, and shutter control means that controls the opening and closing of the shutter based on the value detected by the photoelectric detector, the opening of the shutter is started. The existence time per unit position in the optical axis direction of the best image plane of the projection optical system from the time point to the closing time point is substantially equal at at least two points in the optical axis direction. Stage control means for controlling the movable stage so as to have a maximum value; and a stage control means for controlling the opening of the shutter and the closing time of the shutter according to the existence time per unit position in the optical axis direction of the best imaging plane. An illumination light intensity adjusting means for attenuating or amplifying the illumination light for illuminating the mask for a predetermined time between.