JP2853711B2 - Exposure method and semiconductor element 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 control of irradiation energy amount to an appropriate sensitive object and control of uniformity of illuminance. In particular, pulse energy such as excimer laser light or X-ray is used as exposure light. The present invention relates to an exposure method using an exposure apparatus and a method for manufacturing a semiconductor device using such an exposure apparatus.

[0002]

2. Description of the Related Art Conventionally, exposure amount control in an exposure apparatus using a pulse laser as a light source generally involves a variation of about ± 10% for each pulse, and a short-term and long-term variation in laser density. Because of the phenomenon of reduction, the light amount of each pulse is detected and integrated, and light emission is performed until the integrated result reaches a desired value. The phenomenon of lowering the laser density is remarkable when a gas laser is used, and the output decreases as the internal gas deteriorates.

[0003] Usually, when the internal gas is deteriorated and the output is reduced, the discharge applied voltage or the like is increased, or in some cases, the output is reduced by performing partial gas exchange to reduce the reduction in output. However, it has been difficult to control the exposure amount to a desired value with a fixed number of pulses. Under these circumstances,
For example, Japanese Patent Application Laid-Open No. Sho 60-169136 discloses a control device in which higher-precision exposure amount control is required, such as exposure control in an exposure device for manufacturing semiconductor devices.
Is disclosed in Japanese Unexamined Patent Application Publication No. 2000-205,878.

In this apparatus, exposure energy to be applied to a sensitive body (a wafer with a resist or the like) is divided into a coarse exposure which gives an exposure energy slightly smaller than an appropriate exposure amount and a correction exposure which gives the remaining required exposure energy. By dividing into stages, variations in exposure energy are controlled as a whole. That is, when performing one-shot exposure with a plurality of pulses, the optimum exposure amount is obtained by controlling the exposure amount by the final pulse having a reduced energy amount.

[0005] Here, one shot means that the whole wafer is exposed to exposure energy through a mask in the case of the batch exposure method, and in the case of the step-and-repeat method (which will be described later). That is, the exposure energy is applied to one of the target areas. By the way, in the step-and-repeat method recently used in an exposure apparatus for manufacturing a semiconductor element, when exposing one wafer, an area to be exposed is divided into a plurality of exposure areas, and exposure is performed for each exposure area. When the exposure of one exposure area is completed, the operation of moving to the next exposure area and performing exposure again is repeated, thereby finally exposing the entire surface of the wafer.

Therefore, in order to increase the production of semiconductor devices per unit time, it is important to shorten the movement time between each exposure area as much as possible and to shorten the exposure time in each exposure area. For this reason, JP-A-60-16913
Even when exposure is performed by the means disclosed in Japanese Patent Application Laid-Open No. 6-64, it is necessary to perform rough exposure with as much energy as possible and to shorten the time of correction exposure.

[0007]

However, in the method disclosed in the above-mentioned Japanese Patent Application Laid-Open No. 60-169136, no consideration is given to an error (variation) in the amount of energy contained in the minimum pulse. Therefore, there is a disadvantage that the exposure amount is still not appropriately controlled, and that appropriate exposure cannot be performed.

Further, since the set energy of the final pulse is a value obtained by subtracting the coarse exposure from the proper exposure, the dynamic range of the means for setting the exposure of the final pulse must be widened, and the apparatus becomes complicated. In addition, there is a problem that high control accuracy cannot be obtained. In addition, JP
As disclosed in Japanese Patent Application Laid-Open No. 3-81882, there is a method of obtaining an appropriate exposure amount by using a plurality of correction exposure pulses. However, in this case, the exposure time becomes longer and the dynamic range of the exposure amount setting means becomes larger. The problem remains.

Further, in the above-described method of adjusting the exposure amount by the correction exposure, since the energy amount of one pulse and the number of exposure pulses at the time of the correction exposure depend on the integrated exposure amount at the time when the rough exposure is completed. However, there is a disadvantage that it is not constant for each shot. That is, when the exposure energy source is a laser light source, unevenness of illuminance called speckle may occur on the exposed surface due to the coherence of the laser light, and the above light amount control method effectively reduces this speckle. Can not do. Therefore, this problem will be described below.

Since speckle has a significant effect on the control of the pattern line width in the photolithography process of manufacturing a semiconductor device, for example, Japanese Patent Application Laid-Open No. Sho 59-2263.
As disclosed in Japanese Patent Application Publication No. 17, it is conceivable that the speckle pattern (or interference pattern) is moved (oscillating the laser light) every time the pulsed light is emitted to finally smooth the speckle. I have.

However, the method disclosed in JP-A-59-226317 and the method disclosed in JP-A-63-81882 are combined.
In the case of reducing speckle while optimizing the amount of exposure, a light source image (laser spot) on the pupil plane of the projection lens as disclosed in Japanese Patent Application Laid-Open No. 59-226317.
Must be distributed as uniformly as possible in the pupil plane.
To achieve this, it is necessary to keep the number of laser spots on the pupil plane constant, perform exposure with an integral multiple of the number of pulses, and keep the laser intensity constant during exposure.

However, Japanese Patent Application Laid-Open No. 63-81882 described above.
According to the method of (1), since the exposure intensity is changed at the time of the correction exposure, only the area corresponding to the correction exposure at the pupil plane has a lower integrated energy than other areas, and as a result, speckle cannot be sufficiently reduced. In view of the above, the present invention can control the exposure amount according to the required accuracy, and can evenly uniform the illuminance as needed, and shorten the exposure time. It is an object of the present invention to provide an exposure method that can be used, and a method for manufacturing a semiconductor device using the exposure method.

[0013]

Therefore, in order to achieve the above object, the invention according to claim 1 is a method for forming a pattern of a first object (reticle R or mask) on a photosensitive second object (wafer W). A projection optical system (projection lens PL) for projecting onto a predetermined area (shot) of the light source and an irradiation system (expander 12) for irradiating a plurality of pulse energies oscillated from the pulse light source (10) toward the first object (R). , High-speed darkening unit 1
3, a pattern is formed on the second object (W) by using an exposure apparatus (a reduced projection exposure apparatus such as a stepper) including a speckle reduction unit 20, a mirror 21, an integrator 28, a beam splitter 29, and a condenser lens 30). Applied to the method of exposure.

According to the first aspect of the present invention, while projecting the pattern of the first object (R) onto the second object (W), at a time interval corresponding to a predetermined oscillation frequency (for example, 100 to 500 Hz). Pulse energy is repeatedly emitted from the pulse light source (10), and a value (Pa) corresponding to the amount of exposure on the second object (W) is detected for each emitted pulse energy. The target exposure amount set for each energy (for example, the average light amount value P per one pulse set in step 101 in FIG. 9)
The intensity of the pulse energy is adjusted at high speed within the time interval of repetitive oscillation based on the integrated exposure amount given by the integration of the pulse light or the average light amount value P per unit pulse (for example, by the high-speed dimming unit 13). (Adjustment of transmittance).

According to a second aspect of the present invention, the number of pulses of a plurality of pulse energies to be applied to a predetermined area on the second object, the proper exposure amount (S) to be applied to the second object and the pulse light source defined in the first aspect. The determination is made so as to obtain a predetermined exposure amount control accuracy (A) in consideration of the pulse energy intensity variation (for example, ΔP) from (10).
The invention according to claim 3 is the pulse light source (1) according to claim 2.
When an interference pattern is generated on the first object (R) or the second object (W) by the irradiation of the pulse energy from 0), the number of pulses determined so as to obtain the predetermined exposure amount control accuracy (A) is determined. , The minimum number of pulses required to reduce the visibility of the interference pattern to a predetermined accuracy (Nmin
) Is to set above.

According to a fourth aspect of the present invention, the pulse light source (10) according to any one of the first to third aspects includes a light source (such as an excimer laser) for emitting a laser light in a deep UV region including a band narrowing wavelength stabilizing mechanism. ) Or a light source that emits pulse energy such as X-rays. According to a fifth aspect of the present invention, in any one of the first to fourth aspects, the pulse energy is adjusted by a dimming member (two sheets) capable of continuously changing the transmittance of the pulse energy from the pulse light source (10). Light / dark gratings 16a, 16b; rotatable polarizing plates;
(Etalon by two quartz glass plates 40; or a transmission type double phase grating or the like).

Further, according to a sixth aspect of the present invention, a plurality of pulse energies from a pulse light source (10) are irradiated on a first object (a reticle R or a mask) and a pattern formed on the first object (R) is changed. A semiconductor element is formed on the second object (W) using an exposure apparatus (a reduction projection exposure apparatus such as a stepper) that exposes a predetermined region (shot) on a second object (wafer W) for manufacturing a semiconductor element. Applied to manufacturing methods.

According to the sixth aspect of the present invention, an intensity variation (for example, ΔP) between respective pulse energies from the pulse light source (10), an appropriate exposure amount (S) to be given to a predetermined area of the second object (W), and The number of exposure pulses (N) determined in consideration of the control accuracy (A) for the appropriate exposure amount (S)
exp) based on the average intensity of each pulse energy (P)
And irradiating the first object (R) with pulse energy of the average intensity (P) determined for exposure to a predetermined area on the second object (W); Is the exposure amount actually given by the irradiation of the pulse energy (the value Pa corresponding to the light amount detected by the light receiving element 24) and the target exposure amount set in advance for each pulse energy (for example, in step 101 in FIG. 9). The pulse energy to be irradiated next is adjusted based on the integrated exposure amount given by integrating the set average light amount value P for each pulse light or the average light amount value P per unit pulse (for example, a high-speed light reduction unit). 13) is repeated during the number of exposure pulses (Nexp).

According to a seventh aspect of the present invention, the target exposure amount defined in the sixth aspect is set as an exposure amount integrated in a predetermined area on the second object (W) with an increase in the number of exposure pulses (Nexp). The present invention of claim 8 is to set the target exposure amount defined in claim 6 as an exposure amount for each pulse energy of the number of exposure pulses (Nexp). The ninth aspect of the present invention provides the pulse light source (1) according to the seventh or eighth aspect.
0) is a light source (such as an excimer laser) that emits a pulse laser beam in the deep UV region including a band narrowing wavelength stabilizing mechanism at a repetition frequency of 100 Hz or more, or a light source that emits pulse energy such as X-rays. According to a tenth aspect of the present invention, in the ninth aspect, the first object (R) or the second object (R) is irradiated by irradiation of pulse energy from the pulse light source (10).
When an interference pattern is generated on the object (W), the number of exposure pulses (Nexp) is set to be equal to or more than the minimum number of pulses (Nmin) necessary to reduce the visibility of the interference pattern to a predetermined accuracy.

According to an eleventh aspect of the present invention, there is provided a projection optical system in which the exposure apparatus defined in the tenth aspect projects an image of a pattern formed on the first object (R) onto a predetermined area on the second object (W). (Projection lens PL) and an integrator (fly-eye lens 2) for making pulse energy from the pulse light source (10) into a substantially uniform illuminance distribution on the first object (R).
8) Irradiation system including (expander 12, high-speed darkening unit 1)
3, a speckle reduction section 20, a mirror 21, a beam splitter 29, and a condenser lens 30), and a pulse emitted from an integrator (28) to provide an exposure amount (Pa) actually given by irradiation of pulse energy during exposure. This defines that the energy is detected by the light receiving element (24) that receives the energy.

According to a twelfth aspect of the present invention, the adjustment of the pulse energy according to the eleventh aspect is performed in such a way that the transmittance of the pulse energy from the pulse light source (10) can be continuously changed by being arranged before the integrator (28). It is specified that the process is performed by a member (two light / dark gratings 16a and 16b; a rotatable polarizing plate; an etalon formed by two quartz glass plates 40; or a transmission type double phase grating).

According to a thirteenth aspect of the present invention, a plurality of pulse energies from a pulse light source (10) are irradiated on a first object (a reticle R or a mask) to change a pattern formed on the first object (R). Applied to a manufacturing method for forming a semiconductor element on a second object (W) using an exposure apparatus (a reduction projection exposure apparatus such as a stepper) for exposing a predetermined area on a second object (wafer W) for manufacturing a semiconductor element. Is done.

According to the thirteenth aspect, a variation (for example, ΔP) of each pulse energy output from the pulse light source (10) and an appropriate exposure amount (S) to be given to a predetermined area of the second object (W) are considered. To determine the number of exposure pulses (Nexp) required for exposure, and the number of exposure pulses (Nexp)
Exp), a target exposure amount to be given to a predetermined area of the second object (W) (for example, a diagram given by integrating pulse light of an average light amount value P per pulse set in step 101 in FIG. 9). 10) and setting a predetermined area of the second object (W) to the number of exposure pulses (N
exp)), the pulse energy actually output from the pulse light source (10) is actually measured (light receiving element 24).
The first object (R) is detected based on the actually measured energy and a target exposure amount (eg, a target integrated exposure amount as shown in FIG. 10) for each preset exposure pulse number (Nexp). Of adjusting the pulse energy applied to the laser beam (for example, step 104 in FIG. 9)
To 109).

According to a fourteenth aspect of the present invention, the target exposure amount according to the thirteenth aspect is set as an exposure amount integrated in a predetermined area on the second object (W) with an increase in the number of exposure pulses (Nexp). According to a fifteenth aspect of the present invention, the target exposure amount in the thirteenth aspect is defined as the number of exposure pulses (Nex
p) is set as an exposure amount for each pulse energy, and the invention according to claim 16 is directed to the pulse light source (10) according to claim 14 or 15, wherein the pulse light source (10) includes a deep band including a band narrowing wavelength stabilizing mechanism. The light source (e.g., an excimer laser) that emits pulsed laser light in the UV region or a light source that emits pulsed energy such as X-rays is defined.

[0025]

DESCRIPTION OF THE PREFERRED EMBODIMENTS In the present invention, since the correction exposure is not performed by reducing the light quantity of the last several pulses or one pulse as in the prior art, the exposure time for each shot becomes shortest and constant. obtain. Further, in the present invention, the exposure energy is controlled to be a predetermined average light amount value over all pulses to be exposed, and the number of pulses for each shot is made substantially constant, so that the speckle pattern is By moving the speckle pattern in synchronization with light irradiation, the speckle pattern can be finally almost completely smoothed.

Further, in the present invention, the light quantity is actually measured for each pulse, and the pulse energy is adjusted based on the target exposure set in advance for each pulse. Compared to the case where the pulse energy is adjusted at the stage where the exposure is fairly close to the appropriate exposure, the dynamic range of the measuring means for actually measuring the pulse energy and the adjusting means for adjusting the next pulse energy can be reduced, and the exposure can be controlled. Accuracy can be improved. The configuration and operation of an exposure apparatus suitable for carrying out the method according to the present invention will be described below with reference to the drawings.

FIG. 1 shows a configuration of an embodiment of the present invention. Here, a reduction projection exposure apparatus (stepper) for semiconductor production for projecting and exposing a pattern of a reticle R (first object) onto a wafer W (second object). The configuration applied to the above is shown. In FIG. 1, a trigger control unit 9 that outputs an external trigger pulse controls the oscillation of a pulse laser light source 10 that emits pulse light such as excimer laser light.

The pulse laser light source 10 has a band narrowing wavelength stabilizing mechanism composed of an etalon, a dispersion element, and the like at a part between two resonance mirrors disposed at both ends of the laser tube. And a laser light source having a stable resonator. The emission beam LB0 from the pulse laser light source 10 is, for example, a deep UV region having a wavelength of 248 nm, and its beam cross section is a rectangle having an aspect ratio of about 1/2 to 1/5. The laser beam enters a beam expander 12 combined with a cylindrical lens, and the expander 12 shapes the beam cross section from a rectangle to a substantially square, and emits the beam as a beam LB1.

The high-speed darkening section 13 attenuates the beam light amount (energy) of the incident beam LB1 in, for example, six discrete steps or continuously. The high-speed dimming unit 13 is configured to switch the dimming rate at a high speed while exposing the pattern of the reticle R to the wafer W via the projection lens PL. As an example, a structure in which two light-dark gratings having a transparent portion and an opaque portion with respect to a laser beam are arranged at a predetermined interval so as to be relatively movable.

FIG. 2 is a block diagram showing an example of the high-speed dimming section 13. The high-speed dimming section 13 includes two light-dark gratings 16a, 1
6b and a grid driving section 16c. Bright and dark grid 16
a and 16b are formed by depositing chromium or the like on a glass plate made of a material having a high transmittance to ultraviolet light such as quartz glass and forming opaque portions to ultraviolet light at appropriate intervals on slits. Made, two grids 16a, 16b
Are arranged in parallel at appropriate intervals. One of the gratings 16a is fixed with respect to the laser beam LB1, and the other grating 16b is coupled with a grating driving unit 16c so that the grating 16a moves slightly in the pitch direction of the grating by a command from the light amount control unit 14. It has become.

FIG. 3 shows a partially enlarged view of the light and dark grid and its operation. Let GP be the pitch of the grating and GW be the width of the opaque part (shielding part). FIG. 3 (a) shows a case where the two gratings are not displaced at all, and the transparent portion and the shielding portion coincide with each other, and the extinction ratio for the incident light LB1 is Gw / Gp. FIG.
(B) is a case where the light-dark grating 16a (fixed) and the light-dark grating 16b (movable) are shifted by Δd, and the extinction ratio with respect to the incident light LB1 is (Gw + Δd) / Gp. Where Δd is
Gw> Δd.

For example, when Gp is 10 μm and Gw is 2.5 μm, the output beam LB1 can be reduced from 75% to 50% with respect to the incident beam LB1 by this configuration. In the case of this example, since the maximum movement amount of Δd is defined by Gw, the maximum movement amount may be 2.5 μm. Therefore, the grid driving unit 16
c can sufficiently function even when a piezo element or the like is used, and can be controlled by completely following each pulse emission with respect to a repetition frequency of 100 Hz to 500 Hz in the case of a general excimer laser or the like. , High-speed dimming can be realized.

Returning to the description of FIG. 1, the substantially parallel beam LB1 which has undergone a predetermined attenuation passes through a speckle reduction unit 20 in order to smooth a speckle pattern including interference fringes, so that the beam LB1 has a small angle. After becoming a vibrating beam that swings in one dimension (or two dimensions), it is turned back by the mirror 21 and a fly-eye lens 28 as an optical integrator
Incident on.

The fly-eye lens 28 is formed by bundling a plurality of rod-shaped element lenses.
Each condensed spot of the partial light beam B1 is formed.
Therefore, the beam LB1 incident on the fly-eye lens 28
The angle of incidence on the incident surface of the fly-eye lens 28 changes every moment.

FIG. 4 is a schematic explanatory view showing the relationship between the incident beam of the fly-eye lens 28 and the two-dimensional light source image (spot light) according to the principle disclosed in Japanese Patent Application Laid-Open No. 59-226317. Each of the rod lenses 28a of the fly-eye lens 28 is a quadrangular prism made of quartz with convex spherical surfaces formed at both ends. When the beam LBb (parallel light beam) enters the fly-eye lens 28 in parallel to the optical axis AX, the fly-eye lens 2
The spot light SPb is condensed at the exit end of each of the rod lenses 28a of FIG.

Although this spot light SPb is shown for only one rod lens in FIG. 4, it is formed on all the exit sides of the rod lens irradiated with the beam LBb, and each spot light SPb is The light is focused on almost the center of the lens exit surface. Further, a parallel beam LBc inclined rightward with respect to the optical axis AX is formed by the fly-eye lens 2.
8, the light is condensed as spot light SPc on the left side of the exit surface of each rod lens 28a.

Therefore, due to the one-dimensional vibration of the beam in the speckle reduction unit 20, all of the plurality of spot lights generated on the exit side of the fly-eye lens 28 are one-shot with respect to the fly-eye lens 28 (optical axis AX). In the same direction. In this way, a plurality of beams forming each spot light formed on the exit side of the fly-eye lens 28 are mostly transmitted by the beam splitter 29 as shown in FIG. 1, enter the condenser lens 30, and enter the reticle R on the reticle R. Each is superimposed. Thus, reticle R
Illumination is performed with a substantially uniform illuminance distribution, and the pattern of the reticle R is transferred to the resist layer of the wafer W mounted on the stage by the projection lens PL at a predetermined exposure amount.

That is, the speckle reduction section 20 vibrates the beam incident on the fly-eye lens 28,
The interference fringes (one-dimensional speckle pattern) generated on the reticle surface or wafer surface are moved by a very small amount, and when exposure is completed, the resulting light and dark fringes transferred to the resist layer are smoothed, thereby improving the visibility of the interference fringes. Reduce. (However,
Exposure unevenness occurs due to interference of light from each spot light. ) Image-centric (wafer) side telecentric projection lens P
On the L pupil (entrance pupil) ep, a plurality of spot lights formed at the exit end of the fly-eye lens 28 are re-imaged, thereby forming a so-called Koehler illumination system.

In this embodiment, when smoothing the speckle pattern, the laser light incident on the fly-eye lens 28 is vibrated. In addition to this, for example, a rotary diffuser is synchronized with the emission of pulsed light. It may be configured to rotate. Next, a part of the beam split by the beam splitter 29 is focused on the light receiving element 24 by the focusing optical system 23. The light receiving element 24 outputs the beam LB0 (or LB1).
), A PI having sufficient sensitivity in the ultraviolet region so as to accurately output a photoelectric signal corresponding to the light quantity of each pulse emission
It is composed of N photodiodes and the like. The photoelectric signal is input to the light amount monitor 26, where the light amount for each pulse emission is sequentially integrated.

The light receiving element 24 and the light amount monitor 26 constitute light amount measuring means, and the measured actual value (the light amount may be a value corresponding to the actual light amount, and need not be the light amount itself; the same applies hereinafter). , To the main control system 8. This main control system 8
Sends a command signal to the trigger control unit 9, the light amount control unit 14, and the speckle control unit 22 to control the operation of the entire apparatus. The input / output device 7 provided in the main control system 8 is a man-machine interface between the operator and the device main body.
Various parameters required for exposure are received from the operator, and the state of the apparatus is notified to the operator.

The memory 6 stores parameters (constants), tables, and the like necessary for the exposure operation and various calculations input from the input / output device 7. In the present embodiment, information for determining the minimum number of light emission pulses required for satisfactorily smoothing the speckle pattern while the beam vibrates for a half cycle is stored in the memory 6. Here, the half cycle of the beam means that the spot light is SPa → SP in FIG.
In order to move in the order of b → SPc (or reverse), it corresponds to tilting the beam by the swing angle α ° in the order of LBa → LBb → LBc (or reverse).

In the main control system 8, the number of pulses necessary for smoothing the speckle pattern stored in the memory 6 in advance and the data on the proper exposure amount of the resist applied to the wafer W are calculated based on the data. An average light amount value of each pulse light is determined by the first calculating means (not shown).
The main control system 8 further calculates a target value corresponding to a target integrated light amount to be given to the wafer W when each pulse is irradiated with the average light amount value by a second calculating means (not shown), and a third calculating means. (Not shown), the difference between the target value and the actually measured value sent from the light amount monitor 26 described above is calculated. Then, based on the difference, a command for adjusting the light amount in the high-speed darkening unit 13 is output to the light amount control unit 14.

Further, the main control system 8 outputs a drive signal to the speckle control unit 22 so that the pulse emission of the laser light source 10 and the deflection angle of the beam by the speckle reduction unit 20 are synchronized. In this synchronization, an oscillation start and stop signal may be output to the trigger controller 9 following the output of a detector that monitors the beam deflection angle with high accuracy.

By the way, in the above description, the high-speed dimming unit 1
3 is composed of two light-dark gratings, but the high-speed darkening section 13 is not limited to this. For example, when the laser beam LB0 (LB1) is linearly polarized, a polarizing plate is used. May be rotated. In this case, the transmittance of the beam can be continuously varied depending on the rotation position of the polarizing plate.

Further, an etalon used for narrowing the wavelength band of the laser may be used as the high-speed dimming unit 13. Generally, the etalon is composed of two glass plates (in this case, quartz glass is used to increase the transmittance with respect to ultraviolet light).
Are held in parallel at regular intervals, and have a high transmittance for a specific wavelength according to the intervals.

FIG. 5 shows the structure of the high-speed light reduction section when this etalon is used. The high-speed dimming unit in FIG. 5 includes two quartz glass plates 40, a spacer 41 for holding the two quartz glass plates 40 in parallel at a predetermined interval, and a holding frame 45 for holding the etalon. , An actuator 42 for finely rotating the etalon about a fulcrum 44 and a spring 43 for attracting the holding frame to the fixing portion 46.

In the etalon having such a configuration, assuming that the interval between the two glass plates 40 is d and the inclination angle with respect to a plane perpendicular to the optical axis is θ, the wavelength satisfying the relationship 2d cos θ = mλ (1) The highest transmittance is obtained for λ.
Here, m is an integer value, and as an example, θ = 0 ° and d = 5
Under the conditions of 9.984 μm and λ = 248.380 nm,
It has a maximum transmittance at a value of m = 483. Conversely, under the condition that m is constant, the wavelength λ at which the maximum transmittance is obtained can be changed by changing the angle θ.

FIG. 6 shows a typical example of the transmittance with respect to the wavelength of the etalon. As shown in FIG. 6, the transmittance characteristic of the etalon with respect to the wavelength is a so-called "comb-shaped" Band-Pas.
The filter has an s-filter characteristic, and there are high transmittance portions at substantially constant intervals depending on the order of m. The interval R between the transmittance peaks is generally called a free spectral range, and when the relationship of the formula (1) holds, R = (2d cos θ / m) − (2 d cos θ / (m + 1)) ( 2) The Q value of the filter in each band uses an expression called finesse, and is expressed by R / ΔB. ΔB indicates the half width of one peak of the filter as shown in FIG.

When the etalon described above is used as the high-speed darkening section 13, it is convenient to use an etalon having a bandwidth slightly wider than the wavelength bandwidth of the laser light. The wavelength bandwidth of a laser used in an ordinary exposure apparatus is 5/100.
Since it is about 0 to 3/1000 nm, 10/10000 n
It is preferable to use an etalon having a band of not less than m. FIG. 7 shows the principle in the case where the laser beam is dimmed using an etalon. FIG. 7A shows the intensity distribution with respect to the wavelength of the laser emitted from the pulse laser light source 10, and its bandwidth is about 3/1000 nm to 5/1000 nm. The center wavelength is stabilized inside or outside the pulse laser light source 10, and the stability is controlled to 1/1000 nm or less.

On the other hand, the characteristics of the etalon used as the high-speed darkening portion 13 are shown in FIG. 7B, and have a comb-shaped filter characteristic. The etalon is actually used as the high-speed darkening unit 13
, The center wavelength λm of the filter is
The bandwidth ΔB of the filter is controlled between λo ± ΔB (actually, control may be performed between λo + ΔB and λo or control may be performed between λo−ΔB and λo). Control of the center wavelength λm of the filter can be realized by changing the tilt angle of the etalon as described above.

The laser intensity after passing through the high-speed darkening portion is determined by the product of the profile shown in FIG. 7A and the profile shown in FIG. 7B, and the laser having the power shown in FIG. Is done. Here, if λo and λm are completely matched, the maximum transmittance can be obtained, and in practice, the transmittance is 90% or more. On the other hand, if .lambda.m is set to .lambda.m = .lambda.o -.DELTA.B, the transmittance is about 10%. (Λ
The same effect can be obtained when m = λo + ΔB. )or,
If .lambda.m is set to a value slightly smaller than the value of .lambda.m = .lambda.o -.DELTA.B, a transmittance at a level close to 0% can be obtained.

In actual control, the transmittance of the light with respect to the angle of the etalon is measured in advance, this data is stored in the memory 6, and the amount of angle control of the etalon with respect to the desired transmittance is obtained. What is necessary is just to give a drive command to 42. As an example, using an etalon in which the distance between the glass plates is 60 μm, λo = 24
When dimming control is performed on the 8.380 nm laser, λo and λm match at θ = 6.5262 degrees. At this time, the order of m is 480. Next degree m = 48
Since the center frequency at 1 is λ = 247.864 nm, the free spectral range of this etalon is 0.51
6 nm. If the finesse is 10, the half width of the band is about 0.05 nm.

Therefore, if the center wavelength is shifted by 0.05 nm, the transmittance becomes about 10%, and the angle at this time becomes 6.4246 degrees. Thus, a tilt of the etalon of about 0.1 ° allows control of the transmittance from 90% to 10%. When driving at a position with a radius of 50 nm with respect to the rotation center 44 of the etalon, a tilt of 0.1 ° can be controlled by moving about 90 μm at the driving point. Since control is possible at this level, it is possible to drive at a sufficiently high speed by using a piezo element or the like as an actuator.

The high-speed dimming unit 13 can be inserted not only at the position shown in FIG. 1 but also between the pulse laser light source 10 and the expander 12 or between the resonator mirrors of the pulse laser light source 10. Similar effects can be obtained. Further, when the speckle reduction unit does not employ the above-described method of causing the beam to vibrate at a small angle, the beam may be inserted between the speckle reduction unit 20 and the fly-eye lens 28. However, in any case, the high-speed dimming section 13 needs to be placed in a stage before the laser light is incident on the fly-eye lens 28.

The reason is that the transmittance of the etalon is higher than that of Ga.
Since it depends on the p-value, it is preferable to keep the Gap value as uniform as possible with respect to the translucent portion of the etalon, but it is practically impossible to maintain the same transmittance over the entire surface due to manufacturing errors. . Therefore, in order to eliminate the deterioration of the illuminance uniformity (unevenness of the intensity distribution in the beam cross section) due to the etalon, it is desirable to insert this etalon in front of the fly-eye lens. The features of using an etalon include a large controllable extinction ratio, a high maximum transmittance, and uneven illumination and interference (from the edge of the grid) as compared with the method using the light-dark grid shown in FIG. Is less likely to occur).

As still another example of the high-speed dimming unit 13,
There is a transmission type double phase grating as shown in FIG. In such a phase grating, two diffraction gratings having a predetermined step at a predetermined pitch are held in parallel at a predetermined interval. The step d of the diffraction grating is such that d = λ / 2 (n−1) (3) where n is the refractive index of a member of the grating (in this case, a member transparent to ultraviolet light is used). It is made. Here, λ is the wavelength of light transmitted through this phase grating.

By arranging the two phase gratings with a certain amount of shift as shown in FIG. 8, the paths through which light passes (hereinafter referred to as paths) are A (concave-convex), B (convex-convex), and C (convex-convex). Convex
Concave) and D (concave-concave). A in this
And C have exactly the same optical path length, but the path of B has an optical path difference of d (n-1) as compared with the path of A (however, the refractive index of air is set to 1).

Therefore, from equation (3), the optical path difference between A and B is λ /
The light in the portion A and the light in the portion B interfere with each other and cancel each other, and as a result, the light transmittance of the two phase gratings decreases. The degree of this dimming depends on the amount of displacement (Δx) between the two phase gratings. B and C, C as above
And D, and the light passing through the portions D and A interfere with each other and are dimmed. Also in the case of this method, the pitch of the unevenness of the diffraction grating may be about 10 μm. Therefore, the actuator for relatively moving the two diffraction gratings can be driven at a sufficiently high speed by using the piezo element.

Next, the operation of the embodiment will be described with reference to the flowchart of FIG. After the start of the operation, first, in step 101, the pulse number Nmin necessary for reducing speckles previously stored in the memory 6, the optimum (target) total exposure amount S found in the resist applied to the wafer, and the exposure An average light amount per pulse is calculated by the first calculation means provided in the main control system 8 based on each data of the allowable accuracy of the amount control.

Here, the number of pulses Nmin required to reduce the speckle is a value determined by how much the visibility of the speckle (including one-dimensional interference fringes) is reduced. However, if the value is too large, the exposure time becomes longer and the throughput of the exposure apparatus is reduced. Therefore, the value is set in consideration of these factors.

On the other hand, the average exposure energy per unit pulse (under the initial dimming value set by the high-speed dimming unit 13 before exposure) is set to P, and the total energy for one exposure area (target appropriate (Exposure amount) is S, the number of exposure pulses Nexp for one exposure area is Nexp = iNT (S
/ P). Here, iNT (α) is 0.5 to the real value α.
And rounding off to convert to an integer value.

The number of exposure pulses Nexp must be an integral multiple of Nmin to reduce speckle. Therefore, Nexp = (S / Pβ) = mNmin (4) Here, m is an integer value of 1 or more, β is a coefficient of 1 or less, and 0 or more, and is referred to as an average dimming rate.

The variation of the light intensity for each pulse is ΔP
In the case where the light amount is controlled for each pulse, the error in the final integrated exposure amount is the intensity variation ΔE = ΔP · β of the final pulse. You can do it. That is, assuming that the exposure control accuracy is A, (ΔP · β) / S ≦ A (5) β ≦ AS / ΔP From Expression (4), β = S / (P · mNmin). Therefore, the relationship of S / (P · mNmin) ≦ A · S / ΔP holds. Therefore, m ≧ ΔP / (APNmin) (6) Since β is 1 or less, S / (P · mNmin) ≦ 1 ≦ m ≧ S / (P · Nmin) (7) Therefore, (6) , (7), an integer value m that satisfies the expression
The number of exposure pulses and the average extinction rate β (that is, the average light amount Pβ) may be determined.

As an example, the average energy P of one pulse
Is 2 mJ / cm2, the proper exposure is 153 mJ / cm2,
If Nmin is 50 pulses, the number of pulses required for proper exposure control alone is 77 pulses. However, since m · Nmin must be an integer multiple of 50, exposure is performed with m = 2, that is, 100 pulses. I will do it. The average extinction ratio β at this time is 0.765. If the variation of the exposure energy of one pulse is ± 10%, ΔP is ± 0.2 m
J / cm2 and β · ΔP = ± 0.153 mJ / cm2
Becomes

Subsequently, at step 102, the high-speed darkening section 13
Is set to β, and a value S corresponding to the integrated light amount of the pulse counter n and the light amount monitor unit 26 is set in step 103.
Set a to Nexp and zero, respectively. Then, in the next step 104, it is determined whether or not the value of the pulse counter is zero. If not, the process proceeds to step 105, in which a trigger signal is sent from the trigger control unit 9 to the light source 10 to emit one pulse. At the same time, a value Pa corresponding to the actual light amount of the pulse emitted from the light receiving element 24 is detected. In the subsequent step 106, the setting of the integrated light amount in the light amount monitoring section 26 is set to Sa + Pa, and the setting of the pulse counter is set to n-1.

Next, in step 107, the second calculating means calculates the target integrated light quantity to be given by the average light quantity Pβ determined in step 101 according to equation (8), and the third calculating means calculates the target integrated light quantity. And the difference D between the actually measured integrated light amounts is obtained. D = (Nexp−n) · P · β-
Sa (8) Then, based on the difference D, the extinction ratio βn in the next pulse is determined by the equation (9).

Βn = (P · β + D) / P (9) where βn> 1−a / 100 when the variation of the exposure energy of each pulse light is a%, βn> 1−a / 100
= 1−a / 100 (maximum transmittance). Next, in step 109, the light amount control unit 14 sets the dimming rate of the high-speed dimming unit 13 to βn determined in the previous step 108, and returns to step 104. In step 104, it is determined whether the value of the pulse counter is zero as described above. If not, the process proceeds to step 105 to perform the same operation as described above, and returns to step 104 again. If so, the exposure operation ends.

Next, the state of the exposure control in this embodiment will be described with reference to the graph of FIG. 10 showing the relationship between the number of pulses and the integrated exposure. FIG. 10 shows a case where exposure is completed with eight pulses, where the horizontal axis represents the number of pulses and the vertical axis represents the integrated exposure amount. In the figure, the straight line indicated by the two-dot chain line
A target value of the integrated light amount to be given by the pulse light having the average light amount determined in step 01 is shown. In this embodiment, the light amount is controlled for each pulse light so that the exposure is performed according to the target value.

Assuming that the first pulse light is emitted with an exposure amount of P1 as a target and the actually detected exposure amount after emission is P'1, the second pulse light is a target exposure amount 2P1
The light is emitted with the difference between (P2) and P'1 (2P1-P) = P2. Similarly, if the actually measured value of the light quantity of the second shot is P'2, the third shot is 3P1-P '
Light emission is performed with the light amount set to 1−P′2. By repeating this, the exposure is completed at the eighth pulse in a state where the displacement of the two-dot chain line from the target line is small. The final control accuracy (error with respect to the proper exposure amount) is the light amount error of the eighth pulse.

In the above example, when the next pulse light is set, the light amount of the next pulse is set from the difference between the target value of the integrated exposure amount after the emission of the next pulse light and the past integrated light amount. However, if there is some tendency in the variation method for each pulse, the ratio of the target value for each unit pulse and the actual measurement value for each unit pulse is averaged over past multiple pulses, and the target value is the average value of this ratio. A new target value may be set by dividing by.

In the above embodiment, the case where the pulse energy oscillated from the light source is a coherent laser beam has been described. However, the case where the light source of the exposure apparatus emits a non-coherent pulse light, For example, in the case of emitting pulse energy other than light such as X-rays, it is not necessary to consider the reduction of speckle, so the necessary number of pulses is determined based on the pulse energy variation range and the allowable control accuracy. A target value for each pulse may be set from the required number of pulses and the value of the optimum exposure amount. That is, since the error with respect to the final proper exposure amount is determined by the error of the final pulse, the energy amount of one pulse may be set so that the variation of one pulse falls within the allowable error.

[0072]

As described above, according to the present invention, the number of pulses emitted during exposure to a predetermined area on the second object (photosensitive substrate) has a predetermined constant value, and each pulse has a predetermined value. Since the exposure amount control is performed for each pulse, it is convenient for smoothing the interference pattern relative to the first object or the second object for each pulse, and the exposure is performed with the minimum necessary number of pulses. Since it is performed, productivity can be improved. Further, since the exposure amount control is performed with reference to the target exposure amount set for each pulse, the light amount (energy amount) can be controlled more accurately than in the related art.

[Brief description of the drawings]

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

FIG. 2 is a configuration diagram illustrating an example of a high-speed darkening unit.

FIG. 3 is an explanatory diagram in the case of performing dimming by the configuration of FIG.

FIG. 4 is a diagram schematically showing a relationship between a fly-eye lens as an optical integrator and an incident beam.

FIG. 5 is a configuration diagram showing another example of the high-speed light reduction unit.

FIG. 6 is an explanatory diagram in the case of performing dimming by the configuration of FIG. 5;

FIG. 7 is an explanatory diagram in the case of performing dimming by the configuration of FIG. 5;

FIG. 8 is a configuration diagram showing still another example of the high-speed darkening unit.

FIG. 9 is a flowchart showing the operation of the embodiment shown in FIG. 1;

FIG. 10 is a graph showing a state of exposure amount control in the embodiment shown in FIG. 1;

[Description of Signs of Main Parts]

 R: Reticle W: Wafer PL: Projection lens 8: Main control system 10: Light source 13: High-speed darkening unit 14: Light amount control unit 20: Speckle reduction Unit 22: Speckle control unit 26: Light amount monitor unit 1

──────────────────────────────────────────────────続 き Continuation of front page (56) References JP-A-63-110722 (JP, A) JP-A-63-81420 (JP, A) JP-A-63-190333 (JP, A) 257327 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) H01L 21/027

Claims (16)

(57) [Specification] [Claims] 1. A projection optical system for projecting a pattern of a first object onto a predetermined area on a photosensitive second object, and irradiation for irradiating the first object with a plurality of pulse energies oscillated from a pulse light source. And exposing a pattern on the second object using an exposure apparatus having a system, wherein the pattern of the first object is projected onto the second object at a time interval corresponding to a predetermined oscillation frequency. Pulse energy is repeatedly emitted from the pulse light source, and a value corresponding to the exposure amount on the second object is detected for each emitted pulse energy, and the detection result and a target exposure amount set in advance for each pulse energy are detected. And adjusting the intensity of the pulse energy at high speed within the time interval of the repetitive oscillation based on 2. The number of pulses of a plurality of pulse energies to be applied to a predetermined area on the second object is determined in consideration of a proper exposure amount to be applied to the second object and a variation in intensity of pulse energy from the pulse light source. 2. The method according to claim 1, wherein the predetermined exposure amount control accuracy is determined. 3. When the irradiation of pulse energy from the pulse light source causes an interference pattern on the first object or the second object, the number of pulses determined so as to obtain the predetermined exposure amount control accuracy is: 3. The apparatus according to claim 2, wherein the number of pulses is set to be equal to or more than a minimum number of pulses necessary to reduce the visibility of the interference pattern to a predetermined accuracy.
The method described in. 4. The light source according to claim 1, wherein the pulse light source is a light source that emits laser light in a deep UV region including a band narrowing wavelength stabilizing mechanism, or a light source that emits pulse energy such as X-rays. The method according to any one of claims 1 to 3. 5. The pulse energy adjustment according to claim 1, wherein the pulse energy is adjusted by a dimming member capable of continuously changing the transmittance of the pulse energy from the pulse light source. The method described in. 6. An exposure apparatus for irradiating a first object with a plurality of pulse energies from a pulse light source and exposing a pattern formed on the first object to a predetermined region on a second object for manufacturing a semiconductor device. A manufacturing method for forming a semiconductor element on the second object by using the pulse light source, the intensity variation between pulse energies from the pulse light source, an appropriate exposure to be given to a predetermined area of the second object, and control accuracy for the appropriate exposure. Determining the average intensity of each pulse energy based on the number of exposure pulses determined in consideration of: and the first object by the determined average intensity pulse energy for exposure to a predetermined area on the second object. Starting irradiation to the substrate; during exposure, an exposure amount actually given by irradiation of pulse energy and a target exposure preset for each pulse energy Semiconductor device manufacturing method characterized by adjusting the next pulse energy irradiated on the basis of the bets, and a step of repeating between the number of exposure pulses. 7. The method according to claim 6, wherein the target exposure amount is set as an exposure amount integrated in a predetermined area on the second object as the number of exposure pulses increases. . 8. The method according to claim 6, wherein the target exposure amount is set as an exposure amount for each pulse energy of the exposure pulse number. 9. The pulsed light source emits a pulsed laser light in a deep UV region including a band narrowing wavelength stabilizing mechanism at a frequency of 100 Hz.
9. The method according to claim 7, wherein the light source emits light at the above repetition frequency or emits pulse energy such as X-rays. 10. When an interference pattern is generated on the first object or the second object by irradiation of pulse energy from the pulse light source, the number of exposure pulses is necessary to reduce the visibility of the interference pattern to a predetermined accuracy. The method according to claim 9, wherein the number of pulses is set to be equal to or greater than a minimum pulse number. 11. A projection optical system for projecting an image of a pattern formed on the first object onto a predetermined area on the second object, and a pulse energy from the pulse light source to the first object. An irradiation system including an integrator for making the illumination distribution substantially uniform on the upper side, the exposure amount actually given by irradiation of pulse energy during exposure is performed by a light receiving element that receives pulse energy emitted from the integrator. The method of claim 10, wherein the method is detecting. 12. The pulse energy is adjusted by a dimming member that is disposed in front of the integrator and that can continuously change the transmittance of pulse energy from the pulse light source. 12. The method according to 11. 13. An exposure apparatus for irradiating a first object with a plurality of pulse energies from a pulse light source and exposing a pattern formed on the first object to a predetermined region on a second object for manufacturing a semiconductor device. A method for forming a semiconductor element on the second object by the method described above, wherein a variation in pulse energy output from the pulse light source and an appropriate exposure amount to be given to a predetermined area of the second object are required for exposure. Determine the number of exposure pulses and
Presetting a target exposure amount to be given to a predetermined area of the second object for each number of exposure pulses; and a pulse light source while exposing a predetermined area of the second object with the number of exposure pulses. The pulse energy actually output from is measured, and the first energy is measured based on the actually measured energy and the target exposure amount for each of the preset number of exposure pulses.
Repeating the step of adjusting the pulse energy applied to the object. 14. The method according to claim 13, wherein the target exposure amount is set as an exposure amount that is integrated in a predetermined area on the second object as the number of exposure pulses increases. . 15. The method according to claim 13, wherein the target exposure amount is set as an exposure amount for each pulse energy of the exposure pulse number. 16. The pulse light source according to claim 1, wherein the pulse light source is a light source that emits a pulse laser beam in a deep UV region including a band narrowing wavelength stabilizing mechanism, or a light source that emits pulse energy such as X-rays. 16. The method according to 14 or 15.