KR100317684B1 - An exposure dose control device, a scanning type exposure device, and a device manufacturing method using these devices - Google Patents

Abstract

An object of the present invention is to always obtain a desired exposure amount or uniformity of illumination on a photosensitive substrate when applied to an exposure apparatus that performs exposure in a slit scan exposure system using a pulp light source.

The illuminance distribution on the scanning direction of the illumination area of the sleeve-like agent, and a trapezoidal shape, and the length of 1/2 of the width in the scanning direction of the wafer on the light intensity distribution on both sides of the inclined portion is a trapezoidal shape with each ₁ ΔD and ΔD ₂ , And the average value thereof is referred to as? D ₁₂ . In addition, D-half the width of the illumination area, when the minimum number of exposure pulses required for controlling the light exposure integrator within a predetermined accuracy in the respective points on the wafer as N _min, Lt; / RTI >

Description

An exposure dose control device, a scanning type exposure device, and a device manufacturing method using these devices

In the present invention, for example, a so-called slit scan, in which an illumination region such as a rectangular or arcuate shape is illuminated by using a pulse light source as an exposure light source, and a pattern on the mask is projected onto the photosensitive substrate by scanning the mask and the photosensitive substrate in synchronism with the illumination region, The present invention relates to an exposure amount control apparatus suitable for application in the case of controlling an exposure amount and a luminance uniformity to a photosensitive substrate within a predetermined range in an exposure type exposure apparatus.

2. Description of the Related Art Conventionally, when a semiconductor device, a liquid crystal display element, a thin film magnetic head, or the like is manufactured by using a photolithography technique, a pattern of a photomask or a reticle (hereinafter collectively referred to as a reticle) A projection exposure apparatus that projects a coated wafer or a photosensitive plate such as a glass plate is used. In recent years, one chip pattern or the like of a semiconductor device tends to be enlarged. In a projection exposure apparatus, a larger area for exposing a pattern of a larger area on a reticle onto a photosensitive substrate is sought.

In order to improve the resolution of the projection optical system, it is difficult to increase the exposure field of the projection optical system in terms of design or manufacture, in order to improve the resolution of the projection optical system in response to the miniaturization of the pattern of the semiconductor element or the like have. Particularly, when a reflective refraction system is used as the projection optical system, the shape of the exposure field of the aberration aberration may be an arc-shaped region.

In order to cope with the enlargement of the transfer target pattern and the limitation of the exposure field of the projection optical system, the reticle and the photosensitive substrate are synchronized with respect to an illumination area (such as a slit-shaped illumination area) of a rectangular, A so-called slit scan exposure type projection exposure apparatus has been developed in which a pattern having a larger area than that of the slit-shaped illumination region of the reticle is exposed on the photosensitive substrate. In general, in the projection exposure apparatus, the conditions of the proper exposure amount and the uniformity of the light intensity for the photosensitive material on the photosensitive substrate are determined. Therefore, even in the projection type artillery apparatus of the slit scan knockout type, the exposure amount for the photosensitive substrate is set to a predetermined There is provided an exposure amount control device for maintaining the illumination uniformity of the exposure light with respect to the wafer at a predetermined level.

In recent years, it is also required to increase the resolution of a pattern to be exposed on a photosensitive substrate. One of the methods for increasing the resolution is to shorten the exposure light. In this regard, among the light sources currently available, there is a pulse oscillation type laser light source (pulse light source) such as an excimer laser light source, a metal vapor laser light source, or the like in which the wavelength of emitted light is short. (Pulse light quantity) of the pulse light emitted by the pulse light source fluctuates within a predetermined range for each pulse light emission, unlike a continuous light emission type light source such as a prism, a granna, or a mercury lamp.

Therefore, the conventional exposure amount control apparatus sets the parameter? P / < p, which indicates the fluctuation of the pulse light amount, to? P, where? P is the pulse light amount of the pulsed light from the pulsed light source on the photosensitive substrate, > Is a normal distribution (in fact, random). Then, a pulse irradiated to a region having a predetermined width in the scanning direction on the photosensitive substrate (which is referred to as a pulse number integration region) scanned relative to the slit-shaped illumination region by pulse light and the exposure region which is a conjugate projection If the number of lights is N, control is performed so that the integrated exposure amount reaches the appropriate exposure amount within a predetermined allowable range by using the fact that the variation of the integrated exposure amount after exposure is (? P / <p>) / N ^{l / 2} there was

Also, as disclosed in U.S. Patent No. 4,822,975, in the case of a method of measuring the position of a photosensitive substrate or a reticle and exposing the photosensitive substrate or reticle by emitting a light emission trigger signal in synchronization with the measurement result, A variation in time from the start of the actual measurement in the measurement instrument (laser interferometer or the like) to the output of the measurement result is taken into account.

Such a fluctuation of the light emission position may cause a region having a large integrated exposure amount locally or a small region locally on the gauge-silvered substrate after the scan embedding. This is because, assuming that the light intensity in the slit-shaped illumination area rises from the non-illumination area before and after the scanning direction stepwise to zero to 100%, the scanning direction width of the slit- (Corresponding to the case where the number of pulses N irradiated to the above-described pulse number integration area is 1), and the adjacent pulse This is because the integrated exposure amount at the joints between the integrating areas may be doubled or become 0 due to the positioning accuracy of the photosensitive substrate.

Regarding this, regarding a case where a screen (chip pattern) is connected in the non-scanning direction by using a continuous light emission source such as a mercury lamp (equivalent to the case where the number of pulses N is 1 in the scanning direction using a pulse light source) 3,538,828 discloses a technique of making a shape of a light intensity distribution obtained by integrating a slit-shaped exposure area in a scanning direction in a non-scanning direction (that is, a direction connecting the screen) into an isosceles trapezoidal shape. This method is applied to the case where the pulse number N irradiated to the point of the photosensitive substrate steel is 1, so that it is possible to alleviate the problem that the integrated exposure amount fluctuates at the joint. In addition, when the number of pulses N is 1 in the scanning direction by using a pulsed light source, it is apparent that the content is substantially disclosed in the above US Patent No. 3,538,828. A method of making the illuminance distribution in the direction of the screen joint (scanning direction) into an isosceles triangle shape or an isosceles trapezoid shape is disclosed in US 4,822,975.

In the above-described conventional technique, when the half of the light intensity width in the scanning direction of the slit-shaped illumination area and the light-emission position interval are equal (in the pulse number integration area, the number of pulses N is 1) And the conditions for obtaining the desired exposure amount and uniformity of illumination have not been clarified. The conditions for obtaining a desired exposure amount and uniformity of luminance were not clear even in the case of exposing a plurality of pulses with the pulse number N being 2 or more in each pulse number integration area on the photosensitive substrate.

An object of the present invention is to provide an exposure control apparatus capable of always obtaining a desired exposure amount or uniformity of illumination on a photosensitive substrate when applied to an exposure apparatus that performs exposure in a slit scan exposure system using a pulse light source do.

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

2 is a flowchart showing an example of an exposure operation in an embodiment of the present invention.

FIG. 3A is a diagram showing the distribution state of the pulse light amount in the embodiment of the present invention. FIG.

FIG. 3B is a diagram showing a distribution state of light emission timings in the embodiment of the present invention. FIG.

4 is a view showing the illuminance distribution by the pulse light on the exposure surface of the wafer in the embodiment of the present invention.

5 is a timing chart showing an emission trigger pulse supplied to the pulse light source according to the embodiment of the present invention.

6 is an enlarged plan view showing a certain shot area on the wafer and a slit-shaped arcuate area.

Description of the Related Art [0002]

1: pulse light source 2: binocular optical system

3: Light amount adjusting means 13:

14: Operation unit 21: Trigger control unit

22: input / output means 23: memory

The exposure amount control device according to the present invention is an exposure amount control device that illuminates pulsed light from a pulsed light source in a predetermined illumination area and synchronizes the mask on which the transfer patina is formed and the photosensitive substrate relative to a predetermined illumination area, Wherein the illuminance distribution in the scanning direction of the predetermined illumination area is set to a trapezoidal shape, and the trapezoidal shape of the trapezoidal shape referred to the average value of one-half the length of the scanning direction of width on both sides of board-shaped inclined portion of the light intensity distribution of ΔD and _l2, d d, and the half of the width of the scanning direction on the substrate of the illumination distribution on the predetermined illumination region, the minimum number of exposure pulses required for controlling the light exposure integrator within a predetermined accuracy at each point on the substrate to as N _min, at least the following Condition as a necessary condition.

In this case, the half lengths of the widths in the scanning direction on the substrate W in both side inclined portions of the illuminance distribution of the predetermined illumination region 24 are denoted by? D ₁ and? D ₂ , respectively, When the allowable value of the smear is set to [U _scan ] _max , it is preferable to satisfy the following condition again.

The visual field stop 7 for illuminating the predetermined illumination area 24 is defocused at a predetermined position at a position optically conjugate with the exposure surface of the substrate W so that a predetermined illumination area 24 May have a trapezoidal shape in the scanning direction on the substrate W.

According to the present invention, the illuminance distribution in the scanning direction on the substrate W in the predetermined illumination area (slit-shaped adjustment area) is made trapezoidal as shown in Fig. In Fig. 4, D is the width of the trapezoidal shape of the illuminance distribution on the substrate W in the scanning direction (width between points where the illuminance is 1/2 of the maximum value), D is the inclination Of the width in the scanning direction are denoted by? D _l and? D ₂ , respectively. Then, the width of the ΔD _l and width average of ΔD ₁₂ of ΔD _2, when the conditions of equations (A) to satisfy a requirement, all points on the substrate (W) is at least one pulse in the inclined portion (edge portion) of the light intensity distribution And the fluctuation of the integrated exposure amount in the scanning direction (thermal uniformity of illumination uniformity) can be made less than the exposure energy for one pulse.

For a width of one-half the average value ΔD and ΔD _{l l2} of ΔD ₂ of the illumination distribution of the scan direction either side inclined section, a conditional expression (A) is charged, but, width ΔD ₁ and the ratio (symmetry) of the two sides ₂ ΔD A predetermined condition is also required. That is, by simply determined by the average value ΔD _12, for example, when the width ΔD ₁ of the inclined portion side is very small, it is larger in width ΔD ₂ inclined portion on the other side is, the variation in the integrated amount of exposure in the scanning direction concerned occur periodically have. Thus, by imposing the conditional expression (B) on the symmetry of the widths? D ₁ and? D ₂ , the fluctuation of the integrated exposure amount in the scanning direction (fluctuation of the illuminance distribution) becomes equal to or smaller than [U _scan ] _max .

By disposing the field stop 7 for illuminating the predetermined illumination area 24 with a predetermined amount of defocus from a position optically conjugate with the exposure surface of the substrate w, may set the value in the case of the illuminance distribution on the scanning direction in a trapezoidal shape, a simple structure with the average value of the width of the inclined portion of the light intensity distribution trapezoidal 1/2 ΔD ₁₂ on the substrate (W) to a desired value . Further, the values of? D ₁ and? D ₂ , which are 1/2 of the widths of both inclined portions of the trapezoidal roughness distribution, are substantially the same, and the symmetry is good.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings. This embodiment applies the present invention to an exposure amount control system of a projection exposure apparatus of a slit scan exposure type having a pulse oscillation type canopy light source such as an excimer laser light source as a light source.

1 shows a projection exposure apparatus of this embodiment. In Fig. 1, a laser beam emitted from a pulse oscillation type pulse light source is incident on a beam shaping optical system 2 composed of a cylindrical lens or a beam expander The cross-sectional shape of the beam is shaped so as to efficiently enter the subsequent fly-eye lens 4. The laser beam emitted from the stepped optical system 2 is incident on the light amount adjusting means 3. [ The light amount adjusting means 3 has a coarse adjustment portion and a fine adjustment portion of the transmittance. The laser beam emitted from the light amount adjusting means (3) enters the fly-eye lens (4). The fly-eye lens 4 is for illuminating the subsequent visual field stop 7 and the reticle R with uniform illumination.

The phaser beam emitted from the fly-eye lens 4 is incident on the beam splitter 5 having a small reflectance and a high transmittance. The laser beam having passed through the beam splitter 5 is reflected by the first relay lens 6, (7) illuminates the image with a uniform illuminance. The shape of the opening of the field stop 7 in this embodiment is rectangular.

The laser beam having passed through the field stop 7 passes the reticle R on the reticle stage 11 through the second relay lens 8, the bending mirror 9 and the main condenser lens 10 to a uniform illumination . A rectangular slit-like illumination area 24 (see FIG. 1) on the reticle R, which is conjugate with the pattern formation surface of the field stop 7 and the reticle R and the exposure surface of the wafer W and which is conjugate with the opening of the field stop 7 ) Is irradiated with a laser beam. The shape of the slit-shaped illumination area 24 can be adjusted by changing the shape of the opening of the field stop 7 through a driving part (not shown).

The pattern image in the slit shaped illumination area 24 on the reticle R is projected and exposed on the wafer W via the projection optical system 15. [ An area that is conjugate to the slit-shaped illumination area 24 and the projection optical system 15 is defined as a slit-shaped exposure area 24W. When the Z axis is held parallel to the optical axis of the projection optical system 15 and the scanning direction of the reticle R with respect to the slit-shaped illumination area 24 in the plane perpendicular to the optical axis is set as the X direction, 11 are scanned in the X direction by the reticle stage driving unit 12. [ The reticle stage driving unit 12 is controlled by a main controller 13 that controls the operation of the whole apparatus. The reticle stage driving section 12 is provided with a length measuring device (laser interferometer or the like) for detecting the coordinates of the reticle stage 11 in the X direction, The coordinates are supplied and returned to the main control system (13).

On the other hand, the wafer W is placed on the XY stage 17 which can be scanned in at least the X direction (left and right direction in Fig. 1) via the wafer holder 16. [ Although not shown in the drawing, a Z stage or the like for positioning the wafer W in the Z direction is provided between the XY stage 17 and the wafer holder 16. In the slit scanning exposure, the wafer W is moved in the -X direction (in the -X direction) with respect to the exposure area 24W through the XY stage 17 in synchronism with the reticle R being scanned in the + X direction Or X direction). The main control system 13 controls the operation of the XY stage 17 through the wafer stage driving unit 18. [ The wafer stage driving unit 18 is provided with a measuring apparatus (laser interferometer or the like) for detecting the X and Y coordinates of the XY stage 17 and measures the X coordinate and the Y coordinate of the XY stage 17 The coordinates are supplied to the main control system 13.

The laser beam reflected by the beam splitter 5 is captured by an exposure amount monitor 19 composed of a photoelectric conversion element so that the photoelectrically converted signal of the exposure amount monitor 19 is supplied to the arithmetic unit 14 via the amplifier 20 do. The relationship between the photoelectric conversion signal of the exposure amount monitor 19 and the illuminance of the pulse exposure light on the exposure surface of the wafer W is obtained in advance. That is, the photoelectric conversion signal of the exposure amount monitor 19 is calibrated in advance.

Not only the fluctuation of the pulse light quantity of the pulse light outputted from the pulse light source 1 but also the light emission timing of each pulse light is measured from the photoelectric conversion signal of the calculation section 14 and the exposure amount monitor 19. [ The variation of the pulse light quantity and the variation of the light emission timing are supplied to the main control system 13. At the time of breakage, the arithmetic unit 14 obtains the integrated exposure amount to the wafer W by integrating the photoelectric conversion signals for each pulse light, and supplies the integrated exposure amount to the main controller 13.

The main control system 13 controls the timing of the pulse light source 1 by supplying the light emission trigger signal TP to the pulse light source 1 via the trigger control unit 21. [ The arithmetic section 14 calculates the light emission amount of the pulse light source 1 based on the timing at which the light emission trigger signal TP is transmitted from the trigger control section 21 to the pulse light source 1 and the light reception timing detected by the arithmetic section 14. [ The variation of the time until the pulse light source 1 actually emits, that is, the variation of the light emission timing of the pulse light source 1 can be obtained after the trigger is supplied. The main control system 13 adjusts the output power of the pulse light source 1 as necessary or adjusts the transmittance in the light amount adjusting means 3. [ The operator can input the pattern information of the reticle R to the main controller 13 through the input / output means 22 and the main controller 13 has the memory 23 capable of storing various kinds of information .

In addition, as described in U.S. Patent Application Serial No. 08 / 209,269 filed on March 14, 1994 by the present applicant, when exposure is performed by a slit scan exposure method using a pulse light source, The fluctuation of the emission timing of the pulse light source with respect to the position (hereinafter referred to as the emission position fluctuation) causes fluctuations in the integrated exposure dose. As factors of the light source side among the factors of the light emission position variation, there is a variation in the time from the emission of the light emission trigger signal to the pulse light source until the pulse light source actually emits light. On the other hand, as a factor of the exposure apparatus side, as described in the above-mentioned U.S. Patent Application Serial No. 08 / 209,269, the substrate and the reticle are scanned at a constant speed in synchronization with each other, and the light emission trigger signal is transmitted at the same time interval In the case of the exposure method, the scanning speed is not uniform.

It is described below that the influence due to the fluctuation or speed variation is reduced.

Next, an example of the operation in the case of exposing the letter W (R) pattern onto the wafer W in this example will be described with reference to the flowchart of Fig. First, in step 101 of FIG. The operator sets the desired exposure amount S (mJ / cm2) on the wafer surface to the subject agent 13 via the input / output means 22. [ Next, in step 102, the main control system 13 instructs the trigger control unit 21 to emit dummy light. Thereafter, the test light emission (dummy light emission) of the pulse light source 1 is performed in a state where the wafer W is retreated to a place where the wafer W is not exposed (an area outside the exposure area 24W). In the dummy light emission, for example, approximately 100 pulses of pulse light are emitted, and the distribution of the pulse light amount and the distribution of the light emission timings, which are known from the photoelectric conversion signals detected by the exposure amount monitor 19, It becomes a normal distribution type.

3A shows a distribution of the pulse light amount p (amount converted on the exposure surface of the wafer) (mJ / cm 2) of each pulse light measured by the dummy light emission, and FIG. 3B shows a distribution (Sec) of the light-emitting element 1 according to the present invention. In step 103, the arithmetic unit 14 obtains the average pulse light quantity p (mJ / cm2 占 pulse) on the exposure surface of the wafer from the distribution data of the pulse light quantity p shown in Fig. 3A, The average value < [delta] > of the variation of the light emission timing is obtained from the distribution data of the light emission timing delta shown in Fig.

Thereafter, in step 104, the arithmetic unit 14 obtains the deviation? P of the pulse light quantity at three times (3?) Of the standard deviation from the distribution data of the pulse light quantity p shown in Fig. 3A, From the distribution data of the emission timing delta, the deviation DELTA delta of the emission timing at three times the standard deviation is obtained. Then, the calculating section 14 calculates the variation? P /? P> of the pulse light quantity and the variation? Delta /?> Of the light emission timing.

Next, in step 105, the desired exposure amount S (mJ / cm2) designated through the input / output means 22 is sent from the main control system 13 to the calculation unit 14, The number of exposure pulses N is calculated from the following equation using the desired exposure amount S and the average pulse light amount <p> calculated in step 103. [

Here, int (A) represents an integer obtained by subtracting the decimal part of the real number (A). The width D (cm) in the scanning direction of the slit-shaped open area 24W on the wafer surface and the width D (cm) of the slit-shaped open area 24W on the wafer surface from the memory 23 via the main controller 13, information on f (Hz) is sent to the arithmetic unit 14. The arithmetic unit 14 obtains the scanning speed v (cm / sec) on the wafer surface using the number of exposure pulses N, the width D,

In the subsequent step 106, the arithmetic unit 14 calculates the minimum number of exposure pulses N _min necessary for controlling the integrated exposure dose and the illumination uniformity on the exposure surface of the wafer W within a predetermined precision However, the calculation formula will be described later in detail. The number of exposure pulses N and the minimum number of exposure pulses N _min are supplied to the main controller 13.

Further, when both of the variation amount? P / <p> of the pulse light amount and the variation?? Of the light emission timing are all small, the variation amount? P / The operation of step 104 is omitted by storing the number of exposure pulses N _min in the memory 23 and the minimum number of pulse repetitions N _min in the step 106 is read out from the memory 23 do.

Next, at the step 107, the main control system 13 compares the exposure pulse number (N) and the minimum number of exposure pulses (N _min), the case of N <N _min, the processing proceeds to step 108 , The main control system 13 roughly reduces (roughly adjusts) the transmittance of the light amount adjusting means 3 in Fig. Then, steps 102 to 107 are repeated to compare the number of exposure pulses N and the minimum number of exposure pulses N _min again. Therefore, The transmittance of the light amount adjusting means 3 is set. As an example of means for roughly adjusting the transmittance, a device equipped with a ND filter having a plurality of transmittances on a turret plate as disclosed in Japanese Patent Application Laid-Open No. 63-316430 and U.S. Patent No. 4,970,546 .

Next, in step 107, , The process proceeds to step 109 to finely adjust the light amount of the pulse light. That is, fine adjustment of the average pulse light amount < p > is performed so that S / < p > At this time, since the scanning speed v is also determined in accordance with the number N of exposing pulses obtained from the equation 1 in step 105, the value of the exposure pulse number N is not changed, it is preferable to perform fine adjustment of the pulse light amount in the direction of slightly increasing <p>. Conversely, when the number of exposure pulses N becomes N + 1 as the average pulse light amount <p> becomes slightly smaller by fine adjustment of the pulse light amount, the scanning speed v is again You can get it.

As an example of the light amount fine adjustment means for finely adjusting the energy of the pulse light, as disclosed in Japanese Patent Application Laid-Open No. 2-135723 (U.S. Patent Application No. 438,091 filed on November 16, 1989) Two gratings which are arranged along the same line and in which a line and space pattern is formed at the same pitch, and means constituted by a mechanism for displacing these two gratings slightly transversely. When two gratings are used, the pulse light in the area where the bright portion of the first grating overlaps with the bright portion of the second grating is irradiated to the wafer W side. Therefore, the relative amount of deviation of the two gratings in the lateral direction It is possible to finely adjust the pulse light amount irradiated to the wiper W side.

The main control system 13 starts scanning of the reticle R and the wafer W via the reticle stage 11 and the XY stage 17 on the wafer side. In Fig. 1, for example, when the reticle R is scanned in the X direction, the wafer W is scanned in the -X direction. In this example, although the scanning speed v (conversion value on the exposure surface of the wafer W) of the reticle R and the wafer W is determined by the equation 2, after the start of scanning, the XY stage on the wafer side To is the set time until the scanning speed of the scanning optical system 17 reaches the scanning speed v.

In this embodiment, the light emission trigger of the pulse light source 1 is assumed to emit light at the same time intervals as disclosed in " U.S. Patent Application Serial No. 08 / 209,269, 1994.3.14 ". Here, at the start of scanning, the time t is reset to 0 and the parameter j is reset to 0. Then, as shown in step 111, the main control system 13, when the time t becomes (To + j? T) , The light emission trigger signal TP is turned on (pulse of high level " 1 ") to the pulse light source 1 via the trigger control section 21. [ In response to this, the pulse light source 1 generates one pulse light, and the pattern of the reticle R is exposed on the wafer W. [

5 shows the light-emission trigger signal TP of this embodiment. As shown in Fig. 5, the light-emission trigger signal TP is turned on at a certain period DELTA T from the point of time t reaching To. Therefore, the pulse light source 1 emits light at a constant period DELTA T, and the oscillation frequency f of the pulse light source 1 is represented by 1 / DELTA T. The oscillation frequency f is a value stored in the memory 23 in advance. When the parameter j is not equal to the integer Nt in step 113 and the parameter j is 1 in step 112, the process proceeds to step 111 where the pulse light source 1 emits light The pulsed light source 1 emits N _T pulsed light at a constant frequency f (at a constant cycle? T).

1, assuming that the width in the scanning direction (X direction) of one shot area on the wafer W is L1 and the width in the scanning direction of the exposure area 24W is D, Since the distance that the wafer W is scanned in the cycle is v / f, the minimum value of the number of light emissions N _{T in the} pulse field is as follows.

In practice, a predetermined number of pulsed light beams are added at the start and end of the scanning. When the number of emitted pulses reaches N _T in step 113, the routine proceeds to step 114, where the main routine 13 finishes the scanning and exposure of the reticle R and the wafer W .

Thereby, the entire pattern of one shot on the reticle R is exposed to one shot area on the wafer W. [ In this case, in this example, the light emission of the pulse light source 1 is performed at a constant frequency irrespective of the X coordinate of the reticle stage 11 and the X coordinate of the XY stage 17 on the wafer side. However, the reticle stage 11 and the XY stage 17 are scanned at a constant speed, respectively. Therefore, in the light emission position fluctuation (fluctuation of the light emission timing of the pulse light source 1 with respect to the position of the woby W) in this example, the fluctuation of the light emission timing obtained in step 104 < 105) multiplied by the scanning speed v obtained from the equation (2), and the speed fluctuation.

Next, a method for calculating the minimum number of exposure pulses N _min in step 106 of FIG. 2 will be described in detail. First, the illuminance distribution (illumination region 24) formed by the field stop 7 in FIG. 1 and the illuminance distribution along the cross section of the exposure region 24W on the wafer W in the scanning direction are approximately in a trapezoidal shape .

A method of forming the trapezoidal shape of the illuminance distribution along the cross section in the scanning direction will be described later.

Fig. 4 shows distribution curves 25A, 25B and 25C in the X direction of the illuminance when the illuminance distribution along the cross section in the scanning direction is approximated to a trapezoid shape for each pulse light. Actually, since the width of the exposure area 24W in the scanning direction is several millimeters and the width? D of the faint part of the roughness is about 100 to several hundreds of micrometers, the roughness distribution along the cross- to be. The peak values of the distribution curves 25A, 25B and 25C are p ₁ , p ₂ and p ₃ , respectively, and in the distribution curve 25A, 25B and 25C, And the width in the scanning direction is D in common. D, which is half the width, can be considered as the width in the scanning direction of the exposure area 24W.

4, the peak of the first pulse (distribution curve 25A), the second pulse (distribution curve 25B), and the third pulse (distribution curve 25C) The amount of light varies, and the light emission interval of each pulse light is not constant due to the variation of the light emission position. Here, with respect to the first pulse second distribution in the inclined portion of either side of the curve (25A), each value is the peak value of l / 2 area under the (width ΔD ₁ and scope of ΔD _2), first after the 2nd pulse When the number of overlapped and exposed pulses is set to N ₁ and N ₂ , the following equation is established.

That is, since the N represented by a pulse number of formula l which is exposed between for scanning the exposure area of width D, the width ΔD _1, the width ΔD _2, width 2ΔD _1, and the width 2ΔD _2, is inversely related to the width of each of the regions of The number of pulses corresponding to the number of pulses is exposed.

However, since attention is focused on the vicinity of a point embossed by the exposure light at the center (the position for determining the 1/2 width D) of both inclined portions of the trapezoidal shape illuminance distribution in each pulse number accumulation region on the wafer, 2ΔD ₁ and the width of the pulses in the region of 2ΔD ₂ was set to each of (2N ₁ +1) and (2N ₂ +1).

Next, the 1/2 width D of the slit illumination area is set to 1/2 at a plurality of positions in the non-scanning direction (Y direction) perpendicular to the scanning direction in advance, as disclosed in Japanese Patent Application No. 5-14483, 2 width is measured, and the average value is set to < p >. At this time, the pulse light emission interval Xs, which is the distance that the wafer W moves in the scanning direction during one period of the pulse light emission of the pulse light source 1, is as follows.

At this time, the difference between the actual 1/2 width D of the exposure region 24W having a trapezoidal illumination distribution at a certain position in the non-scanning direction and the average measured value < D > ΔX is as follows.

Generally, this difference DELTA X arises from the measurement accuracy of the dimensional accuracy of the field stop 7 in Fig. 1, the aberration of the illumination system, and the average value < D > of the l / 2 width.

Here, assuming that the stray energy on the wafer W of the pulse light when the pulse light source 1 occurs at the i-th time is p _i and the scanning direction (X direction) of the wafer W when the i- (Light emitting position offset) from the target position of the actual position of the light emitting element is set to? _I. (X) in the X direction after the wafer W is scanned with respect to the slit-shaped exposure region 24W, when the average value and the variation of the light emission position offset _i are respectively < alpha > and DELTA alpha, (mJ / cm ² ), the center point of the inclined portion having the width 2ΔD ₁ is set as X = 0, as follows.

Alternatively, with respect to a point on the wafer W where one pulse is not exposed, the illuminance distribution I (X) is as follows at an inclined portion of the trapezoidal shape illuminance distribution.

At this time, the exposure amount in the scanning direction and the accuracy U (X) of the luminance uniformity are given by the following equations after slightly cumbersome calculation in Equation 7 and Equation 8, taking into account the variation of the pulse energy pi and the light emission position offset? _I. However, it is equivalent that the position X is different from the other and the average value < alpha > of the light emission position offset alpha is different.

Alternatively, in the inclined portion of the trapezoidal shape of the illuminance distribution, the accuracy U (X) is as follows with respect to the point on the wafer W where no pulse is exposed.

The right and left inclined portions of the illuminance distribution are approximately symmetrical and approximate as follows.

By this approximation, Equation 9 is simplified as follows.

The second term on the right side of Expression (12) represents the exposure amount control reproduction accuracy A _rep in the pulse number accumulation region on the wafer (W). This exposure control control reproduction accuracy A _rep needs to satisfy the following conditional expression.

The right side of the inequality in Equation 13 becomes equal to the exposure amount control reproducibility obtained from (Equation 10). From equation (13), the number of exposure pulses N is as follows.

Here, assuming that the variation (? P / <p>) of the pulse energy is 5% (at 3?) By setting the desired exposure control reproduction reproduction accuracy [A _rep ] _max to 1% (with 3σ being three times the standard deviation) As to the number of pulses N from 14, (Pulse) is obtained. Therefore, the minimum number of exposure pulses N _min in each pulse number integration area may be set to 25 pulses. However, in practice, when the coherence of the pulse light source 1 is high, as disclosed in the US Patent No. 4,970,546 by the present applicant, the contrast due to the interference pattern or the speckle pattern ) Is added to the minimum number of exposure pulses N _min .

Next, the term of (l / N) (1/2) 1 {(2N ₁ +1) / D ₁ - (2N ₂ +1) / D ₂ } X in the _third right- Is the thermal spraying uniformity in the scanning direction in each pulse number accumulation area caused by the asymmetry of the inclined portions on both sides of the illuminance distribution.

The term of (1 / N) ((2N ₁₂ +1) / (2ΔD ₁₂ )} ΔX in the third term of the right side is the exposure amount bias generated by the difference ΔX of width 1/2 shown in the equation X is constant), or the thermal spray uniformity in the non-scanning direction (when the difference DELTA X in the non-scanning direction is not constant).

The fourth term on the right side of Expression 12 is the reproduction precision of the third term on the right side and not only the pulse energy variation (? P / <p>), but also the fluctuation DELTA alpha of the light emission position offset represents the deterioration of the luminance uniformity in the scanning direction Contributes to precision. Assuming here that the thermal spraying uniformity in the scanning direction is U _scan , the exposure amount bias or the thermal spraying uniformity in the non-scanning direction is U _Bias , and the reproduction precision of (U _scan + U _Bias ) is A _scan , .

Here the illuminance uniformity of heat in the scanning direction shown in equation 15 and equation 16 pots U _scan, and the amount of exposure bias or illumination uniformity of heat pollen of the non-scanning direction U _Bias is to contribute to the illumination uniformity and exposure bias, and The state is shown in FIG.

6 shows a state of a shot area 26 on the wafer W in Fig. 1, and a slit-shaped exposure area 24W is formed on the right side of the shot area 26 in the X direction. For convenience of explanation, the shape of the exposure region 24W is such that the width in the scanning direction becomes narrower at the central portion in the non-scanning direction (Y direction). That is, the width of the columnar direction at both ends is (<D> + ΔX), and the width in the scanning direction at the center in the Y direction is (<D> -ΔX).

The pattern of the reticle R is sequentially projected and exposed in the shot area 26 by scanning the shot area 26 having the width L1 in the X direction with respect to the exposure area 24W. If the interval at which the wafer moves in the X direction during one light emission period of the pulse light source is Xs, the regions 27 ₁ , 27 ₂ , 27 ₃ , ..., and 27 ₂ , Are the pulse number integration areas. In this case, the thermal spray U _scan of the uniformity of illumination in the scanning direction represented by the equation (15) corresponds to the pulse number integrating areas 27 ₁ , 27 ₂ , ..., The irregularity of illumination occurring periodically in the scanning direction. The exposure amount bias shown in the expression (16) or the uniformity uniformity thermal uniformity U _Bias in the non-scanning direction can be calculated by the following equation: U _Bias in the exposure area 24W (in the irradiation field) Direction, the unevenness of illumination in the non-scanning direction in the shot area 26 on the wafer appears.

Specifically, in FIG. 6, a portion in the shot region 26 passing through a region wider in the scanning direction in the exposure region 24W than the average value <D>, such as (D) + ΔX to D Excess exposure occurs in the regions 26a and 26c and exposure is carried out in the exposure region 24W through the region passing through the region narrower than the average value <D> such that the width in the scanning direction is from <D> to <D> Within the fraction area 26b in the area 26, the exposure becomes insufficient. On the straight line partial area passing through the exposure area 24W having the width in the scanning direction <D>, that is, on the straight line of the Y coordinate Y ₁ and Y ₂ on the shot area 26, the appropriate exposure amount do.

In addition, the illuminance distribution in the scanning direction of the elongated region 24W in Fig. 6 is actually trapezoidal as shown in Fig. One example of such a method of making the illuminance distribution into a trapezoidal shape is as shown in Fig. 1 in which the field stop 7 is moved in the direction of the pattern forming surface of the reticle R and in the plane conjugate with the exposure surface of the wafer W Defocus. When the illuminance distribution in the scanning direction of the exposure area 24W is made trapezoidal by the defocus of the field stop 7 as described above, the inclined portion of the trapezoidal illuminance distribution is projected on the edge of the field stop 7 The oblique portion of the illuminance distribution is referred to as a faint portion of the edge portion of the exposure region 24W and the width of the oblique portion in the scanning direction is called the width of the faint portion.

Therefore, the allowable value of the width of the faint portion of the edge portion in the scanning direction in the slit-shaped exposure region 24W is obtained from Expression (16). Considering Equation 4 and Equation 11, Equation 16 is as follows. However, the permissible value of the exposure dose bias or the uniformity of the uniformity U _{bias in} the non-scanning direction was set to [U _Bias ] _max .

Or Equation 16 is as follows.

Is the following formula 20 in which the difference ΔX condition is obtained from equation 18, the exposure area (24W) faint width of 1/2 of the condition of the following equation 21, the average value ΔD ₁₂ of the parts on both sides of the equation 19 is obtained from.

In order for Equation 21 to always hold, the following relationship is necessary.

If the half width D in the scanning direction of the slit-shaped exposure area 24W is 5 mm and the allowable value U _{Bias max} of the hot plate U _{Bias in} the exposure amount bias or in the non-scanning direction is 0.5% From equation 20 . In addition, with respect to the minimum number N _min exposure pulse, , The average value DELTA D &lt; ₁₂ &gt; Is obtained.

From Equation 9, , Equation (15) becomes as follows.

The right side of this equation becomes maximum when N = N _min , and N ₁ = 0 and N ₂ = 0. Thus, release from the formula 23 with N = N _min, also _{N 1 = 0, N 2 =} 0, the following relation is obtained.

From equation (24), the symmetry of? D ₁ and? D ₂ (see FIG. 4), which is 1/2 of the width of the faint part of both edge portions in the scanning direction in the slit-like elongated region 24W, is obtained. Here, if the permissible value of the thermal spraying uniformity U _{scan in} the scanning direction is set to [U _ssan ] _max , Expression 24 is as follows.

This is a condition for symmetry with respect to the faint part of the both side edge parts in the scanning direction of the exposure area 24W. For example, _{supposing that} the allowable value of the U _scan [U _scan ] _max is 0.2%, the minimum number of exposure pulses N _min is 25, and the average value of the width ΔD ₁₂ is 100 μm, Expression 25 is as follows.

Finally, by substituting the equations (4) and (15) into the equations (17), the following equations are obtained.

When X = -ΔD ₁₂ and N ₁₂ = 0, the right side of equation (27) becomes maximum, so equation (27) becomes as follows.

When the tolerance of this _scan is set to [A _scan ] _max , Equation 28 can be modified as follows.

Here, the minimum exposure pulse number N _min of 25, half the width D of the exposure area (24W) 5mm, and the variation of the energy pulse (Δp / <p>) with 5%, the [U _{scan] max} 0.2% , [U _Bias ] _{max is set} to 0.5%, and [A _scan ] _{max is set} to 0.5%, equation (29) becomes as follows.

Thereby, the standard deviation (tolerance) of the variation of the offset of the light emission position (nonuniformity of the light emission position) is obtained.

[A _rep ] _max , [U _Bias ] _max , [U _scan ] _max , [A _scan ] _max D, and (Δp / <p |), The minimum number of exposure pulses N _min , the difference |? X | between the widths in the scanning direction of the trapezoidal shape of the illuminance distribution in the exposure region 24W, the width of the faint portions on both sides of the illuminance distribution, 2 the average value of ΔD _l2, the symmetry of light parts of the both sides of _{| (l / ΔD 1) -} (1 / ΔD 2) | a, and the allowable value of variation Δα of the offset of the light emission position is obtained.

Next, a description will be given of a method of forming both inclined portions of the illuminance distribution in the scanning direction of the exposure area 24W. First, the first method is to change the film thickness of an opaque material such as chromium on a substrate (quartz or the like) transparent to light having the same wavelength as the exposure light so as to have a trapezoidal transmittance distribution as shown in Fig. 4, A method using a mask (optical filter) can be mentioned. This mask may be provided in close contact with the field stop 7 in Fig. 1 as an illumination field stop.

As a second method, there is a method in which an illumination field stop having a rectangular transmittance distribution along the cross section is defocused from a pattern formation surface of the reticle R or a position conjugate to the exposure surface of the wafer W . At this time, if the light from one point on the field stop 7 is blurred in a circular form with a radius? R on the exposed surface of the wafer W, the light flux passing through the cross section in the scanning direction (X direction) Assuming that the illuminance distribution of the edge portion of the illuminance distribution is represented by a function I (X /? R) of X /? R, the following is obtained.

However, in expression (30), the exposure energy p is normalized to 1. The equation 30 is expanded by Taylor expansion around the origin 0 with respect to X /? R.

Comparing Equation 31 and Equation 7, (? / 2)? R ₁₂ and 2? D ₁₂ correspond to each other. Therefore, with respect to the condition of the average value? D ₁₂ of 1/2 of the width of the faint part of the illuminance distribution obtained from equation (21), the radius? R of the faint part of the point shape may be determined so that the following condition is satisfied.

As described above, according to the present embodiment, since the exposure amount control is performed in consideration of not only the pulse energy of the pulse light source 1 but also the fluctuation of the light emission timing, the advantage of improving the control accuracy of the exposure amount and the luminance uniformity . Since the light emission trigger of the pulse light source 1 is supplied to the pulse light source 1 at regular intervals and the scanning speed of each of the mask R and the wafer W at the time of exposure is made constant, The influence of the fluctuation in the reading timing of the measurement result from the measuring apparatus is alleviated, and the control accuracy of the exposure amount and the luminance uniformity is further improved.

In addition, the above-described embodiments are examples in which the present invention is applied to a projection exposure apparatus provided with a projection optical system, and in addition, a projection type projection apparatus, a proximity type projection apparatus, or a contact type exposure apparatus The present invention can be applied. As described above, the present invention is not limited to the above-described embodiments, and various configurations can be adopted without departing from the gist of the present invention.

According to the above, in a scanning type exposure apparatus using a pulsed light source as an exposure light source, in accordance with a conditional expression for obtaining a desired exposure amount and illumination uniformity accuracy, a predetermined (slit-shaped) The necessary condition for satisfactorily maintaining the uniformity of the exposure amount or the illuminance distribution in the non-scanning direction is satisfied.

When the conditions of the symmetry of the widths of the inclined portions at both sides in the scanning direction of the illuminance distribution are set, the uniformity of the exposure amount or the illuminance distribution in the scanning direction becomes good.

By disposing the field stop for illuminating the predetermined illumination area with a predetermined amount of defocus at a position optically conjugate with the exposure surface of the substrate, the illuminance in the scanning direction on the substrate of the predetermined illumination area In the case where the distribution has a trapezoidal shape, the illuminance distribution can be formed into a trapezoidal shape with a simple configuration, and the widths of both inclined portions of the trapezoidal shape of the illuminance distribution can be easily set to a desired value.

Claims (11)

The mask pattern is irradiated onto the substrate while the pulse light from the pulse light source is illuminated in a predetermined illumination area and the mask on which the transfer pattern is formed and the photosensitive substrate are scanned relative to the predetermined illumination area, An apparatus for controlling an integrated exposure dose of pulse light to a substrate within a predetermined precision in sequential exposure, And the illuminance distribution on the scanning direction of the predetermined illuminated area to one-half the length of the average value of the width of the scanning direction on the substrate in a trapezoidal shape and light intensity distribution on both sides of the inclined portion to the trapezoidal shape ΔD _12, wherein the predetermined the half width of the scanning direction on the substrate of the illumination distribution on the illumination region to D and, when the minimum number of exposure pulses required for controlling the light exposure integrator within a predetermined precision at each point on the substrate as N _min , Of the exposure amount control unit. The method according to claim 1, Assuming that the length of one half of the width in the scanning direction of the illuminance distribution on both sides of the illuminance distribution of the predetermined illumination area is ΔD ₁ and ΔD ₂ respectively and the permissible value of the illuminance unevenness in the scanning direction on the substrate is represented by [ _scan ] _max . Of the exposure amount control unit. The method according to claim 1, Wherein the illuminance distribution in the scanning direction on the substrate of the predetermined illumination area is set to a trapezoidal shape in the scanning direction on the substrate by disposing a field stop for illuminating the predetermined illumination area in a predetermined amount defocused at a position optically conjugate with the exposure surface of the substrate The exposure amount control device comprising: A scanning type exposure apparatus for scanning a substrate with a pulsed light from a pulse light source in a predetermined illumination area and moving the substrate to be exposed in the scanning direction with respect to the predetermined illumination area, The illuminance distribution in the scanning direction of the illumination region is substantially trapezoidal, Wherein an average value of a length of one half of the width of the scanning direction of both inclined portions of the trapezoidal shape illuminance distribution is? D ₁₂ , and a width of the illumination region in the scanning direction defined by 1/2 of the maximum illuminance of the illuminance distribution Wherein a value relating to a half value difference is |? X |, a tolerance for degradation of illumination uniformity in a non-scanning direction intersecting with the scanning direction is defined as [U _Bias ] _max , and an accuracy of an integrated exposure amount for the substrate is set within a predetermined range to the minimum number of exposure pulses required for controlling to N _min, Conditional expression Of the exposure apparatus. 5. The scanning-type exposure apparatus according to claim 4, wherein the symmetry of the widths of both inclined portions of the illuminance distribution is defined based on the minimum exposure pulse number _Nmin . The scanning exposure apparatus according to claim 4, wherein the symmetry of the widths of both side inclined portions of the illuminance distribution is defined based on a tolerance of deterioration of illuminance uniformity in the scanning direction. The exposure amount control apparatus according to any one of claims 1 to 3, wherein the minimum exposure pulse number _Nmin is determined based on a variation in the amount of pulsed light emitted from the pulse light source. 4. The exposure amount control apparatus according to any one of claims 1 to 3, wherein the minimum exposure pulse number _Nmim is determined based on a variation in light emission timing of the pulse light source with respect to the position of the substrate. A method of manufacturing a device using the device according to any one of claims 1 to 4. The scanning exposure apparatus according to any one of claims 4 to 6, wherein the minimum number of exposure pulses N _min is determined based on a variation in the amount of pulsed light emitted from the pulse light source. The scanning-type exposure apparatus according to any one of claims 4 to 6, wherein the minimum number of exposure pulses N _min is determined based on a variation in light emission timing of the pulse light source with respect to the position of the substrate.