JPH02102515A - Exposure period determining method - Google Patents

Abstract

PURPOSE:To determine optimum exposing period in a highly precise manner at all times even when the type and the like of the base layer is changed by a method wherein the optimum exposing period is determined by exposing the first mark for measurement of the optimum exposing period and the second mark for correction of the relations among the period of exposure, the line width of L/S and the length of the first mark. CONSTITUTION:The first mark for measurement of optimum exposing period and the second mark, to be used for correction of the relations among the exposing period, the line width of line and space (L/S) and the length of the first mark, are exposed on a photosensitive substrate, and the length of the first mark formed on a resist layer and the state of formation of the second mark, namely, the width of mark, are detected. The optimum exposing period is computed while the relations between the length of the first mark and the width of the second mark are being corrected. As a result, the lowering of the accuracy of detection and the decrease and the like of a throughput, caused by the change and the like of the type of the lower layer and the film thickness of the resist layer, can be prevented, and the optimum exposing period can be determined in a highly accurate manner.

Description

Detailed Description of the Invention [Field of Industrial Application] The present invention relates to a method for determining the optimum exposure time of an exposure apparatus used in a semiconductor integrated circuit manufacturing process, particularly in a lithography process for forming a circuit pattern on a wafer. It is related to.

[Conventional technology]

Step-and-repeat reduction projection exposure devices (steppers) have come to play a central role in the lithography process in the manufacture of semiconductor integrated circuits. In this type of stepper, a circuit pattern of a semiconductor element drawn on a mask or reticle (hereinafter referred to as a reticle) is sequentially exposed onto a wafer through a projection lens. If it is not set accurately, it will not be possible to expose a circuit pattern on the wafer with the desired line width, and it will not be possible to obtain a circuit pattern of a semiconductor element that satisfies the desired characteristics. Therefore,
For example, a reticle (for test) on which a plurality of rectangular marks having different line widths, including marks smaller than the minimum line width of a circuit pattern, are formed is sequentially exposed onto a wafer while changing the exposure time for each shot. Then, the line width of the resist image (1 to several μm thick) of the rectangular mark formed on the wafer that has been subjected to development processing, etc. is measured using the 88M length measurement method using a scanning electronic W4 microscope and a television camera (ITV). The measurement is performed using an image processing method according to the present invention, or a method of irradiating a resist image with spot light and detecting the scattered light.

The optimum exposure time is calculated from this measurement result, that is, the line width according to the exposure time. In addition, a reticle on which a wedge-shaped mark is formed in advance is sequentially exposed on the wafer while changing the exposure time, and the ratio of the bright part to the dark part is determined according to the line width of a specific pattern, for example, a circuit pattern to be transferred, so-called duty ratio. -Measure the line and space (L/S) line width with a ratio of 1:I and the predetermined mark length of the wedge-shaped mark (the length in the direction in which the line width changes continuously). The relationship between these exposure times and the L/S line width and mark length is stored. A method has also been proposed in which the mark length of a wedge-shaped mark formed on a wafer to be exposed is detected, and the optimum exposure time is calculated from this detected value and the above relationship.

[Problem to be solved by the invention]

However, in the conventional technology as mentioned above, 88M
The problem with the length measurement method is that the scanning electron microscope itself is expensive and that it takes time to measure because it requires exhaust, etc. when mounting the wafer.

In addition, with image processing methods and scattered light detection methods, as the numerical aperture (N, A,) of the projection lens increases, it becomes difficult to detect line widths when the line width of the circuit pattern to be transferred becomes sub-micron. Furthermore, there is a problem in that the amount of change in the line width of a resist image due to a change in exposure time is small, and measurement errors increase, making it impossible to accurately determine the optimum exposure time. Furthermore, in the method of detecting the mark length of a wedge-shaped mark, if a resist is used, in particular, resist IIIrg is severely reduced (film wear) due to exposure, development, etc. The above-mentioned primary relationship between the exposure time and the L/S line width and mark length changes depending on the type of base such as a metal film or oxide film, the thickness of the resist layer, etc. If the above unique relationship is applied to all wafers, the optimum exposure time cannot be calculated accurately. Therefore, in order to accurately calculate the optimum exposure time, there is a problem in that the above-mentioned unambiguous relationship must be determined for each wafer in advance depending on the type of substrate and the like.

The present invention has been made in consideration of the above points, and can respond to changes in the type of base material, the thickness of the resist layer, etc.
The purpose of this invention is to obtain an exposure time determination method that can calculate the optimal exposure time with high precision and in a short time.

[Means to solve problems]

In order to solve this problem, in the present invention, one or more linear marks whose line width changes continuously or stepwise are formed in a region corresponding to a scribe line in the pattern region Pa or attached to the pattern region Pa. The reticle R on which the first mark RMI and the 27th rectangular mark RM2 are formed is sequentially exposed onto the wafer W through the projection lens 1 while changing the exposure time T. And the laser step alignment system CL installed in the stepper
The first mark RMI and the second mark RM2 are formed on the wafer W using the LSA-based spot light spy.
Each of the resist images WMI and WM2 is sequentially scanned to detect the mark length of the resist image WMI (the dimension in the direction perpendicular to the width of the wedge) and the mark widths Dt and Db as the formation state of the resist image WM2. Configure. Furthermore, L.S.
The main controller 20, which centrally controls the operation of the entire temperer including the A system, is configured to calculate the optimum exposure time depending on the type of base material etc. based on the mark length and mark widths Dt and Db.

[For production]

According to the present invention, the first mark for measuring the optimum exposure time and the second mark for correcting the relationship between the exposure time, the L/5(7) line width, and the mark length of the first mark are placed on the photosensitive substrate. The mark length of the first mark formed on the resist layer and the formation state of the second mark, that is, the mark width are detected using an alignment system. Then, the optimum exposure time is calculated while correcting the above relationship based on the mark length of the first mark and the mark width of the second mark.

Therefore, it is possible to prevent a decrease in detection accuracy or a decrease in throughput due to changes in the type of undercoat or the thickness of the resist layer, etc., and it is possible to determine the optimum exposure time with high precision.

〔Example〕

Embodiments of the present invention will be described in detail below with reference to the drawings. FIG. 1 is a plan view showing a schematic configuration of a stepper suitable for applying the method according to the first embodiment of the present invention. The structure of this stepper is disclosed in, for example, Japanese Unexamined Patent Application Publication No. 130742/1988, which was previously filed by the applicant of the present application, and will therefore be briefly described here.

In FIG. 1, illumination light sources for exposure (not shown) are g-line and i-line.
Illumination light having a wavelength (n wavelength) that exposes resist layers such as lines is generated, and this illumination light illuminates the pattern area pa of the reticle R with uniform illuminance. Here, the reticle R has a first mark RMI for measuring the optimum exposure time, an area corresponding to the scribe line in the pattern area Pa, or a first mark RMI for measuring the optimum exposure time, a line width of the exposure time and L/S, and a first mark attached to the pattern area Pa. No. 27- for correcting the relationship between mark RMI and mark length
RM2 is formed. An example of the schematic configuration of the first mark RMI and the second mark RM2 used in this embodiment is shown in FIGS. 2(a) and 2(b). RM1 is a linear mark whose line width changes continuously or stepwise, for example, three marks formed by symmetrically combining two wedge-shaped marks in parallel at regular intervals. Further, as shown in FIG. 2(b), the 27th mark RM2 is a rectangular mark whose line width in the lateral direction is thicker than the minimum line width of the circuit pattern, for example.

Now, the projection lens 1, which is telescopic on one side (or on both sides), projects images of the circuit pattern, the first mark RMI, and the 27th mark RM2 formed in the pattern area Pa onto the wafer W on which the resist layer is formed. . In this embodiment, the optical system is non-telecentric on the reticle R side and telecentric on the wafer W side. The wafer W is located in a plane (imaging plane IM) substantially perpendicular to the optical axis AX of the projection lens 1
, an X-Y stage that moves two-dimensionally in the Y direction, and an
It is installed on the Y stage, and the optical axis direction (Z
The wafer is placed on a wafer stage 2, which includes a Z stage that moves in a wafer holder (not shown). Further, the wafer stage 2 is moved to the imaging plane 1 by the drive unit 3.
It is configured to move two-dimensionally within M and also move in the Z direction. The two-dimensional position of the wafer stage 2 is determined by a light wave interferometer (hereinafter referred to as a laser interferometer) 4.
The sensor is configured to be constantly detected with a resolution of, for example, 0.02 μm.

In addition, two sets of TTL are used as alignment systems to respectively detect the positions in the X and Y directions of the wafer mark for alignment that is printed on the chips on the wafer W, especially the diffraction grating mark consisting of 7M grating elements. A (through-the-lens) type laser step alignment (LSA) system is provided. Note that FIG. 2 shows only the LSA system that detects the position of the alignment mark in the Y direction. Now, the LS that detects this position in the Y direction
The A system is, for example, a He-Ne laser (wavelength: 633 nm)
A laser light source 5 whose light source is a laser beam with a linear polarization different from the exposure wavelength is generated, and this laser beam
The beam is expanded to a predetermined diameter by the cylindrical lens 7.
The cross section is shaped into an elongated elliptical beam. and,
The laser beam shaped in this way is reflected by a mirror 8, passes through a lens 9, a polarizing beam splinter 10, a quarter-wave plate 11 as a phase rotation member, and an objective lens 12, and then is directed to the lower surface of the reticle R by a mirror 13. It is reflected upward from the The laser beam from the mirror 13 once converges into a slit shape, and then reaches the entrance pupil 1a of the projection lens 1 via the mirror 14, which has a reflection plane parallel to the reticle R below the reticle R, and forms an elongated strip on the wafer W. A spot light SPy is formed. In addition, spot light SPy
extends in the X direction within the exposure field of the projection lens 1, is formed toward the optical axis AX of the projection lens 1, and
It is formed at a predetermined distance away from the optical axis AX on the axis. Then, when the wafer stage 2 is slightly moved in the Y direction so that the spot light SPy scans the alignment mark (diffraction grating mark) in the Y direction, in addition to specularly reflected light (0th order diffraction light), scattered light is emitted from the diffraction grating mark. or diffracted light (1
This optical information is arranged approximately conjugate with the entrance pupil 1a of the projection lens 1 through the projection lens 1, polarization beam splinter IO, etc., and has an aperture that matches the diffracted light or scattered light distribution. A spatial filter 15 is reached.

Only the diffracted light or scattered light is extracted by the spatial filter 15 and focused onto the light receiving surface of the light receiving element 17 via the condensing lens 16 . Light receiving element 1 as this photoelectric detector
7 outputs a photoelectric signal according to the intensity of the diffracted light or scattered light, and this photoelectric signal is sent to an alignment signal processing circuit (hereinafter referred to as
(called LSAC) 18. The LSAC 18 also inputs position information from the laser interferometer 4, samples photoelectric signals in synchronization with up/down pulse signals generated every 1 t (0.02 μm) of unit movement of the wafer stage 2, and performs predetermined calculation processing. This detects the scanning position of the mark in the Y direction. In addition, in order to simplify the explanation, only the LSA system that detects the position in the Y direction is shown in Figure 1, but in reality, an LS system with a similar configuration that detects the position in the X direction is shown.
Another set of A system is arranged, and in Fig. 1 mirror 14
Only the mirror 19 corresponding to the above is shown.

Now, FIG. 1 shows an irradiation optical system that supplies an imaging light beam 11 from an oblique direction with respect to the optical axis AX to form a pinhole or slit image toward the imaging surface IM of the projection lens 1. 21a, and a light receiving optical system 21b that receives the reflected light beam 7!2 of the imaging light beam xl on the surface of the wafer W. The configuration of the focus detection system 21 is disclosed in, for example, Japanese Patent Application Laid-Open No. 168112/1989, which was previously filed by the applicant of the present application, and is disclosed in the vertical direction (
The position in the Z direction) is detected, and the in-focus state of the wafer W and the projection lens 1 is detected.

In addition, in this embodiment, so that the imaging plane IM serves as the zero point reference,
It is assumed that the angle of a parallel flat glass (not shown) provided inside the light receiving optical system 21b (plane parallel) is adjusted in advance, and the focus detection system 21 is calibrated. In addition to calculating the optimum exposure time using the LSA system described above, the main controller 20 also operates the two sets of LSA systems described above.
This system controls the entire operation of the stepper including the focus detection system 21 and the like.

Next, the operation of this embodiment will be explained with reference to each drawing. In FIG. 1, the main control unit W20 closes the optical path of the exposure light,
A shutter (not shown) that opens is controlled, and the exposure time T,
That is, the first mark RM1 and the second mark RM2 are sequentially exposed on the wafer W while changing the exposure amount per shot. Then, the resist images WMI and WM2 of the first mark RMI and second mark RM2 formed on the wafer W after the development process and the like are detected using the LSA system. In this embodiment, the resist image WML is created using the LSA-based spot light SPy that detects the position of the alignment mark in the Y direction.
In order to detect WM2, the alignment direction of the resist image WMI (
It is assumed that the first mark RM1 and the second mark RM2 are transferred onto the wafer W so that both the pitch direction) and the longitudinal direction of the resist image WM2 coincide with the X direction. Therefore, first, the detection operation of a predetermined mark length L (the length in the direction in which the line width changes continuously (Y direction)) of the resist image WMI will be explained using FIGS. 3(a) and (b). . FIG. 3(a) shows the state in which the spot light SPy scans the resist image WMI, and FIG. 3(b) shows the waveform of the special photoelectric signal S1.

Now, the main controller 20 slightly moves the wafer stage 2 in the Y direction so that the spot light SPy relatively scans the resist image WMI in the Y direction (FIG. 3(a)), and
The light receiving element 17 photoelectrically detects the scattered light (or diffracted light) generated from the resist image WMI, and outputs a photoelectric signal S1 according to the intensity of the scattered light (or diffracted light) as shown in FIG. 3(b).
Output to SACI8. LSAC18 uses this photoelectric signal S
1 is sampled in synchronization with the position signal of the laser interferometer 4, and the photoelectric signal S1 is processed at a slice level set by a predetermined calculation to calculate the mark length of the resist image WM1. The control device 20 repeatedly performs the same operation as described above, calculates the mark length of the resist image WMI for each shot on the wafer W, and stores these calculated values. Note that the mark length L and exposure time T obtained as a result of this
The relationship with is shown in FIG. 4 by a dashed line.

Next, the formation state of the resist image WM2 formed in an arbitrary region t+il on the wafer W, that is, the predetermined mark width (
The width in the lateral direction) is detected using an LSA system. Therefore,
This detection operation will be explained using FIGS. 5(a) to 5(d). As shown in FIG. 5(a), the main controller 20 uses the focus detection system 21 to detect the height position of the upper end surface of the resist image WM2, and this upper end surface is the imaging plane IM of the projection lens 1. The drive unit 3 is used to move the wafer stage 2 in the Z direction so that the wafer stages 2 and 2 are substantially aligned. After the focusing of the upper end surface of the resist image WM2 is completed, the driving unit 3 is used so that the spot light SPy relatively scans the resist image WM2 in the Y direction as shown in FIG. 5(b). wafer stage 2
Move slightly in the Y direction. At this time, as the spot light SPy scans the resist image WM2, scattered light is generated in the spaces facing the edges E1 and E2, and the spot light S
When the center of the laser beam forming the py coincides with the edges E1 and E2, the intensity of this scattered light reaches its peak. The scattered light generated from this resist image WM2 is photoelectrically detected by the light receiving element 17, and the light receiving element 17 sends a photoelectric signal S2 having two peaks corresponding to the intensity of the scattered light as shown in FIG. 5(C) to the LSAC 18. Output. And LSACl
8 detects the Y coordinate values Yl and Y2 of the position where the photoelectric signal S2 reaches its peak based on this photoelectric signal S2 and the position signal of the laser interferometer 4. Furthermore, this Y coordinate value Yl,
The mark width Dt at the upper end surface of the resist image WM2 shown in FIG. 5(a) is calculated from Y2, and this mark width Dt is stored. Next, as shown in FIG. 5(d), the focus detection system 21
is used to detect the height position of the surface of the wafer W, and the drive unit 3 is used to move the wafer so that the surface of the wafer W, that is, the lower end of the resist image WM2 substantially coincides with the imaging plane IM of the projection lens 1. Move stage 2 in the Z direction. After this focusing is completed, the main controller f20 photoelectrically detects the scattered light generated from the resist image WM2 using the spot light SPy in the same manner as the detection of the mark width Dt described above.

Then, the mark width Db at the lower end of the resist image WM2 shown in FIG. 5(b) is calculated, and this mark width Db is stored. Note that the detection of the mark width DtSDb of the resist image WM2 does not need to be performed in all shot areas transferred onto the wafer W with different exposure times, but may be performed in at least one shot area. However, in order to improve accuracy, it is desirable to detect mark widths Dt and Db in a plurality of shot areas and use a value obtained by averaging these calculated values.

Here, in this example, a wafer W on which an arbitrary resist layer has been formed with a predetermined thickness by exposure, development, etc. is used, and L/R as shown by a dashed line in FIG. The relationship between the line width P of S and the mark length is determined and this information is input to the main controller 20. Furthermore, L
The relationship between the line width P and mark length of /S is -a, P=A
It is expressed by the linear equation L+B. However, although the coefficient A is a substantially constant value regardless of the type of base, etc., the correction constant B varies depending on the type of base. Therefore, only the coefficient A is manually input to the main control device 20 as the above information. 720 is a correction constant B (B
=KxDb/Dt.

However, K is a constant and is uniquely determined by the type of base material, etc.), and the relationship between the line width P of L/S and the mark length (P -AL+KxDb/
Find Dt). From this, by detecting the mark length of the resist image WM2 formed on the wafer W, the line width of the reticle pattern formed on the wafer W can be determined.

Next, the main side? 11 Based on the mark widths Di and Db, the apparatus 20 corrects the relationship between the mark length shown by the dashed line in FIG. Seek a relationship according to. Therefore, the main controller 20 controls the mark width Dt.

Based on Db, a correction value L'(L'=L+B') of the mark length detected as described above is calculated.

Here, the correction constant B' is B' = 'xDb/Dt(
However, K' is a constant and is expressed as '=/A).

Then, using the mark length L′ corrected in this way,
The relationship between the mark length L and the exposure time T is calculated depending on the type of base, etc. as shown by the solid line in FIG. Then, the main controller 20 controls the line width P of L/S shown by the solid line in FIG.
From the relationship between and the mark length, the desired L/S line width P,
That is, the mark length when the reticle pattern is formed with the optimum line width is found in the 0th order, and this mark length and the mark length shown by the solid line in Fig. 4 are...! : Calculate the optimum exposure time of the stepper based on the relationship with n light time T. As a result, the reticle pattern is formed on the wafer W with a predetermined line width.
The optimum exposure time for forming the film is determined.

Note that in this embodiment, as the first mark RMI, three linear marks in which two wedge-shaped marks were symmetrically combined were arranged in parallel at a constant pitch.
The first mark used in the present invention is not limited to the mark shown in FIG. Any mark that has a plurality of marks may be used.For example, a pattern in which wedge-shaped marks are arranged radially at predetermined angular intervals (so-called Siemens star pattern) may be used.Similarly, the second mark RM2 may be Figure 2 (b
) is not limited to the rectangular mark shown in
As shown in figures (a) and (d), the mark width Dt at the upper end
Any mark may be used as long as it forms a resist image in which the mark width Db and the mark width Db at the lower end are different.

Further, in the above embodiment, a TTL type LSA system is used as the pattern detection system, and the mark length and mark width Dt,
Db was detected, but within a certain bandwidth (about 200 nm)
It is also possible to use an off-axis wafer alignment system installed separately with a projection lens 6 that uses light with a broad wavelength distribution of good. The numerical aperture (N, A, = about 0.2 to 0.3) of these devices is TT
It is larger than the L-type LSA system and can detect mark width DtSDb, etc. with higher accuracy.

〔Effect of the invention〕

As described above, according to the present invention, the first mark for measuring the optimum exposure time and the second mark for correcting the relationship between the exposure time, the line width of L/S, and the mark length of the first mark are attached to the photosensitive substrate. The optimum exposure time is determined based on the predetermined mark length of the first mark formed on the resist layer and the predetermined mark width of the second mark. , the optimal exposure time can always be determined with high precision.

In addition, compared to the amount of line width change due to exposure time change in circuit patterns, etc., the sensitivity of length change to exposure time change is several tens of times higher. The mark length of one mark is detected using an alignment-based scattered light detection method to calculate the optimal exposure time, so even if the line width of the circuit pattern to be transferred is sub-micron, it can be performed with high precision and at high speed. The optimal exposure time can be determined.

[Brief explanation of the drawing]

FIG. 1 is a plan view showing a schematic configuration of a stepper suitable for applying the method according to the first embodiment of the present invention, and FIGS. 2(a) and (b) are one embodiment of the present invention. An example of a schematic configuration of a first mark for measuring the optimum exposure time used in and a second mark for correcting the relationship between the exposure time, the line width of L/S, and the predetermined mark length of the first mark is shown. Figure 3(a),
(b) is a diagram for explaining the operation of detecting the mark length of the first mark formed on the wafer, and Figure 4 shows the line width of L/S and the mark length of the first mark with exposure time as a parameter. 5(a) to 5(d) are diagrams illustrating the operation of detecting the predetermined mark width of the second mark formed on the substrate, and FIG. FIG. 3 is a diagram showing the relationship between mark length and exposure time. [Explanation of symbols of main parts] 1... Projection lens, 2... Wafer stage, 3...
- Drive unit, 4... Laser interferometer, 5-17... Laser step alignment system, 18... Laser step alignment system processing circuit, 20... Main side j1
Device, 21a, 21b...Focus detection system, spy...
Spot light, AX...optical axis, R...reticle, area, W...wafer. Pa...pattern

Claims (4)

[Claims] (1) In a method for determining an appropriate exposure time when transferring a pattern formed on a mask onto a photosensitive substrate with a predetermined line width using an exposure device, at least a first step of exposing a resist layer on the photosensitive substrate to a linear first mark whose line width changes continuously or stepwise, and a second mark whose shape is different from the first mark; a second step of detecting a predetermined mark length of the resist image of the first mark formed on the layer and a formation state of the resist image of the second mark by a pattern detection system;
Based on the mark length of the first mark detected in the second step and the formation state of the second mark, calculate an appropriate exposure time for transferring the pattern onto the photosensitive substrate with the predetermined line width. An exposure time determining method, comprising: a third step of determining an exposure time. (2) The exposure time determining method according to claim 1, wherein the first mark is a wedge-shaped mark and the second mark is a rectangular mark. (3) The pattern detection means, as the formation state,
3. The exposure time determining method according to claim 1, wherein the mark width is detected at each of an upper end and a lower end in the thickness direction of the resist image of the second mark having a predetermined thickness. (4) The exposure time determining method according to any one of claims 1 to 3, wherein the pattern detection means is an alignment system of the exposure apparatus that aligns the mask and the photosensitive substrate.