JPS62291919A - Observation apparatus - Google Patents

Abstract

PURPOSE:To observe only information of a wafer except the influence of an interference with a resist by obliquely emitting a wafer by an illumination system provided between a projecting optical system and the wafer. CONSTITUTION:A contraction lens 3 projects to expose a pattern 2 on a reticle 1 on a wafer 4. An alignment mark 6 for the part 5 of a wafer to be printed from this time is obliquely illuminated by an illumination system (a laser light source 10, a mirror 11 and a lens 12) existing between the wafer 4 and the lens 3. Thus, the component of the illuminated light reflected on the surface of the resist is not incident to a detecting optical system, but escaped out. Accordingly, only the reflected light from the wafer substrate 4 can be collected by the detecting optical system. Here, the optical system means the whole observing optical system of the wafer 4 through the lens 3, an alignment scope optical system 7 to photoelectric converting means 8.

Description

Detailed Description of the Invention 3. Detailed Description of the Invention [Field of Industrial Application] The present invention relates to an observation device, and in particular, an optical system for observing an object having a transparent thin film on its surface and detecting information thereof. Regarding the system. A representative example of this type of optical system is an exposure device such as a stepper that aligns a wafer coated with photoresist with respect to a reticle.

[Prior Art and Problems to be Solved by the Invention] The two basic performances of this type of exposure apparatus are resolution and overlay accuracy. Regarding resolution, handling is very simple. This is because there are only a few parameters that determine resolving power, and in devices called steppers, the resolving power of the optical system can be easily estimated by knowing the wavelength used and the numerical aperture (NA) of the projection lens. Further, even in the case of X-ray exposure, there are only limited parameters such as penumbra blur due to the size of the light source.

Since the realization of single-transistor memory cells, advances in lithography, that is, fine line width printing technology, and advances in process technologies such as etching have played a role in achieving higher integration of semiconductors. In terms of resolution, optical systems have been steadily improving, as can be seen by tracing the history of stepper lenses. Optical systems have broken the 1 μm barrier, and lenses compatible with the submicron era are being announced one after another.

On the other hand, in the process, new ideas such as trenching methods and the combination of lower and higher height differences are being realized using three-dimensional IC concepts. The greatest point of contact between advances in resolution on the exposure equipment side and advances on the process side is the overlapping of patterns in each process. In this sense, it can be said that overlay accuracy is becoming increasingly important in exposure equipment.

It is difficult to express overlay accuracy using simple parameters such as resolution. This shows the diversity of the Unihachi process, but on the other hand, the configuration of the alignment system for superimposition is multi-f! ! ! This may be due to the fact that it is used frequently. What makes wafer process factors even more complex is that this problem is not limited to just one substrate, but also needs to be discussed in conjunction with the photoresist coated on the wafer. One of the obvious trends in current semiconductor technology is the trend toward three-dimensional IC configurations. Among these, a high level difference on the surface of the sea urchin is unavoidable, but this high level difference has a clear negative effect on the coating condition of the photoresist.

Furthermore, wafers are becoming increasingly larger from 6 inches to 8 inches and even 10 inches. When applying photoresist to a large-diameter wafer using a spin method, it is obvious that the state of resist application differs between the center and the periphery, and this difference becomes more pronounced as the level difference on the surface of the sea urchin increases. It is also clear that In fact, it is known that the alignment state changes due to the influence of resist coating, and research is being conducted on how to achieve uniform coating.

Another thing to keep in mind when it comes to photoresists is the trend toward multilayering in the submicron era. Since means for improving resolution, such as a multilayer resist process and CEL, are inevitably employed in several processes, countermeasures for this are also necessary. It can be said that exposure equipment is being forced to deal with these new wafer processes in the field of stacking.

On the other hand, the diversity of alignment systems is a testament to the flexibility and difficulty of system configuration.

Currently, no two alignment systems that have been proposed and implemented are the same, and each system has its own advantages and disadvantages. For example, one example is ``Exposure Apparatus'' published in Japanese Patent Application Laid-Open No. 58-25638 filed by the present applicant. This system uses a projection optical system that is telecentric to both the reticle and the wafer.
This is one of the excellent configuration examples that realize the concept of nAxis. Although the projection lens has aberrations corrected for the g-line (436 nm), it is designed to exhibit similar performance at the wavelength of the He-Cd laser (442 nm). In one embodiment disclosed in this patent application, a laser beam scanning method using a He-Cd laser is adopted as an alignment signal detection method, and as a result, TTL on Ax
In other words, it is possible to start an immediate exposure operation in an aligned state. TTL on Axis
In the sense that the only error factor for the exposure apparatus is the alignment signal detection error, the system has the configuration with the least systematic error factors, and is close to an ideal system. The only drawback to this system is that it is sensitive to processes that absorb wavelengths near the exposure wavelength, such as multilayer resists.

On the other hand, wavelengths other than the exposure wavelength, specifically e
line (546nm) or He-Ne laser (633nm)
), many system configuration examples using longer wavelengths have also been proposed. This system has the advantage of being robust against absorption-type processes such as multilayer resist because it uses a wavelength longer than the exposure wavelength. However, normally, the image height for alignment is fixed relative to the projection lens due to various chromatic aberrations of the projection lens, and the error factor of moving the urchin to the exposure position after alignment detection is introduced. become. For this reason, alignment systems using light other than the exposure wavelength are inevitably TTL.
This results in an offAxis system.

However, in recent years, requirements for overlay accuracy have become increasingly strict, and even detection errors in alignment signals, which are the cause of errors in ideal systems, as shown in Japanese Patent Application Laid-open No. 58-25638, have even become a problem. ing. The present invention is characterized in that the alignment accuracy is improved by analyzing the detection error component of the alignment signal based on the conventional example and removing the error factor.

According to the inventors' analysis of the detection error component of the alignment signal, it has been found that the majority of the error component is due to problems in coating the photoresist. There are various error factors caused by the photoresist, but the following two factors are thought to be the largest among them.

The first is the interference effect between the light reflected from the surface of the resist and the light that AA-performs the resist and returns after hitting the Urchin eight substrate. In particular, as described above, the photoresist is not necessarily coated uniformly within the wafer, and the coating state often differs between the center and the periphery. The wafer substrate itself also has problems with internal wafer uniformity due to etching, sputtering, etc. Therefore, the structure of the alignment mark for each shot on the wafer differs depending on the location when considering the resist coating.
Therefore, the interference effects are also different. This interference seems to be the most significant cause of alignment errors due to the influence of resist coating.

The second factor is multiple reflection. The resist has the character of a leading wave path. Therefore, a portion of the light that is reflected by the substrate to the sea urchin is reflected at the interface between the resist and the air, returns to the sea urchin, and is reflected again.

The higher the reflectance of the substrate, the more pronounced this shadowing becomes, and this multiple reflected light eventually causes interference and becomes a factor that deteriorates alignment accuracy.

Other resist factors can be considered, such as image deviation due to refraction, but these are only secondary factors, and alignment is achieved by eliminating the two factors mentioned above, especially the first interference effect. The analysis results confirmed that this greatly contributes to improving the accuracy of.

Accordingly, it is an object of the present invention to provide a system that reduces interference effects and achieves higher alignment detection accuracy.

[Means for Solving the Problems] In order to solve the problems described above, the observation device of the present invention projects a pattern on a first object onto a second object via an imaging optical system. In the transfer apparatus, an observation optical system is arranged which is coupled to the projection optical system and observes the second object, and the observation optical system is arranged between the projection lens and the second object through the projection lens. It is characterized by detecting light from an illumination optical system that provides illumination.

In addition, features of the present invention, such as various embodiments of the present invention, are described in the following Examples.

[Example] An example of the present invention is shown in FIG. FIG. 1 shows an example in which the present invention is applied to a stepper, that is, a reduction projection exposure apparatus.

In the figure, a reduction lens 3 serves to project and expose a pattern 2 on a reticle 1 onto a wafer 4. The reduction lens 3 is used to reduce the irregularities of the wafer, which is the object to be printed,
Normally, the wafer side is telecentric so that distortion and magnification do not change due to changes in focus due to measurement drive errors of the autofocus system placed between the lens 3 and the wafer 4. It is. Note that the location corresponding to 2 on the wafer corresponds to 5 in FIG.

The most important feature of the present invention is the illumination system placed below the projection lens 3. In Figure 1, what should be burned from now on,
The alignment mark 6 for the part 5 of the wafer is in this case obliquely illuminated by an illumination system located between the wafer 4 and the projection lens 3 .

This is the most important part of the invention.

As already mentioned, the largest cause of alignment signal detection errors is the interference between the surface reflection of the resist and the reflected light from the Urchin substrate. There are several ways to eliminate this shadow, but the most fundamental solution is to eliminate one of them. When we observed the state of the resist coating using a SEM or an interference microscope, we found that even on a wafer with a very large step structure, the slope of the surface of the resist coated on top of the wafer is only about 5 degrees at most, and it is not steeper than that. It turned out that the slope did not exist. Due to step coverage issues, it is common practice to apply a resist thicker than the large step difference, and the result is approximately 5 degrees. Therefore, the present invention is characterized by adding the following conditions to the illumination light as shown in FIG. That is, (maximum angle A+lO° that the detection optical system has on the wafer)≦(
Angle of incidence of illumination light on the wafer B) (1) If this is done, the component of the illumination light reflected on the resist surface will not enter the detection optical system and will escape to the outside (see Figure 2). ).

Therefore, only the reflected light from the sea urchin eight substrate can be captured by the detection optical system. The detection optical system in FIG. 1 refers to the entire wafer observation optical system from the projection lens 3 and the alignment scope optical system 7 to the photoelectric conversion means 8 such as a CCD. If the exposure wavelength and observation wavelength are different, a correction optical system such as 9 may be installed to enable simultaneous observation of the reticle and wafer, or it may be possible to observe only the wafer as shown in Figure 3. Therefore, an alignment scope optical system may be placed between the reticle and the projection lens. In the case of FIG. 3, the reticle is independently aligned by another alignment scope system, and reference value calibration is performed so that the wafer-based alignment reference and the reticle-based alignment reference have a predetermined relationship.

The maximum angle A that the detection optical system has on the wafer matches the value determined by the NA of the projection lens, unless a special aperture is inserted on the alignment scope side to regulate the NA of the projection lens. In other words, if the lens has an NA of 0.35, A =
Since sin −' 0.35 = 20.49°, the incident angle B of the illumination system must be greater than or equal to 30.49°. Also, if the alignment scope has an aperture to regulate the NA of the projection lens, the value of A is N.
It will be smaller than the value determined by A. All TTL methods proposed for steppers so far have been methods in which incident light is illuminated from above through the alignment scope side. In contrast, the present invention is a system in which the wafer is illuminated from outside the projection lens, and by setting a regulation value of 10 degrees, the resist surface reflection has been successfully eliminated.

Being able to disperse and extract the scattered and diffracted light from the wafer substrate at the bottom of the resist means that it directly leads to improved alignment accuracy.

In the present invention, as described above, the scattered and diffracted light from the Urchin 8 substrate itself is taken, so it is advantageous to enter the illumination light as P-polarized light so as to reduce reflection loss on the resist surface. P (
Since m-light has less surface reflection than S-polarized light, more light reaches the substrate. Furthermore, it is advantageous to set the incident angle at Brewster's angle because surface reflection is minimized.

It is also possible to determine other degrees of freedom, such as the shape of the mark, to meet this condition.

In FIG. 1, 10 is a laser, and illumination light from the laser 10 illuminates the alignment mark 6 on the wafer 4 via a mirror 11 and a lens 12.

Another feature of the illumination system shown in FIG. 1 is that light is incident from a predetermined direction corresponding to a pattern direction known in advance. For this reason, it is desirable that the illumination light has a directionality such that it is incident from a direction perpendicular to the edge of the mark to be detected.

FIG. 4 shows how to enter illumination light when dot-shaped marks are used. Here, it is assumed that the center position C in the width direction of a mark composed of dots is to be determined as an alignment signal. In the example of FIG. 4(A), the illumination light is incident from a direction that illuminates the edges P and P', which are directions indicating the width of the grating. The direction of incidence is indicated by an arrow D'. In this case, the image actually observed on the image plane is as shown in Fig. 5 (A), and by receiving this image with a COD or scanning it with a slit-shaped aperture,
The center position C can be determined. P, P which is edge
The way the light is illuminated is in the direction of incidence and corresponds to D'. however,
The incident light from D does not always make P shine; P' may be brighter than P depending on the amount of step on the substrate and the state of resist coating. If illumination light is input from two directions, D and D', the asymmetry of P and P' will be alleviated, but the symmetry of both is not particularly necessary when calculating C, as shown in Figure 5 (A). I understand more. Therefore, when considering the system configuration, D' does not necessarily have to be inserted symmetrically from two directions; it may be inserted only from one side, or even if it is inserted from both directions, it may be arranged asymmetrically, that is, as shown in Figure 4 (
Even if the values of α and α' in A) are not made the same, the position of C can be detected with sufficient accuracy.

In the example of FIG. 4(B), illumination light is incident from a direction perpendicular to the direction of illumination light in FIG. 4(A). In this case, the image observed is as shown in FIG. 5(B), with the edges shining in the opposite direction to that in FIG. 5(A).

The direction in which the edges shown in (A) and (B) are illuminated depends on the mark signal processing method used.

Further, although a row of dots has been adopted as a mark in the explanation because it has edges in two directions, it is of course not necessary to be limited to this example.

By using the above illumination method, it is possible to observe a bright image with a good S/N ratio even with monochromatic light without the influence of surface reflection. The illuminated mark portion 6 of the wafer is once imaged at the position of the observation optical system 1 via the projection lens 3 and the correction optical system 9 in FIG. It then enters the alignment scope section and is captured as an image at position 8. The eight-line method for the exposure operation after detecting the alignment state includes the configuration of a known system, such as feeding a predetermined amount to the exposure position or retracting the correction optical system 9 out of the exposure beam. It is possible to take

FIG. 3 is an example of a system in which the wafer signal in the system of FIG. 1 is taken between the reticle and the projection lens. The reticle 1 is separately aligned using an optical system for reticle alignment (not shown). A laser 10 is used for illumination, but in this example, the directivity of the laser beam is utilized to directly direct the beam to the alignment mark 6 on the wafer. The wafer signal is guided through a projection lens 3 to an observation optical system for up to 20 sea urchins. In the figure, 20 is an objective lens, 21 is an erector lens, and a galvano mirror 23 is placed after the erector. The image of wafer 6 is a galvano mirror.
23, and the scanned image is guided to a photomultiplier 25 via a slit 24 placed on the image plane, and an auto-alignment signal is detected accordingly. If the wavelengths of the dark-field illumination light and the exposure light are different, the correction optical system corresponding to 9 in Figure 1 should be placed in front of 20 or 20.
It may be necessary to insert it into the following optical system.

Figure 6 is a modification of Figure 3, but in addition to 10, there is another 10.
It is characterized by the fact that laser beams with different wavelengths than 10 are incident. 10, for example, He −N e
Laser (633 nm), Ar as 10'. An ion laser (515 nm) or a green He-Ne laser (543r+m) can be used. laser 10'
was added to the laser 10 because there are cases where the scattered light from the He--Ne laser with a wavelength of 633 nm becomes extremely weak depending on the stepped structure of the square. The conditions for weakening are closely tied to the wavelength of the laser, so if lasers of different wavelengths are incident, their output characteristics will become complementary, allowing more stable measurements. In this case, the wafer detection optical systems are also provided separately, taking into account the difference in chromatic aberration of the projection lens 3 depending on the wavelength. Furthermore, it is efficient and advantageous to use a dichroic film for the beam splitter 13 that combines two wavelengths.

The conceptual diagram of the present invention, which provides illumination from the outside of the projection lens 3 in such a way that the surface reflection of the resist does not enter the detection optical system, is shown in Section 1.
This is clearly shown in Figure 3.6.

Conventionally, an autofocus function has been installed between the projection lens 3 and the wafer 4 since the stepper era. As shown in the present invention, no positive significance was given to the illumination method, and thus there had been no attempt to make the two coexist. However, as is the gist of the present invention, the condition that satisfies conditional expression (1) can be given positive significance to the detection of wafer information, so the coexistence of both has been considered.
It turns out that mechanically the two can coexist.

Further, if autofocus is also performed by TTL, the autofocus mechanism between the projection lens wafers can be removed, so this illumination method can be easily implemented between the projection lens wafers. Hereinafter, embodiments of the present invention based on specific implementations will be described.

One of the most important problems with the dark field method of the present invention is the amount of light. Therefore, the present invention has consistently featured a combination of dark field illumination using a laser and a projection lens. The reason for this is the chromatic aberration of the projection exposure system 3 when an observation wavelength different from the exposure wavelength is used. Generally, the projection lens 3 is designed to exhibit the best performance at the exposure wavelength, and its performance at the observation wavelength is often left unresolved. For example, a lens for the g-line (436 nm) is designed taking into consideration the spectral width of the g-line, and exhibits changes in wavelength VS and focus such that they touch at the g-line as shown in FIG. For this reason, if the wavelength differs from the observation wavelength, the coefficient of change due to the color of the focus becomes extremely small. For example, it is conceivable to use mercury's e-line (546 nm) as a dark-field illumination light source, but if you consider the extent to which the e-line's spectrum spreads, this could result in a focus change of nearly 10 μm. . This chromatic aberration is the first
There is no way that the alignment optical system shown in the figure can correct this. One possible solution would be to insert a narrow band pass filter, but in that case the drop in light intensity would be unignorable. On the other hand, if the spectrum is not narrowed, there is a problem of a decrease in contrast due to out of focus within the spectrum. The reason why a laser is used for dark field illumination as in the present invention is that the spectral width of the laser is so small that it can be almost ignored, so it can be freed from such restrictions of the projection lens. In addition to changes in focus, various aberrations occur due to color, but these can also be corrected by utilizing the monochromatic nature of the laser. 9 in Figure 1 corresponds to this. Regarding these amendments, etc., please refer to the patent application "Observation Device" filed by the applicant on May 30, 1986 (
A detailed explanation is given in the following article (inventor: Dogi Suzuki).

Since the light source itself is high-intensity, lasers are very advantageous in terms of light quantity.Also, as a countermeasure against signal loss due to interference due to monochromaticity, multiple lasers are stacked one on top of the other as shown in Figure 6 to generate multi-wavelength signals. You can also view it at Here, regarding the incident angle, the mark has a pitch as shown in Figure 4, and the fourth
When illuminating from a direction as shown in FIG. 3A, the angle of incidence is preferably such that the first-order diffracted light exits parallel to the optical axis of the projection lens 3. Since something with such a pitch is a type of grating, the well-known diffraction theory of gratings holds true. Generally (1)
In the case of conditions that satisfy the formula, what is detected by the detection system is often fairly high-order light such as third-order light or fourth-order light. Normally, the higher the order of diffracted light, the smaller the amount of light, so if the direction of diffraction is determined in advance, it is preferable to make the diffracted light of some order parallel to the optical axis of the projection lens 3. This is related to the fact that the projection lens 3 has a telecentric configuration on the wafer side so that the distortion does not change due to the unevenness of the wafer or the accuracy of autofocus. In other words, in the example shown in Figure 1.3.6, the position observed by the detection optical system is off the optical axis of the projection optical system 3, but due to telecentricity, the chief ray at this observation point also It is parallel to the optical axis and perpendicular to the wafer surface.

In the case of a normal silicon wafer, a signal with a sufficiently good S/N ratio can be obtained by such a simple illumination method.

However, in order to further increase the S/N ratio and detection rate, it is also possible to use an illumination method as shown in FIG.

FIG. 8(A) shows an example in which the illumination light emitted from the laser 10 is vibrated by a galvanometer mirror 31 to illuminate the alignment mark 6 on the wafer. In this example, the reflection point of the vibrating mirror is placed at a position that almost resonates with mark 6,
The angle of incidence on the mark 6 is changed according to the rotation of the mirror. A change in the incident angle changes the scattering diffraction conditions at the edge of the alignment mark, and therefore the signal output changes. This change in incidence angle is particularly effective when the step amount of the edge matches the wavelength of the laser and the signals cancel each other out. The change in the incident angle can be realized not only by a reflective type as shown in FIG. 8(A) but also by a transmissive type as shown in FIG. 8(B), and various modifications thereof are possible. When changing the incident angle over time, synchronization with the detection system may be achieved. For example, in the example shown in Figure 1, the detection is by CCD, but in this case, it is necessary to set the cycle to the readout timing of the CCD, or to make the change in the angle of incidence very fast compared to the readout. It is necessary to ensure that there is no impact.

Furthermore, as another means, it is of course possible to control the change in the incident angle so as to stop it at the most advantageous position for signal output.

The embodiment shown in FIG. 9 is an example in which illumination is performed using a fiber without directly using a mirror or the like. Due to space issues between the projection lens and the sea urchin, it is difficult to route the laser beam, so it may be better to use a fiber as the illumination means. FIG. 9(A) is an example of Koehler illumination using the fiber exit 35 as a secondary light source. Depending on the distance, the lens 12 is not necessarily required. In this embodiment using a fiber bundle, there is a disadvantage due to the use of fibers, that is, speckle interference of laser light occurs on the wafer surface. In order to eliminate the influence of speckle, in this example, the laser beam entering the fiber bundle is
1 and a motor 32 to perform temporal integration, and as a result, speckles on the wafer surface are removed. The galvanometer mirror can be replaced with a rotating polygon mirror or an AO element, but the periodic relationship with the detection system is the same as that explained with reference to FIG.

FIG. 9B shows an example of critical illumination, in which the exit 35 of the fiber is imaged onto the mark 6 on the wafer surface by the lens 12 as a secondary light source. In such an example, non-uniform illumination due to the core of the fiber may appear on speckles.

For this reason, in FIG. 9(B), a galvanometer mirror is placed in the optical path midway for scanning to remove uneven illumination. The period relationship with the detection system is the same as before.

Although the principle of the preferred embodiment has been mainly explained using two-dimensional cross-sectional views from FIG. 7 to FIG. 9, in reality, it is necessary to further consider the correlation with the mark for the illumination system on the apparatus side. For example, in the example shown in FIG. 9, the auto-alignment mark 6 is a line extending in a direction perpendicular to the paper surface. It is fine if the alignment mark is composed only of edges in this direction, but if there is a line perpendicular to this within the mark, that line will not be detected by the detection system at all (Figures 4 and 5). reference). For example, consider a case where the auto-alignment mark 6 is made up of four rectangles as shown in FIG. 10(A). For example, if these four lines are observed in the embodiment shown in FIG. 1, they will be as shown in FIG. 10(B), and alignment is achieved by sandwiching the reticle marks indicated by dotted lines.

The marks in Fig. 10 are the ones shown in Fig. 1'L (A) because of their functions.
) is divided into two areas U and L separated by a line 40. When light is applied to each region from the direction indicated by the arrow in the figure, the edges for alignment appear to shine as shown in FIG. 11(B). At this time, there is a method in which light is incident from four directions at once, but if the target is a wafer with a rough surface, if the vertical light necessary for the U area also irradiates the L area, In some areas, it may become background noise and reduce the S/N ratio.

Conversely, the same thing will happen if the light in the left and right direction required for the L area mixes into the U area. Therefore, a preferred embodiment of the present invention is characterized in that, as shown in FIG. 12, a field stop is provided in the illumination system in accordance with the area where the mark is to be illuminated to limit the irradiation area. In the figure, reference numeral 41 denotes an illumination area limiting slit, and this slit is projected onto the wafer mark through the imaging lens 12, thereby limiting the illumination area.

This figure shows an illumination method using a fiber of the type shown in FIG. 9(A).

In order to more precisely limit the illumination area, it is convenient to tilt the slit with respect to the optical axis so as to satisfy the tilting conditions, as schematically shown in FIG. 13, although it is difficult to see in FIG.

Further, FIG. 14 shows a mark suitable for entering the U area with the direction of incidence perpendicular to the direction of FIG. 11.

The slit 41 provided within the illumination system also has other benefits. Since the feature of the present invention is that surface reflection of the resist is removed, the wavelength is not limited to non-exposure light, and the same applies to exposure light. However, if the exposure light is applied without the slit 41, the actual element mark near the alignment mark will be exposed. Therefore, when using light that exposes the resist, this problem can be solved by installing a field stop in the illumination system that illuminates only the desired portion of the alignment mark.

In addition, although the field of view limiting slit 41 was used in FIG. 12,
Even with a method that simply switches the lighting direction, for example, the third
When using a scanning system like the one shown in the figure, it is also possible to improve the S/N ratio similar to a field stop by manually turning the four fibers on and off using an AO element, etc. in synchronization with the scanning. .

Various modifications between the projection lens 3 and the wafer 4 have been described above, but in all cases, the irradiation angle satisfies formula (1), and the non-reflection of the resist can be eliminated. , the wafer signal is captured correctly and alignment accuracy is significantly improved.

On the other hand, the detection system, unless the wafer has a rough surface or a particularly high step, may require the insertion of a correction optical system, etc. It is enough to connect them. However, in the case of a rough surface, it is impossible to extract only the edge signal from the light scattered in all directions by the rough surface, and in the case of a high level difference, it is impossible to ignore the wraparound of light due to multiple reflections.

For example, FIG. 15 shows the light amount distribution at the pupil position of the detection system objective lens, and FIG. 15A shows a rough surface that spreads uniformly over the entire pupil. On the other hand, FIG. 4(B) shows the diffracted light distribution from the alignment mark when the dot pattern shown in FIG. 4 is irradiated with light as shown in FIG. 4(A). In this case, three high-order diffracted lights enter the pupil. Therefore,
When a spatial frequency filter as shown in FIG. 15(C) is inserted into the objective lens, the S/N ratio is improved and a signal of a desired pattern can be extracted. This filter can have various shapes depending on the mark used. On the other hand, creating a filter corresponding to the alignment mark in this way can also reduce the influence of multiple reflections.

[Effects of the Invention] As explained above, according to the observation device according to the present invention,
By flexibly taking the measures described above for various wafers using the illumination system and detection system, it has become possible to observe only the wafer information without the influence of interference with the resist. This method is a very effective means not only for auto-alignment but also for accurately extracting information on a wafer coated with resist, such as auto-focusing.

[Brief explanation of the drawing]

FIG. 1 is a schematic configuration diagram of an observation device according to an embodiment of the present invention, FIG. 2 is a diagram showing illumination light to a wafer and its reflected components on the resist surface, etc., and FIG. 3 is a diagram showing a reticle and the like. A diagram showing an example in which an alignment scope optical system is arranged between the projection lens, FIG. 4 is a diagram showing the irradiation direction of illumination light when dot-shaped marks are used, and FIG. Figure 6 shows an example of an image actually observed on the image plane due to illumination. Figure 6 shows an example where laser beams of two different wavelengths are incident. Figure 7 shows the illumination light on the wafer and the scattering at the mark. , a diagram showing diffracted light, FIG. 8 is a diagram showing an example of vibrating illumination light and illuminating the alignment mark on the wafer, FIG. 9 is a diagram showing an example of illuminating using a fiber, and FIG. 11 is a diagram showing an example of an alignment mark, FIG. 11 is a diagram showing the direction of illumination light with respect to the alignment mark in FIG. Fig. 13 is a diagram showing an example in which the slit is tilted with respect to the optical axis so as to satisfy the tilting conditions, and Fig. 14 is a mark suitable for illuminating from a direction perpendicular to the direction shown in Fig. 11. FIG. 15 is a diagram showing the light amount distribution at the pupil position of the detection system objective lens, etc. FIG. 16 is a graph showing wavelength vs. focus change. 1 reticle, 4: wafer, 6: alignment mark, 7: alignment scope optical system, 8: photoelectric conversion means, 9: correction optical system, 10, lO': laser light source, 12: imaging lens, 23.32: galvano mirror -, 35: Fiber exit, 41: Irradiation area limiting slit. Patent Applicant Canon Co., Ltd. Agent Patent Attorney Tatsuo Ito Agent Patent Attorney Tetsuya Ito Figure 1 Figure 3

Claims (1)

[Scope of Claims] 1. An observation device for a second object in an apparatus for projecting and transferring a pattern drawn on a first object onto a second object via a projection optical system, comprising: an illumination optical system that illuminates the second object from between the optical system and the second object without going through the projection optical system; and an illumination optical system that is arranged coupled to the projection optical system and that emits illumination light from the illumination optical system. and an observation optical system for observing the second object. 2. N of the luminous flux that the observation optical system has on the second object
Observation according to claim 1, wherein the illumination optical system is arranged so that when A is sinA and the angle that the illumination optical system makes with the optical axis of the projection optical system is B, B≧A+10° is satisfied. Device. 3. The observation device according to claim 1 or 2, wherein the illumination light is incident at a Brewster's angle of a substance on the uppermost surface of the second object. 4. The observation device according to claim 1, 2 or 3, wherein the illumination light is incident from a direction perpendicular to a direction of an edge to be detected of a mark marked in advance on the second object. . 5. The observation device according to claim 1, 2, 3, or 4, wherein the incident angle of the illumination light changes with time. 6. Claim 1, wherein the illumination system is provided with a diaphragm therein, and the diaphragm limits the direction and area of the light hitting the mark previously marked on the second object. , 2, 3, 4 or 5. 7. The observation device according to claim 6, wherein the diaphragm is inclined with respect to the optical axis of the illumination optical system. 8. Claims 1, 2, 3, 4, 5, 6, or 7, wherein a spatial frequency filter corresponding to a mark marked in advance on the second object is arranged in the observation optical system. The observation device described. 9. The observation device according to claim 1, 2, 3, 4, 5, 6, 7, or 8, wherein the light source of the illumination light of the illumination optical system is a laser. 10. The observation device according to claim 9, wherein the illumination light is P-polarized and enters the second object. 11. Claim 9, wherein the laser beam of the illumination light is composed of a plurality of lasers having mutually different wavelengths, and has a number of observation optical systems corresponding to the number of wavelengths.
Or the observation device according to item 10. 12. The scope of the present invention is set at an angle such that when the laser beam is incident on a mark marked in advance on the second object, the diffracted light is substantially parallel to the optical axis of the projection optical system. Observation device according to item 9, 10 or 11.