JPH0288906A - Alignment optical system - Google Patents

Abstract

PURPOSE:To always obtain a stable misalignment detecting signal regardless of the magnitude of reflectance on a wafer surface by inclining a misalignment detecting light beam in a direction orthogonally crossed with a misalignment detecting direction so that the light beam is made incident and providing an aperture for cutting multiple reflected light on a mask. CONSTITUTION:A beam projecting optical system is inclined in the direction perpendicular to the misalignment detecting direction so as to make the beam 1 obliquely incident and the aperture which can cut the multiple interference light between the mask 3 and the wafer 4 is provided on a mark. When the misalignment detecting beam 1 inclined by an angle alpha is made incident just on the aperture which opens to the mask 3 placed close to the wafer 4, the light exiting to the outside from the aperture of width alpha again after being reflected on the surface of the wafer becomes very little. Thus, gap dependent component included in the light obtained through the aperture can be made very little.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to an alignment optical system that cuts multiple reflections between a mask and a wafer that adversely affect positional deviation detection accuracy in proximity exposure.

Conventional technology Multiple reflections between a mask set at a gloximity gap and a wafer have been a major problem in the past, and are a cause of fatal errors, especially in optical systems that detect light intensity from alignment marks. It has become. Figures 7 to 9 show examples of measures to prevent multiple reflections (USP 4.614.
(See specification No. 433).

FIG. 8 shows what kind of multiple reflections occur in a combined optical system using concentric Fresnel zone plates. 112 is a polyimide film with a thickness of 2 μm,
IIO is a BN film with a thickness of 4 μm, and a Fresnel zone plate indicated by 118.120 is formed on this mask 111. 122 is a wafer and 128 is a resist. Fresnel zone plates designated 124 and 126 are formed on the wafer 122. 321.123
shows the circuit pattern on the mask 111. A portion of the incident laser light 138 and 140 undergoes reflective diffraction as shown at 142 and is focused at 144 . On the other hand, as shown by 146, there is also light that passes through the mask 111, and from now on 1
46 is reflected on the resist 128 and wafer 122 surface to form light 148 (again Fresnel zone play) 118
．． 120, and the diffracted light passes through the 1420 optical path.

Since the diffracted light 148 from the pickle 140 has a phase difference depending on the gap interval α1, the focal point 144
The light intensity is affected by the gap variation. It is reported that the amount of variation is nearly 90% for a gap change of 0.15 μm.

Further, the incident beam 130 passes through the mask 111 and becomes a beam 150, which is reflected by the resist 128 and the wafer 122 to become light 151, and then passes through the polyimide film 112.
, and the front and back surfaces of the BN film iio to become beams 153.155.157. These interfere with the reflected beam 130, so that the light 15 illuminating the wafer 122 is
0, the intensity changes due to gap fluctuation. Therefore, Fresnel zone plate 124.12 on wafer】22
This means that the light intensity that irradiates the lens 6 varies, and naturally the light intensity that reaches the focal point also varies.

Also, the light 152 diffracted by the Fresnel zone plates 124 and 126 is reflected by the light 158 to 16 within the mask.
0, this also causes light intensity fluctuations at the focal point 136. Fluctuations in light intensity lead to a decrease in the S/N ratio of the detection signal, resulting in a decrease in detection accuracy.

The above-mentioned patent takes the following measures to deal with the effects of multiple reflections as described above. As shown in Figure 9,
If the light transmission prevention layer 72 is provided below the marks 118 and 120 on the mask 111, the amount of light transmitted through the mask 111 can be significantly reduced. As the material for this layer 72, a vapor deposited film of aluminum or the like or a resist containing dye is used. 88 is an antireflection film, and the polyimide film 112 is made to have an optical thickness that is an odd multiple of λ/4.

It has been reported that these measures significantly reduce reflection within the mask.

Problems to be Solved by the Invention However, applying the device shown in FIG. 9 on the mask not only complicates the mask manufacturing process, but also makes it difficult to change the pattern by selectively attaching a light transmission prevention film. There are problems such as distortion.

The present invention has been made in order to solve the conventional problems as described above, and an object of the present invention is to provide an alignment optical system that can extremely easily eliminate or significantly reduce multiple reflections between a mask and a wafer. The purpose is to

Means for Solving the Problems To achieve this object, the present invention tilts the beam projection optical system in a direction perpendicular to the positional deviation detection direction, makes the beam obliquely incident, and furthermore performs mask-to-wafer multiplexing on the mark. The above object is achieved by providing an aperture that can cut interference light.

Function The function of the aperture formed on the mask in the present invention is as follows. In other words, when a positional deviation detection beam tilted by an angle α is directed into an aperture opened in a mask placed close to the wafer, sea urchins and
- The amount of light that is reflected by the surface and exits again from the aperture with width d becomes extremely small if d<2g5inα, and the gap-dependent component included in the light obtained through the aperture can be greatly reduced.

EXAMPLES Below, examples of the present invention will be described in detail with reference to the drawings. First, the principle of the present invention will be explained using FIGS. 5 and 6.

Figure 5 shows that the positional deviation detection pim l tilted by the angle α is
It is not shown that the light is incident on the aperture opened in the mask 4 placed close to the wafer 3 (actually, an alignment mark such as a diffraction grating is made in the aperture).

The light 2 that is reflected on the wafer surface and goes out again from the aperture with width d becomes extremely small if d<2g5inα (in the proximity gap, d must be sufficiently small, and the influence of Fresnel diffraction is large). It is difficult to completely reduce light 2 to 0)
. In other words, the gap-dependent component contained in the light obtained through the aperture under the above conditions is extremely small. Generally, during proximity exposure, the gap is set appropriately within a certain range, but d calculated by the above formula may be used for the minimum value of the gap interval. In addition, the operation of the present invention is based on the above principle, and no matter how high the reflectance of the wafer is, the mask can be used without being affected by multiple reflections.
This is to align the wafer. Further, as shown in FIG. 6, if an aperture is provided on the mask 4 for marks on the wafer 3, the influence of multiple reflections can be prevented. In other words, if d<4g5inα, the light 5 that is reflected from the wafer, then further reflected from the lower surface of the mask, and then comes out from the aperture after being reflected from the wafer (light that makes two round trips between the wafer and the mask) is It can be made small enough.

Then, the light obtained through the aperture is transmitted to the wafer.
Only the light 6 has gone back and forth between the masks, and as a result, the intensity does not change due to gap fluctuations.

Next, a first embodiment of the present invention will be described based on FIGS. 1 to 4. FIG. 1 is a schematic configuration diagram of an alignment optical system for explaining a first embodiment. Here, a HeNe Zeeman laser 10 was used as a light source for alignment illumination with high spatial and temporal coherence.

The two orthogonal polarization components l and f2 emitted from this HeNe laser are first expanded in beam diameter by a beam expander 11, and then amplitude-divided by a reference diffraction grating 12.

One of the two divided optical paths passes through the λ/2 plate 13! As a result, the fl and f2 polarized light interchange with each other. Both optical paths are 2
The light is projected through a set of Fourier transform lenses 14 toward alignment marks made on the mask 3 and wafer 4 set in the gloximity gap. Further, a spatial filter 15 is placed on the Fourier plane of the Fourier transform lens, and is buttoned so that only the ±1st-order diffracted light can pass through. The marks provided on the mask 3 are shown in FIG.

The width d of the diffraction grating is set between 4 and 12 μm, and
In order to prevent a decrease in the intensity of diffracted light due to the narrowing of the area of the diffraction grating, the length in the longitudinal direction is increased.

When viewed from the direction of the arrow in FIG. 2, it is easy to see that the structure is as described above with reference to FIG. 5, and multiple interference light between the mask and the wafer can be eliminated.

Here, the width of the diffraction grating (corresponding to d in FIG. 5) was set to 7 μm, assuming that the minimum gap setting value was m μm. The diffraction grating on the wafer is square, as shown in FIG. 3, and has the same area as the diffraction grating on the mask.

For the diffraction grating on the mask, we do not take measures against multiple interference (the method explained earlier using Figure 6), but this is because detection errors due to multiple reflections of the diffraction grating on the mask are Since the upper diffraction grating is not very large, it is omitted in this example. The emitted diffracted lights from the two diffraction gratings are imaged onto a knife surface by an imaging lens 16. A polarizer 17 is provided up to the image plane, and this is to cause the fl and f22 portions of the laser beam to interfere with each other to generate a beat. By Naifetsuji Miller 18,
Two diffracted lights (the diffracted light from the wafer and the diffracted light from the mask) are guided to different photomuls 19. At Hotomaru, fl
, f2 are observed, and the amount of positional deviation between the two diffraction gratings can be determined by measuring the phase difference between the two beat signals (optical heterodye method).

The diffraction grating on the mask in this embodiment may be replaced by the one shown in FIG. In this diffraction grating, the longitudinal dimension of the diffraction grating is reduced, and the diffraction gratings are stacked in several stages in order not to reduce the area of the diffraction grating. In this embodiment, the laser beam projection system uses a beam splitter 11 instead of a reference grating, a Fourier transform lens, and a spatial filter.
It is also possible to configure it with a mirror 31 and a photodetector 32 (see FIG. 7).

However, it can be said that this method of cutting multiple reflected light is effective for all alignment optical systems that allow oblique incidence.

According to the alignment method of the present invention as described in detail,
In mask-wafer alignment in the proximity gap, errors in positional deviation detection signals due to multiple reflections, which have been a problem in the past, can be almost completely eliminated, and the alignment is always stable regardless of the reflectance of the wafer surface. A positional deviation detection signal is obtained. The lattice shape that is the core of the present invention is very simple and easy to create, and since the incident light and outgoing light are tilted, the optical system does not emit exposure light such as X-rays. Another effect is that the arrangement of the optical system becomes easier.

[Brief explanation of the drawing]

FIG. 1 is an overall schematic diagram for explaining the alignment optical system in the first embodiment of the present invention, FIG. 2 is a detailed diagram of the alignment marks on the mask of the same embodiment, and FIG. 3 is the same embodiment. FIG. 4 is a detailed view of the alignment marks on the wafer of the first embodiment, FIG. 4 is a detailed view showing other embodiments of the marks on the mask of the first embodiment, and FIGS. 5 to 6 are for clarifying the principle of the present invention. FIG. 7 is an overall schematic diagram of the second embodiment of the present invention, and FIGS. 8 to 9 are explanatory diagrams for explaining the conventional technology. 3... Membrane mask, 4... Wafer, 10...
・Seman laser, 11...beam expander, 1
2°゛・Reference diffraction grating, 13...λ/2 plate, 14...
・Fourier transform lens, 15... Spatial filter, 16
...Imaging lens, 17...Polarizer, 18...Naifetsuji. Name of agent: Patent attorney Shigetaka Awano and one other person

Claims (1)

[Claims] A positioning optical system that includes a positional deviation detection light beam that is incident obliquely in a direction perpendicular to the positional deviation detection direction, and an aperture that is formed on the mask to cut multiple reflected light between the mask and the wafer.