JPS61168919A - Optical projection apparatus - Google Patents

Abstract

PURPOSE:To correct magnification errors and fluctuation of a focal point due to the fluctuation of temperature and atmospheric pressure simply, by providing a gaseous lens, in which a gas having a refractive index different from outer air, in a light path from a mask to a substrate, on which light is projected. CONSTITUTION:An annular frame 4, in which two transparent thin films 3a and 3b are provided with a specified interval being provided in the direction of a light axis, is provided in a light path between a mask M and a projecting lens 5. In a space between the thin films 3a and 3b, a gas, which has a refractive index different from that of outer air and has specified pressure, is sent from a gas supplying and exhausting means 11 through a pipe 10. A controller 16 detects the difference between the inner pressure and atmospheric pressure and outputs a control signals S3 to the gas supplying and exhausting means 11 so that said pressure difference becomes a specified value. When the pressure difference is changed, the curving amounts of the thin films 3a and 3b are changed, and the power of the gaseous lens formed by an airtight space 3 is changed. Therefore, the substantial length of the light path between the mask M and the projection lens 5 is changed.

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field of the Invention) The present invention relates to an apparatus that uses a projection optical system to expose a pattern of a mask onto a photoreceptor.

In particular, the present invention relates to a projection optical device that can stably maintain the overall optical characteristics from the mask to the photoreceptor even if the optical characteristics of the projection optical system change.

(Background of the Invention) Reduction projection type exposure apparatuses have recently been introduced into many VLSI production sites, and have brought great results.

One of its important performances is overlay matching accuracy. Among the factors that affect this matching accuracy, an important one is the magnification error of the projection optical system. The size of patterns used in VLSIs is becoming increasingly smaller year by year, and along with this, the need for improved matching accuracy is also becoming stronger. Therefore, the need to maintain the projection magnification at a predetermined value has become extremely high. Currently, by adjusting the magnification of the projection optical system when installing the apparatus, the magnification error can be reduced to a negligible level. However, there is an increasing demand for solutions that do not cause magnification errors even when environmental conditions change, such as slight temperature changes during equipment operation or slight pressure fluctuations in a clean room. Due to changes in conditions, not only the magnification changes, but also the position of the imaging plane of the projection optical system changes in the direction of the optical axis. - So-called focus fluctuations also occur. For this reason.

If this focus fluctuation is left as it is, the projected pattern image of the mask will have poor resolution on the wafer, which is a photoreceptor, and this will also lead to defects in the VLSI.

Such defects are also caused by exposure light passing through the projection optical system. This occurs because a portion of the exposure light energy is absorbed by optical elements within the projection optical system, causing a temperature change. Therefore, even if the environmental conditions are stable, variations in magnification and focus will occur while the device is in operation. The focal point fluctuation can be easily corrected by adjusting the distance between the projection optical system and the projection substrate such as a wafer, for example.

However, variations in magnification cannot be easily corrected unless the projection optical system is a so-called zoom lens (variable magnification optical system).

(Object of the Invention) In order to solve the above-mentioned problems, it is an object of the present invention to provide a projection optical device that can easily correct magnification errors and focus fluctuations caused by changes in temperature and atmospheric pressure.

(Summary of the Invention) The present invention uses two transparent members to form a gas lens in which a gas having a refractive index different from that of the outside air is sealed in the optical path from the mask to the projection target substrate (wafer, etc.). A gas lens container is provided that is held at a predetermined interval in the optical axis direction of the projection optical system, and at least one of the two transparent members has a curved state that changes depending on the pressure difference between the internal pressure of the gas lens container and the external pressure. The optical characteristics (magnification and focal plane position) from the mask to the projection substrate can be adjusted using the fact that the power of the gas lens changes with changes in differential pressure. This is the technical point.

(Embodiment) FIG. 1 is a diagram showing a schematic configuration of a projection optical apparatus, a so-called reduction projection type exposure apparatus, according to a first embodiment of the present invention. Light from an illumination light source (not shown) is transmitted through a condenser lens 1
A mask or reticle (hereinafter collectively referred to as a mask) M is illuminated with uniform illuminance through the illuminance. A circuit pattern to be exposed and transferred onto the projection substrate is formed on the mask M.
The mask M is designed so that the circuit pattern area is not blocked from light.

The mask holder 2 is vacuum-adsorbed. Luminous flux of circuit pattern image created by mask M (I! Luminous flux in optical path of light destination)
The light L enters a reduction projection lens (hereinafter simply referred to as a projection lens) 5 through an airtight space 3 that is sealed with a gas having a refractive index different from that of the outside air. The projection lens 5 forms a circuit pattern image onto a photosensitive wafer W serving as a projection target substrate. Wafer W#i moves in two dimensions! It is placed on the 7th stage 6 via the 2nd stage 7. The second stage 7 vacuum-chucks the wafer W on its upper surface and is provided so as to be movable in the vertical direction relative to the xy stage 6, that is, in the optical axis direction of the projection lens 50. And the vertical movement of stage 7 is x
This is performed by driving a motor 8 provided on the y stage 6. These two stages 7 move up and down for focusing so that the surface of the wafer W and the imaging plane of the projection lens 5 are aligned.

Now, in this embodiment, in order to form the airtight space 3, there is one in the optical path between the mask M and the projection lens 5.

An annular frame 4 is provided on which two transparent thin films 3a and 3b are stretched at a predetermined distance apart in the optical axis direction. Frame 4 is made into a cylindrical shape centered on the optical axis, $9, and thin film 3a.
, 3b are adhered to each end surface of the frame 4 so as to be stretched tightly with a predetermined tension and to maintain airtightness from the outside air. The cylindrical frame was made with a thin film of 3 m,
This is to obtain a spherical surface exactly equivalent to the lens surface when 3b is curved. Further, as the thin films 3a and 3b, it is preferable to use a pellicle made of a polymeric material that is commercially available as a protective film for masks and reticles. The film thickness of this pellicle is relatively thin, and the film thickness is very uniform.

The thin film of this example has an appropriate tensile stress of #
No. 1 provides good optical characteristics. The two thin films 3a
, 3b, gas at a predetermined pressure is fed from the gas supply/exhaust means 11 via the pipe 10. This gas is a single component with a refractive index different from that of the outside air (air).
Or it is a gas with multiple components. Here are some examples of suitable gas components:

Helium, nitrogen, chlorofluorocarbons, argon, benzene.

Examples include carbon dioxide or a suitable mixture of some of these components. For the gas supply/exhaust means 11.

Such gas is supplied via the vibrator 12 from an unillustrated source.

On the other hand, the pressure (internal pressure) of the gas sealed in the airtight air valve 3 is t! , detected and output by the pressure sensor 14. The pressure sensor 14 detects the internal pressure as an absolute value and outputs a signal S spread over that value. The pressure sensor 15 detects the pressure of outside air (atmospheric pressure) as an absolute value, and outputs a signal S according to that value.
, outputs.

The control device 16 inputs the signals Sl and S2, detects the difference between the internal pressure and the atmospheric pressure, and outputs a control signal S to the gas supply/exhaust means 11 so that the differential pressure becomes a predetermined value. . When the differential pressure is changed, the amount of curvature (curvature height) of the thin films 3a and 3b changes, the power of the gas lens by the airtight space 3 changes, and the substantial optical path between the mask M and the projection lens 5 changes. The length changes. In other words, by changing the power of the gas lens, it is equivalent to a minute change in the distance between the mask M and the projection lens 5, and the size of the pattern projected onto the mirror W is expanded or contracted minutely. become. However, changing the power of the gas lens means that the focal position of the entire projection system including the gas lens and the projection lens 5 changes. Therefore, the control device 16
In conjunction with the pressure adjustment in the airtight space 3, it also outputs a signal S4 for driving the motor 8 to correct the height of the wafer W.

This controls the projection lens 5 so that the converging surface of the projection lens 5 and the surface of the wafer W always coincide with each other. Furthermore, the control device 16
, the illumination light is mask M and a gas lens.

A signal S corresponding to the amount of magnification variation caused by passing through the projection lens 5 is input, and the differential pressure (the difference between the internal pressure of the gas lens and the atmospheric pressure) necessary to correct the magnification variation is input. ), and the control signal S is set so that the differential pressure can be obtained.
, outputs. The signal S representing the amount of magnification variation can be obtained in various ways. For example, the opening/closing operation of the shutter provided in the illumination optical system, the light intensity of the light source, and the mask M
A method of determining the cumulative amount of light energy incident on the projection system based on the transmittance of A method of determining the amount of magnification variation by depositing a thin film resistor, detecting the change in resistance value due to the temperature characteristics of the thin film resistor, and examining the change in temperature distribution on the lens surface. Select an optical element that exhibits a temperature variation similar to the magnification variation of the system, and detect the temperature of that element to determine the amount of magnification variation. Alternatively, there is a method in which a wavelength that is not used for exposure is stored deep inside the dichroic mirror in the illumination optical system. A conceivable method is to arrange a temperature sensor that receives the source of , and to obtain the amount of change in magnification equivalently by combining the thermal time constant of this temperature sensor with the time constant of change in magnification.

In addition, the gas lens (airtight space 3) Fi, two thin films 3
Even if a convex lens shape is formed in which both a and 3b are curved outward, if the gas sealed inside is smaller than the refractive index of air, it will not necessarily have positive power. On the contrary, it can also become a negative power.

Regardless of the deviation of b, the power of the gas lens is uniquely determined depending on the relative refractive index between the internal gas and the outside air, and the curvature and spacing of the thin films 3a and 3b.
The power of the gas lens can be adjusted by changing the curvature of the lens using differential pressure.

Now, FIG. 2 is a circuit diagram showing a schematic circuit configuration within the control device 16. As shown in FIG. The entire device is a processor such as a microcomputer or minicomputer (hereinafter referred to as CP).
It is centrally controlled by U) 30. Read-only memory (hereinafter referred to as ROM) 31 as a storage element
The file stores various parameter values and reference tables necessary to correct magnification fluctuations and focus fluctuations of the projection optical system, and software programs for calculations.

Specifically, it is necessary to correct atmospheric pressure P0 and 1 magnification variation ΔM to zero when the magnification from Mask M to Ueno W is precisely adjusted to a constant value (for example, 115 or 1/1G). , the differential pressure ΔDP between the airtight space 3 and the atmospheric pressure is 1. For example, when the magnification variation ΔM is expressed as the amount of expansion and contraction of the projected pattern on the sea urchin/%W, it is actually measured and memorized every 0.05 μm of ΔM. The reference table (hereinafter referred to as the first table), the coefficient of the magnification variation amount ΔM with respect to the atmospheric pressure variation amount ΔP, and the focus correction amount Δz to be corrected at the same time when one magnification variation is corrected.
A reference table (hereinafter referred to as a second table) in which C is determined by actual measurement and stored in correspondence with ΔM is maintained. The ROM 31 also stores a coefficient C representing the relationship between the atmospheric pressure fluctuation amount ΔP and the focus fluctuation amount Δ2 of the projection optical system. It should be noted that the experimental results showed that the atmospheric pressure fluctuation amount ΔP and the magnification fluctuation amount ΔM or focus fluctuation amount Δ2 are proportional to tX, so regarding the relationship between ΔP and ΔM or the relationship between ΔP and Δ2, It is sufficient to simply calculate C as the proportionality coefficient.

On the other hand, the CPU 30 converts a signal S corresponding to the internal pressure from the pressure sensor 14 into an analog-to-digital converter (hereinafter referred to as A).
DC) 33, pressure sensor 15
Signal S according to the atmospheric pressure from! is read in via the ADC 32, and further, the CPU 30 appropriately outputs a signal S for adjusting the internal pressure of the gas lens and a signal S4 for moving the wafer W up and down during focus fluctuation correction. The CPU 3G then controls the projection optical system (from the mask M to the wafer W).
A signal S from a fluctuation detection means 35 that detects the amount of change in height caused by the illumination light passing through a gas lens (including a gas lens) using the b-shift force 1 of some of the methods described above is also input. This variation detection means 35#i also detects the amount of focus variation caused by light incident on the projection optical system. There is a unique relationship between the amount of variation in focus and the amount of variation in magnification.If the amount of variation in either one is known, the amount of variation in the other can be found by simply multiplying it by a constant.

Here, an example of variation in magnification due to incidence of illumination light on the projection optical system will be explained with reference to FIG. Figure 3(a) is! 7 shows the temporal change in the illumination light passing through the projection optical system when the image of the circuit pattern of the mask M is exposed on the wafer W nine times by stepping the stage 6. Figure 3 (
In a).

The horizontal axis represents time, and the vertical axis represents the amount of illumination light passing through the projection optical system, that is, the intensity of the light illuminating the mask M, and the intensity of the illumination light passing through the projection optical system.
represents the incident energy EJ according to the product of the total area (transmission base) of the illumination portion in the circuit pattern. FIG. 3(b) shows the temporal change in the amount of variation in magnification of the projection optical system in correspondence with the step-and-repeat exposure operation (FIG. 83(&)). Assuming that the exposure operation is started at time t0, the 1x high fluctuation amount ΔM increases logarithmically as the incident energy El is integrated at each step, and remains almost constant at time t. reaches the saturation value ΔMS. The magnitude of the saturation value ΔM8 changes in proportion to the magnitude of the incident energy E7. Due to this magnification variation characteristic, from time 1o to 1. The logarithmic change up to 4 depends on the configuration of the projection optical system.

It is difficult to approximate using a formula with a time constant of 1.

Generally, it is approximately expressed by a formula that combines two to four time constants. Now, after reaching the saturation value ΔMS at time t,
The step-and-repeat exposure duty and the incident energy RJ remain unchanged for a limited time.

The saturation value ΔMS does not change. And time t! When the exposure operation is completed, the 1-times high variation amount ΔM decreases exponentially, and returns to zero, which is the initial value of the 1-times high variation amount ΔMFi, unless the exposure operation of the first-order Uni-No is started.

As a result of the experiment, it was confirmed that the characteristics of focus fluctuation are similar to the characteristics of height fluctuation shown in FIG. 3(b). Therefore, the actual focus variation amount Δ2 can be easily determined by multiplying the double height variation amount ΔM by the conversion constant. The fluctuation detecting means 35 sequentially outputs a signal S representing the magnification fluctuation amount ΔM and the focus fluctuation amount Δ2, which are 1 or more, to the CPU 30.

Next, the operation of this embodiment will be explained with reference to the flowchart shown in FIG. This flowchart represents a program executed by the CP03G to correct variations in magnification and focus.

At step 10G, CPU 30i1i pressure sensor 14.
Read the signal Sr-ag from 15, step 101
The atmospheric pressure fluctuation amount ΔP and the differential pressure ΔDP are calculated. The atmospheric pressure fluctuation amount ΔP is the difference between the value of the signal S, and the initial atmospheric pressure Po stored at the time of manufacture, and the differential pressure ΔDlt signal S, and S! is the difference between the values of Next, in step 103, the CPU 30 reads the signal S from the fluctuation detection means 35, and detects the magnification fluctuation amount ΔM and the focus fluctuation amount Δ2 due to the incident energy of the projection optical system at that time. Then, in step IC)4, the CPU 30 calculates a double height fluctuation amount ΔMP and a focus fluctuation amount ΔzP that correspond only to the atmospheric pressure fluctuation amount ΔP. ΔMP and Δz
P can be calculated by multiplying the atmospheric pressure fluctuation amount ΔP by the coefficient. In step 105, the CPU 30 calculates the overall magnification fluctuation amount ΔM caused by both the atmospheric pressure fluctuation and the incident energy.
T and the focus variation amount ΔZT are calculated. In general, the amount of height variation ΔMT is the sum of ΔMP and ΔM, taking into account the positive and negative, and the focus variation ΔZT is the sum of ΔzP and Δ2, taking into account the positive and negative.

Next, in step 106, the CPU 30Fi searches the first table in the ROM 31 and selects the differential pressure ΔDP' corresponding to the one-time high fluctuation amount ΔMT as the target value. Then, in the next step 107, the CPU 30 searches the second table in the ROM 31.

A pointless correction amount ΔzC corresponding to the double height variation amount ΔMT is selected. Furthermore, the CPU 30#i further corrects the previously determined focus variation amount ΔZT by the correction amount ΔzC. That is, Δ
The added value including the positive and negative values of ZT and ΔzC is newly calculated as the focus variation amount Δ
Let's call it ZT. The focus variation amount ΔZT obtained in this way is
This includes both focus fluctuations caused by atmospheric pressure fluctuations and incident energy, and focus fluctuations that occur incidentally when the magnification fluctuations at that time are corrected to zero.

Next, in step 108, the CPU 30 determines the focus variation amount Δ
A drive signal S4 that makes ZT zero is output to the motor 8, and the height of the second stage 7 is corrected.

Subsequently, the CPU 30 determines the target differential pressure ΔDP in step 110.
A control signal S1 corresponding to ' is outputted, and the fluctuation correction routine is ended. Its control signal S.

After outputting the signal 83., the CPU 30 outputs the signal 83. am is read, it is detected whether or not the actual differential pressure ΔDP matches the target differential pressure ΔDP', and if they match, the control signal S is generated so that the supply/exhaust means 11 stops pressurizing, operating, or depressurizing operation. , outputs.

The series of operations described above are constantly repeated at short time intervals (for example, 5 to 10 seconds) while the exposure apparatus is in operation. The time interval can also be made longer if variations in optical properties due only to atmospheric pressure variations are to be corrected. Conversely, when the exposure device is performing an exposure operation, it is necessary to set the time interval as short as possible.

As described above, according to this embodiment, since a gas lens whose power changes depending on pressure is provided in the optical path between the mask M and the projection lens 5, the projection magnification can be finely adjusted without providing any mechanically movable parts. becomes possible.

Stable fold height control is achieved. Note that this embodiment has a configuration that automatically corrects variations in magnification and focus.
The internal pressure of the gas lens may be manually controlled, and the temperature of the gas supplied into the gas lens must be constant throughout the manufacturing, adjustment, and operation of the device. Therefore, it is desirable to provide a semiconductor heating and cooling element (such as a Peltier element) on the inner wall of the frame 4 that holds the gas lens so as not to block the exposure light beam, and to detect the temperature of the gas in the gas lens with a sensor to control the constant temperature. 10 Next, a second embodiment of the present invention will be described with reference to FIG. 5. The basic configuration is the same as the device shown in FIG. 1, and the same components as in FIG. 1 are given the same symbols. This embodiment differs from the first embodiment in that the gas lens is provided in close contact with the lens g of the projection lens 5. For this purpose, a part of the lens barrel of the projection lens 5 is extended upward to form a cylindrical holding frame 5a, and a transparent thin film 3ai is stretched over the upper end surface of this holding frame 5a. When the gas lens (airtight space 3) is configured in this way, the element that changes the power of the gas lens is only one thin film 3a, so the reproducibility during 1 magnification correction is improved by using the two thin films 3a and 3b. There is an advantage that it is improved over the previous case.

Next, a third embodiment of the present invention will be described with reference to FIG. This embodiment differs from the above-mentioned embodiments in that the gas lens is not provided on the exposure apparatus side but on the mask M side. In this embodiment, a protective pellicle 3b attached to the pattern surface side of the mask M at regular intervals is configured to serve as a power variable element of the gas lens. The pellicle 3b is stretched over the frame PF, and the frame PF' is bonded so as not to block the circuit pattern area of the mask M from light. A ventilation hole 22 for communicating the internal space 3 of the pellicle 3b with outside air is formed on the side of the frame PF. Such a mask M with a pellicle protection device is equipped with a mask holder 20 that supports the periphery of the pattern side of the mask M.
placed on top of.

vacuum adsorbed. A gas supply/exhaust hole 20a is formed in the side wall of the mask holder 20 so as to face the ventilation hole 22. A pipe 10 communicating with the supply/exhaust means 11 is connected to the supply/exhaust hole 201L.

In addition, an airtight joint 24 is provided on the side of the ventilation hole 2 (la) facing the frame PF to maintain airtightness with the ventilation hole 22. For this reason, the mask M is placed in a predetermined position of the mask holder 20. When placed, the air supply/exhaust hole 20a and the ventilation hole 22 communicate with each other, and the internal space 3 is sealed with a gas having a refractive index different from that of the atmosphere at a predetermined pressure.

Actually, in order to form a gas lens with such a configuration, after putting on the mask M, the outside air (air) existing in the internal space 3 is expelled, and the gas from the pipe IO is sealed instead. operation is required. The operation after the gas is sealed is the same as in the first embodiment described above. however.

In the case of this embodiment, since the height of the frame PF and the area of the pellicle 3b may be slightly different for each mask M, the values of various data (coefficients and tables) stored in the ROM 31 are replaced with the mask M. It is necessary to correct only a small amount each time you use it.

As described above, in this embodiment, since the mask protection device itself provided on the surface of the mask M facing the projection lens is used as a gas lens, there is no need to make any major modifications to the main body of the exposure apparatus.

It is extremely cylindrical. In this embodiment as well, the curvature of the pellicle 3b allows the substantial optical path length from the mask M to the projection lens to be varied, and correction of 1x height variation and focus variation can be similarly implemented.

In addition to the above-mentioned embodiments, the present invention may have many modifications. For example, the position where the gas lens is provided may be the space from the projection lens 5 to the wafer W-1, or a transparent thin film may be stretched in the air gap between the lenses in the projection lens 5. The airtight space 6 between the thin film and the lens may be sealed with a gas having a refractive index different from that of the atmosphere at a predetermined pressure. Furthermore, in the first to third embodiments of the present invention, the power of the gas lens was forcibly varied, but if only atmospheric pressure fluctuations were to be corrected, a static control method could be used. Conceivable.

For example, the volume of the airtight space 3 of the gas lens is sealed so as to change in accordance with the curvature of the thin film due to fluctuations in atmospheric pressure, so that the power of the gas lens changes in proportion to fluctuations in atmospheric pressure. Then, the tension of the transparent thin film, the refraction height of the gas, the thickness of the gas lens, etc. are appropriately determined so that the power change at this time self-corrects the magnification and focus change due to changes in atmospheric pressure.

In addition, in each of the above embodiments, the refractive index of the gas constituting the gas lens was constant regardless of the pressure, but depending on the type of gas, the refraction may vary depending on the absolute pressure value in the airtight space. The rate also changes somewhat, so even if the difference between atmospheric pressure and internal pressure is constant, if the absolute value of internal pressure differs,
The power of gas lenses is not necessarily the same. Therefore, even more precise control becomes possible if data on the amount of change in refractive index with respect to the amount of change in gas pressure is actually measured in advance and used as a parameter when correcting fluctuations in magnification and focus.

(Effects of the Invention) According to the present invention, even if the characteristics of the projection optical system provided between the mask and the projection substrate vary due to changes in atmospheric pressure, changes in incident energy of illumination light, etc. Since these fluctuations can be easily corrected by adjusting the power of the gas lens without providing any mechanical movable parts, it is possible to obtain the effect that the reproducibility of extremely minute correction control is greatly improved. Furthermore, since it can be easily retrofitted into existing exposure equipment, it is possible to further reduce magnification fluctuations and focus fluctuations that existing exposure equipment has, and it is possible to overlap fine patterns that are considered to be at the limit with existing exposure equipment. Another advantage is that the device is fully practical even in the alignment exposure process. Further, the present invention is applicable to a 1:1 equal-magnification exposure apparatus as long as it has a projection optical system between the mask and the projection target substrate.

Exactly the same effect can be obtained even when applied to a reflection projection type exposure apparatus or the like.

[Brief explanation of drawings]

FIG. 1 is a diagram showing a schematic configuration of a projection exposure apparatus according to a first embodiment of the present invention, FIG. 2 is a circuit block diagram showing a schematic configuration of a control device, and FIG. 3 is a diagram showing an exposure operation. FIG. 4 is a flowchart for correcting fluctuations in optical characteristics; FIG. 5 is a diagram schematically showing a projection exposure apparatus according to a second embodiment of the present invention;
FIG. 6 is a diagram showing the configuration of a gas lens according to a third embodiment of the present invention. (Explanation of symbols of main parts) M: Mask, W: Queha. 2...Mask holder, 3...Airtight space. 3&, 3b...Transparent thin film, 4...
·flame. 5... Projection lens. 11... Gas supply and exhaust means. 14.15...Pressure sensor. 16...Control device.

Claims (4)

[Claims] (1) In a device that projects a pattern formed on a mask onto a projection substrate via a projection optical system, a gas having a refractive index different from that of the outside air is sealed in the optical path from the mask to the projection substrate. In order to form a lens, a gas lens container is provided that holds two transparent members at a predetermined interval in the optical axis direction of the projection optical system, and at least one of the two transparent members is attached to the gas lens container. The mask is a transparent thin film stretched so that its curved state can change depending on the pressure difference between internal pressure and external pressure, and the power of the gas lens changes depending on the change in the pressure difference. A projection optical device characterized by adjusting optical characteristics from to a projection target substrate. (2) The gas lens container is arranged between the mask and the projection optical system, and consists of a cylindrical frame having a predetermined height in the optical axis direction, and has a height on each end surface of the frame. 2. The device according to claim 1, further comprising a transparent thin film made of a molecular material. (3) The gas lens container is a part of the lens barrel of the projection optical system, and the other of the two transparent members is one optical glass element in the projection optical system. The device according to item 1. (4) the other of the two transparent members is the mask;
The gas lens container is a frame fixed to the mask so as not to block light from the pattern area, and the transparent thin film is a pellicle for protecting the mask stretched over the frame. The device according to item 1.