JP2009099902A - Cooling mechanism for discharge-excited gas laser apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a cooling mechanism for a discharge-excited gas laser apparatus that can efficiently cool a peaking capacitor and a high-voltage applied portion and perform highly repetitive operation. <P>SOLUTION: In a laser chamber 1, anode and cathode electrodes 2a and 2b are provided, and the cathode electrode 2b is fitted to the laser chamber 1 having a ground potential with an insulator 5 interposed. A high-voltage supply member 3 is electrically connected to the cathode electrode 2b and one terminal of the peaking capacitor 4, and the other terminal of the peaking capacitor 4 is connected to the laser chamber 1 with the ground potential through an electric conducting member 8 fitted with a water cooling jacket 6. As the insulator 5 provided between the cathode electrode 2b and laser chamber 1, a ceramic-made member is provided which is made of aluminum nitride etc., and has high-insulation and high-heat-conductivity properties, and heat escapes to the side of the laser chamber 1 through the insulator 5 to cool the high-voltage supply member 3 and further cool the peaking capacitor 4. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a cooling mechanism for a discharge excitation gas laser device, and more particularly to a cooling mechanism for a discharge excitation gas laser device capable of effectively cooling a high-pressure side member of the laser device.

As the technology node advances, in the 45 nm to 32 nm node, the exposure apparatus using an ArF excimer laser as the light source increases the NA (1.3 to 1.5) by immersion technique, double exposure, double exposure, etc. Is considered to be the main force.
The requirements for a light source for an ArF laser exposure apparatus are shown below.
1. A high repetition frequency (10 kHz or more) and a high average output (100 W or more) are required with ensuring high dose stability and high throughput.
2. As the NA increases, a further ultra-narrow band (spectrum purity of 95% spectral width of 0.3 pm or less) is required.
This is a requirement from miniaturization of processing dimensions. At present, the spectral line width of ArF laser is 0.5 to 0.35 pm with 95% purity. In the future, this band narrowing will further progress, and it is considered that 0.2 pm or less will be required.
3. In order to reduce the influence of speckle on the mask of the exposure apparatus, the spatial coherence of the output laser light is required to be low.
In particular, in the double exposure, since it is necessary to expose the wafer twice, it is necessary to increase the light source output in order to increase the productivity.

In order to satisfy such requirements, an ArF excimer laser as a light source employs a double chamber system equipped with two chambers. This is because in a laser using a single laser chamber, it is practically necessary to increase the output while maintaining good optical performance such as a narrow spectral width because of the stable operation of the system and the limitation of the lifetime of each module. Because it is very difficult.
In the double chamber system, the part where the first chamber called the oscillation stage is mounted is the part where the laser with high output and high optical performance is made and the next chamber called the amplification stage is mounted. In this way, the above-mentioned difficulties in practical use are overcome, and the above-mentioned industrial application requirements are met.

By the way, the excimer laser is a so-called pulse laser that emits light intermittently. Therefore, the laser output is represented by the product of energy per pulse and repetition frequency. Therefore, two methods are conceivable for increasing the output of the light source in the exposure machine.
One is a method of increasing the energy of the light source. The other is a method of increasing the repetition frequency. Each of these methods has advantages and disadvantages. Which method is adopted is determined by practical difficulties.
At present, a major problem is the durability of optical elements. In an exposure machine using deep ultraviolet light, since the photon energy of deep ultraviolet light is high, the optical element is likely to be damaged by light. In order not to cause such damage, the optical element is required to have high durability performance and various devices are made.
Naturally, the durability performance of the element itself is required, but a device for reducing the load on the element is also devised. It has been found that reducing the peak intensity of the time and position of the incident light on the optical element reduces damage to the optical element.
Therefore, the above-described method of increasing the light source output by increasing the pulse energy increases the load on the optical element, which is not preferable from the viewpoint of the lifetime of the optical element. Therefore, it is considered preferable to increase the repetition frequency and increase the output of the laser from the standpoint of constructing a stable system with a long lifetime of the optical element and thus the exposure system as a whole.

FIG. 5 shows a sectional view of a laser chamber of a general discharge excitation laser.
In the laser chamber 100, a cross flow fan 121 for flowing a laser gas between the electrodes, a heat exchanger 122 for cooling the discharged laser gas, an anode and cathode electrodes 131 and 132 for exciting the discharge, It comprises a preliminary ionization mechanism 125 for generating uniformly and efficiently, and a wind guide 123 for efficiently flowing laser gas flowing through the cross flow fan 121 between the electrodes.
A high voltage supply member 135 that is insulated from the chamber 100 by an insulating material 101 is connected to the power source 130, and the peaking capacitor 134 is charged through the high voltage supply member 135.
When the peaking capacitor 134 is charged, a high voltage is generated between the cathode electrode (high voltage side) 132 and the anode electrode (ground potential side) 131. At that time, the preionization mechanism 125 operates at a certain timing, and predischarges the discharge space. As this preionization mechanism 125, a mechanism using corona discharge is generally used. After pre-ionization occurs, discharge starts when the voltage between the cathode electrode 132 and the anode electrode 131 exceeds the dielectric breakdown voltage. Excimer laser oscillates by selecting and amplifying light of a certain wavelength generated by this discharge with a resonator.

The biggest technical problem in the case of operating an excimer laser at a high operating frequency is discharge stability during high repetition operation. As described above, the excimer laser causes intermittent discharge to excite the laser gas and cause laser oscillation. However, if the repetition frequency is increased, this intermittent discharge cannot be generated stably. This is because if the repetition rate is increased, the discharge changes from a normal glow discharge to an arc discharge or streamer discharge, and the laser gain during discharge is not uniform, and the gain length necessary for laser oscillation cannot be secured. It is believed that.
In the excimer laser, the most effective and high repetition rate is to narrow the discharge width. Generally, the maximum operable repetition rate in an excimer laser is described in relation to the clearance ratio (CR). CR is the ratio between the product of the interelectrode gas flow velocity v and the discharge interval time t and the discharge width W, that is, CR = vt / W. When CR is sufficiently large, the discharge is stable and high repetition operation is possible. If CR is large, the discharge is generated stably and the energy stability of the laser is also improved.

  From the viewpoint of physical phenomena, CR will be described as follows. Ions generated by discharge, discharge products such as active species, scattered dust and debris from the electrode significantly reduce the discharge resistance of the gas. Further, due to the discharge, a gas lean portion is generated, and since this portion has a relatively low gas pressure, the discharge resistance is small. Therefore, if this portion (discharge product) is present in the vicinity of the electrodes, the next generated discharge occurs not in the electrodes but in this portion. A large CR means that this discharge product is kept away from the vicinity of the electrode when the next discharge occurs. FIG. 6 shows a cross-sectional view of the electrode portion and shows a schematic diagram of normal discharge and abnormal discharge. In the figure, 131 and 132 are electrodes, and G is a gap between electrodes.

  The required CR value depends on the laser application. For applications where laser energy stability is not as critical, CR may be as low as 2. However, when used as a light source for semiconductor exposure, since high energy stability is required, two or more CRs are required. Therefore, for example, when the discharge width is 3 mm and the operation is to be performed at 6000 Hz, the inter-electrode gas flow velocity needs to be about 50 m / s or more.

Next, the double chamber described in the background will be described.
When a double chamber system is considered, the oscillation stage chamber has less technical difficulty for high repetition than the amplification stage. This is because the output and energy required for the oscillation stage are usually small, and the energy density of the output beam can be designed small. When the energy density of the output beam is high, optical elements such as an output mirror and an output window are damaged, so that practical performance such as life performance cannot be achieved. Narrowing the discharge width reduces the width of the laser beam. If the discharge width is narrowed in order to produce the same output energy, the energy density increases. In the case of the oscillation stage, since the output energy is small, the allowable discharge width becomes narrow. However, since the output energy is high in the amplification stage, the allowable discharge width becomes large.
As a means for avoiding the problem of high energy density, a large interelectrode distance can be mentioned. By increasing the height as the beam width is reduced, the cross section of the laser beam can be increased and the energy density can be reduced.

Another technical problem in the high repetition operation is exhaust heat.
When the laser is operated, an electrical loss is generated by a peaking capacitor, a metal in the current path, etc., and the heat is changed to heat. In particular, the loss in the peaking capacitor is a tangent loss, so-called tan δ, which is a representative value representing the loss, but tan δ increases in proportion to the repetition frequency.
Therefore, at the time of high repetition, the amount of heat due to loss increases. The increase in temperature of the peaking capacitor causes fluctuations in the capacitance of the peaking capacitor, causing a deviation from the optimum operating point of the laser design. Therefore, in the high repetition laser, cooling near the peaking capacitor is required. There are several methods for cooling (see Patent Documents 1, 2, 3, 4, 5, etc.).

First, a jacket-type cooling method using a coolant such as water is used. For example, as disclosed in Patent Document 5, cooling is performed by attaching a cooling member to the connecting portion of the peaking capacitor.
Second, there is a forced air cooling system that lowers the temperature in the ambient air.
As a third method, there is a method in which the peaking capacitor portion is immersed in insulating oil to cool and circulate the oil. Instead of this insulating oil, a fluorine-based inert liquid (typically known by the trade name Fluorinert (registered trademark)) may be used. In this way, unnecessary heat is exhausted in the prior art.
The thermal problem is considered to be a problem even with a preionization mechanism that has an important effect on laser performance. However, in low repetition lasers, the effect is hardly a problem. This is because the energy input to the preliminary ionization mechanism is smaller than the energy input to the main discharge. In general, the energy input to the preionization is 1/5 to 1/20 of the energy input to the main discharge, and the temperature rise caused by the energy is smaller than that of the main discharge circuit. Is not a particular problem.

Moreover, there is a life of an electrode as a problem when the laser is highly repeated. The electrode life is determined by the number of laser shots if the input energy is the same. Therefore, when the repetition frequency is doubled, the time life of the electrode is halved.
As stated in the background, the purpose of a high repetition laser is to improve throughput. Therefore, halving the time life that shortens the maintenance interval cannot satisfy the performance as an industrial machine. Thus, in high-repetition laser technology development, high repetition lasers have been developed that extend the electrode life and have a long time life by improving electrode materials. Conventionally, high repetition lasers have been designed and developed in this way.

JP-A-5-82863 Japanese Utility Model Publication No. 6-66052 JP-A-7-221372 JP-T 8-505006 JP 2003-249703 A

The most important issue in the practical application of high repetition laser is the required power. In order to increase the repetition frequency of the laser, it is considered that two measures are possible from the viewpoint of the CR described above.
For example, when the repetition frequency of a laser operating at 6000 Hz is increased to 12000 kHz and the stability of the laser energy is to be secured, the gas flow rate v is doubled or the discharge width is halved, so that CR is reduced. It can be 2-3. When making a laser practical, the former of these two measures is extremely difficult. This is because doubling the gas flow rate increases the power required for the motor driving the fan that generates the gas flow by an factor of eight. Assuming from the power of lasers currently in practical use, the power required for the motor is required to be 20 kW or more. This is power comparable to the discharge input for laser oscillation, and is not in an acceptable range in practical use from the viewpoint of efficiency.

Therefore, in order to realize a high repetition rate laser of 10 kHz or more, it is practical to keep the power of the motor within the practical range, and narrow the discharge width to within the necessary range. Actually, it is effective to narrow the discharge, that is, to narrow the discharge width, in order to increase the repetition rate and narrow the bandwidth of the oscillator chamber.
In order to realize this, an oscillator chamber having a spectral purity of 0.2 pm or less and a repetitive operation of 10 kHz or more can be realized by keeping the electrode width and the distance between the electrodes within appropriate ranges.
As described above, if the discharge stability for the high repetition rate of the laser can be realized, as described in the prior art, the exhaust heat due to the loss and the extension of the electrode life are required.

First, let's start with the problem of exhaust heat. The first method described in the prior art, the jacket type cooling method using a refrigerant such as water, has a feature that it is difficult to cool on the high voltage side. This is because the refrigerant generally has electrical conductivity, so if a jacket is installed on the high voltage side, electrical insulation cannot be ensured and the laser cannot operate.
For example, Patent Document 5 states that this problem can be solved by using an insulator as the refrigerant used in the cooling mechanism.
However, since metal is used for the cooling pipe in the cooling mechanism, the electrical insulation is weak at the connection point of the pipe.
Moreover, in order to ensure electrical insulation, it is necessary to lengthen the creeping distance of the insulator. However, when using a metal pipe, there is a drawback that the size of the apparatus becomes large. For this reason, there is a risk that cooling on the high voltage side involves risks and the structure becomes complicated.

Therefore, usually only the ground side cooling, which is a highly reliable cooling method, is performed. The problem in this case is that the cooling efficiency is poor because cooling from the high voltage side is impossible.
This is because the ground side and the high voltage side are electrically insulated by a peaking capacitor, and thus basically have poor thermal conductivity. In general, a dielectric, that is, an insulator used for a peaking capacitor is made for the purpose of increasing the capacitance per unit volume, and its thermal conductivity is low. Therefore, there is a problem that the high voltage side through the peaking capacitor and the capacitor itself cannot be efficiently cooled.

The problem with the forced air cooling method, which lowers the temperature in the ambient air, which is the second method, is that the cooling efficiency is poor.
In order to improve the cooling efficiency, there is a method of efficiently exchanging heat by attaching fins etc. to the heat generating part, but basically it is cooling by gas, and generally the thermal conductivity of gas is small, so efficiency Is bad.
Furthermore, it is indispensable to increase the surface area of fins to enhance the heat dissipation effect in the high-voltage part, but it is indispensable to increase the efficiency. Increased vulnerability to electrical insulation against
That is, in order to increase the surface area of the tips of the radiating fins in order to enhance the heat radiating effect, a structure with a pointed tip such as a needle must be formed. However, such a structure causes concentration of an electric field when a high voltage is applied, and the insulation performance deteriorates. Accordingly, it is difficult to achieve both heat radiation optimization and insulation. Further, as another method for improving the efficiency, there is a forced circulation by a fan, but there is a drawback that the structure becomes complicated. Further, helium gas may be used as a gas having good thermal conductivity, but the structure becomes complicated because it is necessary to improve the airtightness of the upper portion of the chamber.

As a third method, there is a method in which the peaking capacitor portion is immersed in insulating oil or a fluorine-based inert liquid to cool and circulate the liquid. The advantage of this method is that both the high voltage side and the ground side can be efficiently cooled. Further, the disadvantages of this method are that the structure is complicated, the weight of the apparatus is increased, and the maintainability is inferior.
First, in the case where insulating oil that has been frequently used as a liquid is used as a liquid, considering the high viscosity of the insulating oil and attempting to generate convection efficiently, the size of the tank that contains the oil increases. The size is also increased when forced circulation is performed using a fan or the like. When a fluorine-based inert liquid is used, the disadvantage of using insulating oil can be alleviated due to its low viscosity, but there is a concern that the liquid may leak or adversely affect the environment during maintenance. Therefore, a more efficient cooling method in the vicinity of the peaking capacitor is required to achieve a repetition frequency higher than the repetition frequency currently in practical use.

As described in the prior art, an increase in the temperature of the peaking capacitor results in a decrease in the capacitance of the peaking capacitor, causing a deviation from the optimum design capacitance, resulting in a loss of laser performance. As a result, the stability of the laser output is adversely affected.
In addition, the excimer laser electrode is affected by corrosion caused by halogen gas, which is corrosive gas.
The deterioration of the electrode is mainly caused by gas sputtering at the cathode and corrosion by halogen gas at the anode. If the temperature of sputtering or corrosion is high, the reaction rate increases and the electrode life is shortened. As shown in FIG. 5, the cathode electrode is electrically connected to the peaking capacitor. Since it is an electrical circuit, its electrical resistance is designed to be minimal, and therefore the thermal resistance cannot be reduced too much. Accordingly, the heat of the peaking capacitor easily flows to the electrode side, and the electrode is heated and the temperature rises. Therefore, the heat generation of the capacitor during the high repetition operation may shorten the electrode life.
For the reasons described above, efficient cooling of the peaking capacitor and the high voltage application portion becomes a major issue during high repetition operation.
The present invention has been made in view of the above circumstances, and it is an object of the present invention to provide a cooling mechanism for a discharge-excited gas laser device that can efficiently cool a peaking capacitor and a high voltage application portion and can be highly repeated. .

In order to solve the above problems, in the present invention, a cooling means is provided on the ground-side conductive member, and the ground-side cooling means and the high voltage supply member are connected by a ceramic member having high insulation and high thermal conductivity such as aluminum nitride. By doing so, the high voltage supply portion is cooled, and further the peaking capacitor is cooled.
That is, this invention solves the said subject as follows.
(1) The high voltage side of the peaking capacitor is connected to the high voltage side electrode via a high voltage supply member, and the high voltage supply member is attached to a laser chamber at ground potential via an insulating material. In the discharge excitation gas laser apparatus, which is attached to the laser chamber via a ground side conductive member provided in the laser chamber of the ground potential and the ground side conductive member is provided with cooling means, the insulating material exceeds aluminum oxide. A material having thermal conductivity and an insulation performance equal to or better than aluminum oxide is used.
(2) The high voltage side of the peaking capacitor is connected to the high voltage side electrode via the high voltage supply member, and the low voltage side of the peaking capacitor is attached to the ground side conductive member provided in the laser chamber at the ground potential. In the discharge excitation gas laser apparatus in which the member is provided with cooling means, an insulating material that is in thermal contact with the high voltage supply member and the cooling member is provided between the high voltage supply member and the ground side conductive member, and the insulation As a member, a material having a thermal conductivity exceeding that of aluminum oxide and having an insulation performance equal to or higher than that of aluminum oxide is used.
(3) In the above (2), fins are provided on the insulating member that is in thermal contact with the high voltage supply member and the cooling member so as to have a large contact area with the atmospheric gas and enhance the air cooling effect.
(4) In the above (1) and (2), the insulating material is aluminum nitride or silicon carbide.

In the present invention, the following effects can be obtained.
(1) Since the cooling means on the ground side and the high voltage supply member are connected by an insulating member having high insulation and high thermal conductivity, the high voltage supply member is cooled by heat conduction. Can be effectively cooled. Further, it is not necessary to provide a cooling means for the high voltage supply member, and the insulation performance is not deteriorated.
(2) By providing the insulating member with a large contact area with the atmospheric gas and providing fins for enhancing the air cooling effect, the high voltage supply portion and the peaking capacitor can be cooled more effectively.
(3) By using aluminum nitride or silicon carbide as the insulating material, the necessary thermal conductivity can be ensured while maintaining the mechanical strength, and effective cooling is possible.

FIG. 1 is a diagram showing a first embodiment of the present invention. FIG. 1 shows a partial cross-sectional view of a discharge-excited gas laser device cut along a plane perpendicular to the longitudinal direction of the electrode. FIG.
The laser chamber 1 is provided with an anode and cathode electrodes 2a and 2b for exciting the discharge, and a preionization mechanism 7 for generating the discharge uniformly and efficiently. A cross-section (not shown) is provided in the vicinity of the anode electrode 2a. A wind guide 9 for efficiently flowing the laser gas flowing by the flow fan between the electrodes is arranged.
The cathode electrode 2b to which a negative high voltage is applied is attached to the laser chamber 1 having a ground potential via an insulating material 5 made of ceramics or the like.
On the laser chamber 1, a high voltage supply member 3, a peaking capacitor 4, and an energization member 8 are attached, and the high voltage supply member 3 is electrically connected to the cathode electrode 2 b and one terminal of the peaking capacitor 4. . Further, the other terminal of the peaking capacitor 4 is connected to the laser chamber 1 having a ground potential via an energizing member 8.
The current-carrying member 8 is provided with a metal water-cooling jacket 6. The water cooling jacket 6 is provided with a flow path for flowing a cooling liquid, for example, and the cooling liquid is supplied to the flow path to cool the grounding side of the peaking capacitor 6 through the energizing member 8.

In this embodiment, as the insulating material 5 provided between the cathode electrode 2b and the laser chamber 1, a ceramic member having high insulation and high thermal conductivity, such as aluminum nitride, described later is used. Then, heat is released to the laser chamber 1 side through the insulating material 5 to cool the high voltage supply member 3 and further the peaking capacitor 4 is cooled.
Although it is most desirable from the viewpoint of cooling efficiency to add a water cooling mechanism using a coolant such as water to a high voltage portion such as the high voltage supply member 3, the structure becomes complicated, such as the handling of a cooling pipe. In addition, unless a highly reliable cooling pipe is used, there is a possibility that the refrigerant leaks to the high voltage part when the cooling pipe is broken, and the reliability of the apparatus is lowered.
On the other hand, as described above, a ceramic member having high insulation and high thermal conductivity is used as the insulating material 5, the current-carrying member on the ground side is cooled, and the heat of the high voltage portion is transferred by heat conduction through the insulating material 5. Is configured to escape to the cooled grounding side, the peaking capacitor can be cooled with an extremely simple configuration, and the reliability is not lowered.

FIG. 2 is a diagram showing a second embodiment of the present invention. Like FIG. 1, FIG. 2 shows a partial sectional view of the discharge excitation gas laser device taken along a plane perpendicular to the longitudinal direction of the electrodes. A sectional view of the vicinity of the electrode is shown.
As described with reference to FIG. 1, the laser chamber 1 is provided with anodes and cathode electrodes 2a and 2b and a preionization mechanism 7 for exciting the discharge. In the vicinity of the anode electrode 2a, laser gas is interposed between the electrodes. A wind guide 9 for efficiently flowing is arranged.
The cathode electrode 2b is attached to the laser chamber 1 having the ground potential via an insulating material 5. As the insulating material 5, the ceramic member having high insulation and high thermal conductivity shown in the first embodiment can be used.
As described above, the high voltage supply member 3, the peaking capacitor 4, and the energization member 8 are mounted on the laser chamber 1. The high voltage supply member 3 is electrically connected to the cathode electrode 2b and one terminal of the peaking capacitor 4. It is connected. Further, the other terminal of the peaking capacitor 4 is connected to the laser chamber 1 having a ground potential via an energizing member 8. The current-carrying member 8 is provided with a metal water-cooling jacket 6.

In the present embodiment, a ceramic member 11 having high insulation and high thermal conductivity described later is provided between the high voltage supply member 3 and the water cooling jacket 6.
By connecting the water cooling jacket 6 and the high voltage supply unit 3 with a ceramic member 11 having high insulation and high thermal conductivity, the high voltage supply member 3 can be cooled, and further the peaking capacitor 4 can be cooled.
As shown in FIG. 2, if the surface of the ceramic member 11 is uneven, it is also effective to increase the creepage distance between the high voltage side and the ground side to improve the insulation performance. Further, since the surface area can be increased and function as fins, the cooling effect can be enhanced.

FIG. 3 is a diagram showing a third embodiment of the present invention. In FIG. 3, a portion where a high voltage supply member 3, a peaking capacitor 4, a water cooling jacket 6, an energizing member 8 and the like of a discharge excitation gas laser device are attached is shown. The figure seen from diagonally upward is shown.
In general, a discharge excitation gas laser device such as an excimer laser requires a long electrode length (300 to 700 mm) for laser oscillation. It is desirable that the peaking capacitors 4 be arranged uniformly within the length. However, the size of the capacitor is determined by the dielectric constant. If the capacitors are evenly arranged, in most cases, there is a gap where the capacitors are not arranged.
Therefore, in this embodiment, for example, a cylindrical ceramic member 12 is disposed in the gap as shown in FIG. With this configuration, it is possible to cool the high voltage side with a structure that is almost the same as the structure of the current excimer laser. That is, the heat of the high voltage supply unit 3 can be released to the low pressure side where the water cooling jacket 6 is provided via the ceramic member 12, the high voltage supply member 3 is cooled, and the peaking capacitor 4 is further cooled. can do.
Even in this structure, it is possible to increase the creepage distance and enhance the insulation performance by making the ceramic member uneven as in the structure of FIG. Further, by providing unevenness, the surface area can be increased, and a heat dissipation effect to the air can be obtained.

As described above, it is most desirable from the viewpoint of cooling efficiency to add a water cooling mechanism using a refrigerant such as water to a high voltage portion such as the high voltage supply member 3. In addition to the complexity of the structure, if a highly reliable cooling pipe is not used, there is a possibility that the refrigerant leaks to the high voltage portion when the cooling pipe is broken, and the reliability of the apparatus is lowered.
On the other hand, as shown in FIGS. 2 and 3, by using a ceramic member having high insulation and high thermal conductivity, heat can be efficiently released to the laser chamber 1 side as in the first embodiment. With this configuration, the peaking capacitor can be cooled with an extremely simple configuration, and the reliability is not lowered.

In FIGS. 2 and 3, the ceramic member 11 or 12 is in contact with the water-cooling jacket on the ground side, but can be disposed without contacting this portion. Furthermore, if the heat radiation performance to the air is improved by attaching a heat radiation fin or the like to the ceramic member to increase the area, the air cooling performance can be further increased.
If the structure as described above is realized by using a metal member, the insulation performance is impaired. However, by using an insulation member made of a ceramic member having high thermal conductivity, the high voltage side can be obtained without impairing the electrical insulation performance. Efficient cooling is possible.
Further, by appropriately using the cooling means shown in FIG. 1, FIG. 2, and FIG. 3, the high voltage side can be cooled more effectively than when one cooling means is used.

As the insulating material 5 and the ceramic members 11 and 12, various members having high insulation and high thermal conductivity described below can be used.
FIG. 4 shows the thermal conductivity of various materials. In general, copper is used as a metal used in temperature control devices such as a heat sink and the like. Therefore, ideally, ceramics having the same thermal conductivity as copper and electrical insulation are desirable. However, at present, the only common ceramic material having a thermal conductivity exceeding that of copper is carbon.

Although not shown in FIG. 4, industrial diamond is the most suitable material for the purpose of this patent as a carbon-based material. As various characteristics of industrial diamonds available as commercial products, the thermal conductivity is 2000 W / m · K, and the specific resistance representing the characteristic of insulation resistance is 1014 Ω · m. However, diamond has problems in cost, workability, and manufacturable dimensions.
Carbon fibers, graphite sheets and the like having a thermal conductivity of 800 W / m · K are highly available. However, the graphite system has poor electrical insulation and, depending on the material, exhibits conductivity, so it cannot be used in this patent. Therefore, as carbon-based ceramics, diamond that becomes a single crystal can be used depending on the part of use, but not all of them can be used.

On the other hand, the ceramic material shown in FIG. 4 is practical because it has relatively good workability and availability. Taking these into consideration, we consider the extent to which thermal conductivity is desirable.
Stainless steel (SUS) is a material that is generally avoided when a metal is thermally designed. FIG. 4 shows the thermal conductivity of SUS304, which is a kind of SUS. The thermal conductivity of SUS304 is 16 W / m · K. Therefore, the thermal conductivity of the ceramic used in this patent is required to be at least larger than that of the SUS material.

Conventionally, aluminum oxide, so-called alumina, is often used as the insulating material 5 shown in FIG. 1, and the high voltage application portion is thermally connected to the chamber housing through the alumina. Therefore, it can be said that the high-voltage side of the chamber in the current excimer laser is exhausted from the high-voltage side if it is not sufficient.
This is because during the laser operation, the temperature of the housing is from room temperature to about 70 ° C. due to the temperature control of the laser gas, etc., but at least the temperature is lower than the high voltage application portion. Therefore, in order to improve the performance of the excimer laser in which the current excimer laser chamber is used, or to stabilize the excimer laser even when a higher amount of heat is applied, at least a material having a thermal conductivity equal to or higher than that of alumina is preferable. it is conceivable that.

From the above, it can be said from FIG. 4 that desirable ceramic materials are beryllium oxide, aluminum nitride, silicon, silicon carbide (silicon), sapphire (single crystal alumina), and alumina. Among these, beryllium oxide is inferior in practicality because of its toxicity.
Moreover, since silicon is easy to break, handling requires a little care. Moreover, although silicon carbide may have inferior insulation performance depending on its purity, it has a thermal conductivity comparable to that of aluminum nitride.
Therefore, it is desirable to use an insulating material having a thermal conductivity exceeding that of aluminum oxide and having an insulation performance equal to or higher than that of aluminum oxide, and among these, aluminum nitride and silicon carbide (silicon) are the most desirable. It is.
For example, in FIG. 4, the thermal conductivity of aluminum nitride is shown as 162 W / m · K, but this value is an average of various grades and exceeds 230 W / m · K at the prototype level. There are also things.

It is a figure which shows the 1st Example of this invention. It is a figure which shows the 2nd Example of this invention. It is a figure which shows the 3rd Example of this invention. It is a figure which shows the heat conductivity of various materials. It is sectional drawing of the laser chamber of a general discharge excitation laser. It is a schematic diagram of normal discharge and abnormal discharge.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Laser chamber 2a Anode electrode 2b Cathode electrode 3 High voltage supply member 4 Peaking capacitor 5 Insulating material 6 Water cooling jacket 7 Pre-ionization mechanism 8 Current supply member 9 Wind guide 11, 12 Ceramic member

Claims (4)

The high voltage side of the peaking capacitor is connected to the electrode on the high voltage side via a high voltage supply member, and the high voltage supply member is attached to the laser chamber at ground potential via an insulating material,
A cooling mechanism of a discharge excitation gas laser device in which a low-voltage side of a peaking capacitor is attached to the laser chamber via a ground-side conductive member provided in a laser chamber of ground potential, and a cooling means is provided on the ground-side conductive member. ,
A cooling mechanism for a discharge-excited gas laser device, wherein the insulating material has a thermal conductivity exceeding that of aluminum oxide and has an insulating performance equal to or higher than that of aluminum oxide. The high voltage side of the peaking capacitor is connected to the high voltage side electrode via the high voltage supply member, and the low voltage side of the peaking capacitor is attached to the ground side conductive member provided in the laser chamber at the ground potential and cooled to the ground side conductive member. A cooling mechanism of a discharge excitation gas laser device provided with a means,
An insulating material that is in thermal contact with the high voltage supply member and the cooling member is provided between the high voltage supply member and the ground-side conductive member, and the insulation member has a thermal conductivity that exceeds aluminum oxide, A cooling mechanism of a discharge excitation gas laser device characterized by having an insulation performance equal to or better than aluminum oxide. The insulating member that is in thermal contact with the high-voltage supply member and the cooling member is provided with fins that have a large contact area with the atmospheric gas and enhance the air-cooling effect. Cooling mechanism of the discharge excitation gas laser apparatus. 4. The cooling mechanism for a discharge-excited gas laser device according to claim 1, wherein the insulating material is aluminum nitride or silicon carbide.

Images