RU2298271C2 - Line-selectable double-chamber f2 laser system - Google Patents

Abstract

FIELD: laser engineering; narrow-band double-discharge gas lasers used as light sources for integrated-circuit lithography.

SUBSTANCE: two separate gas-discharge chambers are provided of which one chamber is part of master oscillator producing extremely narrow-band seed beam amplified in other discharge chamber. Each chamber can be separately controlled and has one tangential blower affording sufficient gas flow enabling operation at pulse repetition frequency of 4000 pulses per second or more due to removal of discharge products from discharge area within pulse-to-pulse time interval shorter than approximately 0.25 milliseconds. Master oscillator is provided with line choice unit enabling selection of most intensive spectral line F₂.

EFFECT: improved design and ability of regulating beam parameters, including pulse wavelength and energy.

154 cl, 74 dwg

Description

Gas discharge lasers

Gas discharge lasers are well known and appeared shortly after the invention of the laser in the 1960s. A high voltage discharge between the two electrodes excites the laser gas, creating a gaseous active medium. The resonator, in which the active medium is located, provides stimulated amplification of light, which is then removed from the resonator in the form of a laser beam. Many of these discharge lasers are pulsed.

Excimer Lasers

Excimer lasers are a special type of gas discharge lasers and have been known since the mid-1970s. An excimer laser intended for integrated circuit lithography is described in US Pat. No. 5,023,884, issued June 11, 1991, to the invention of the "Compact Excimer Laser." This patent is incorporated herein by reference. The excimer laser described in the '884 patent is a pulsed laser with a high repetition rate. These excimer lasers, when used in integrated circuit lithography, usually operate around the clock around the clock, producing thousands of valuable integrated circuits per hour, so their downtime can be very expensive. For this reason, most of the components of these lasers are organized in the form of modules that can be replaced in a few minutes. The bandwidth of the output beam of excimer lasers used in lithography should be reduced to a fraction of a picometer. Discharge lasers of the type '884 patent use a pulsed power system to create electrical discharges between two elongated electrodes. In such known systems, a direct current source charges the capacitor bank, the so-called "charging capacitor" or "C ₀ ", to a predetermined regulated voltage, the so-called "charging voltage", for each pulse. In known devices, the magnitude of this charging voltage can be in the range of about 500-1000 volts. After charging C ₀ to a predetermined voltage, the semiconductor switch closes, and this allows the electric energy stored in C ₀ to oscillate very quickly through a sequence of magnetic compression circuits and a voltage transformer to generate a high-voltage electric potential of about 16,000 volts (or higher) on electrodes that create discharges lasting about 20-50 ns.

Major advances in light sources for lithography

For the period 1989-2001 excimer lasers, like the laser described in the '884 patent, have become the main light source for integrated circuit lithography. Currently, more than 1000 of these lasers are used in the most modern enterprises for the production of integrated circuits. Almost all of these lasers have the basic design features described in the '884 patent. They include the following:

(1) one pulsed power system that provides electrical pulses on the electrodes with a frequency of about 100-2500 pulses per second;

(2) one resonator, consisting of an output coupler of the type of a partially reflecting mirror, and a line narrowing unit, consisting of a prism beam expander, a rotary mirror, and a diffraction grating;

(3) one discharge chamber containing a laser gas (KrF or ArF), two elongated electrodes, and a tangential fan to allow the laser gas to circulate sufficiently quickly between the two electrodes to clear the discharge region between pulses, and

(4) a beam monitor for monitoring the pulse energy, wavelength and bandwidth of the output pulses by a feedback control system for controlling the pulse energy, dose of energy and wavelength of each subsequent pulse.

During 1989-2001 the output power of such lasers gradually increased, and the requirements for the quality of the beam to ensure stability of the beam energy, stability of the wavelength and bandwidth became more stringent. The operating parameters of the popular lithographic laser model, widely used in the manufacture of integrated circuits, are as follows: pulse energy 8 mJ, frequency 2500 pulses per second (with an average beam power of up to about 20 watts), the bandwidth calculated from the full width of the distribution curve at half maximum ), about 0.5 pm and the stability of the pulse energy +/- 0.35%.

F ₂ lasers

F ₂ lasers are well known in the art. These lasers are similar to KrF and ArF lasers. The main difference is in the gas mixture, which in the F ₂ laser consists of a small fraction of F ₂ with helium and / or neon as a buffer gas. The natural output spectrum of the F ₂ laser is concentrated in two spectral lines with a narrow bandwidth, with a relatively strong line being centered around 157.63 nm, and a relatively weak line centered around 157.52 nm.

Bandwidth F ₂ lasers

A typical KrF laser has a natural bandwidth calculated from the full width of the distribution curve at half maximum (FWHM), about 300 pm and centered at about 248 nm, and for use in lithography the line is usually narrowed to less than 0.6 pm. (In this description, the bandwidth values will refer to the bandwidth of the UWB, unless otherwise indicated). ArF lasers have a natural bandwidth of about 500 centered at approximately 193 nm, usually with a narrowing of the line to less than 0.5 pm. These lasers can be relatively easily tuned over much of their natural bandwidth using the diffraction grating line narrowing module mentioned above. F ₂ lasers, as noted above, traditionally generate laser beams, most of whose energy is in two narrow spectral elements (sometimes called "spectral lines"), centered at a wavelength of about 157.63 nm and 157.52 nm. Often the less intense of these two spectral lines (i.e., the 157.52 nm line) is suppressed, and the laser is forced to work on the 157.63 nm line. The natural line width of the 157.63 nm line depends on the pressure and gas content and ranges from about 0.6 to 1.2 pm (FWHM). A F ₂ laser with a bandwidth in this interval can be used with lithographic devices with a mirror lens, which uses both refractive and reflective optical elements, but for an all-refractive lens, the laser beam bandwidth may need to be reduced to about 0.1 pm for obtaining the required results.

Injection seed

A well-known method for reducing the bandwidth of gas-discharge laser systems (including excimer laser systems) is the injection of a narrow-band "seed" beam into the active medium. In one such system, a seed-beam generating laser called a “master oscillator” is intended to generate a beam with a very narrow bandwidth in the first active medium, and this beam is used as a seed beam in the second active medium. If the second active medium acts as a power amplifier, then such a system is called the "master oscillator-power amplifier" (PGM) system. If the second active medium itself has a resonator (in which laser oscillations occur), then the system is called the “injection seed generator” (GIZ) system or the “master oscillator-power generator” (ZGGM) system, in this case the seed laser is called the master oscillator, and the system located after it is called a power generator. Laser systems consisting of two separate systems usually have a higher price and larger dimensions; they are more difficult to build and operate than comparable single-chamber laser systems. Therefore, the commercial use of such two-chamber laser systems has so far been limited.

There is a need to improve the design of a pulsed gas-discharge F ₂ laser for operation with a repetition rate of the order of 4000 pulses per second or higher, in which all beam quality parameters, including wavelength and pulse energy, could be precisely controlled.

Summary of the invention

A modular injection-discharge gas discharge laser system capable of generating high-quality pulsed laser beams with a frequency of about 4000 Hz or higher and a pulse energy of about 5-10 mJ or higher for an integrated power of about 20-40 W or higher is proposed. Two discharge chambers are provided, one of which is part of a master oscillator generating a very narrow-band seed beam, which is amplified in the second discharge chamber. The cameras can be controlled separately, which allows optimizing the wavelength parameters in the master oscillator and the pulse energy parameters in the amplification chamber. A preferred embodiment is an F ₂ laser system with a ZGUM configuration and is specifically designed to be used as a light source for lithography integrated circuits. In this preferred embodiment, both the cameras and the laser optical system are mounted on a vertical optical table in the laser housing. In this preferred embodiment of the ZGUM system, each chamber contains a tangential fan that generates a sufficient gas flow to operate at a frequency of 4000 Hz or higher by cleaning the discharge products from the discharge region in a time less than about 0.25 milliseconds between pulses. The master oscillator is equipped with a line selection unit for selecting the strongest spectral line F ₂ . This preferred embodiment also includes a pulse multiplier module that divides each pulse from the power amplifier into two or four pulses in order to substantially reduce the rate of destruction of lithographic optics. In preferred embodiments of the invention, a “three wavelength platform” is used. It includes a closed optical table and a general layout of equipment, the same for all three types of discharge laser systems, which are supposed to be used in the production of integrated circuits in the early 21st century, i.e. for KrF, ArF and F ₂ lasers.

Thus, according to the claimed invention, there is provided a system of a very narrow-band double-chamber gas-discharge F ₂ laser with a high repetition rate, comprising

A) a first laser unit containing

1) the first discharge chamber accommodating

a) the first laser gas,

b) a first pair of elongated spaced apart electrodes forming a first discharge region,

2) the first fan to ensure sufficient speeds of the first laser gas in the first region of the discharge to remove from the first region of the discharge after each pulse essentially all of the ions formed as a result of the discharge before the next pulse when operating with a repetition rate of 4000 pulses per second or higher,

3) a first heat exchanger system configured to remove at least 16 kW of thermal energy from the first laser gas,

B) a line selection unit for minimizing energy outside the spectrum of one selected line,

C) a second laser unit containing

1) a second discharge chamber containing

a) a second laser gas,

b) a second pair of elongated spaced apart electrodes forming a second discharge region,

2) a second fan to ensure sufficient speeds of the second laser gas in the second region of the discharge to remove from the second region of the discharge after each pulse essentially all of the ions formed as a result of the discharge before the next pulse when operating at a repetition rate of 4000 pulses per second or higher,

3) a second heat exchanger system, configured to remove at least 16 kW of thermal energy from the second laser gas,

D) a pulsed power system configured to supply electric pulses to the first pair of electrodes and the second pair of electrodes, sufficient to generate laser pulses with a frequency of about 4000 pulses per second with precisely controlled pulse energy above about 5 mJ,

E) a laser beam measurement and control system for measuring the energy of the output laser pulses generated by the two-chamber laser system and for controlling the output laser pulses in a feedback control device, the output laser beams from the first laser unit being used as a seed beam to seed the second laser unit .

In addition, in the specified laser system, the first laser unit is made in the form of a master oscillator, and the second laser unit is made in the form of a power amplifier.

Moreover, in the specified laser system, the first laser gas contains fluorine and neon, or may contain fluorine and helium.

In addition, the first and second laser gas may contain fluorine and a buffer gas selected from the group consisting of neon, helium, or a mixture of neon and helium.

In this laser system, the power amplifier is configured to pass a beam through the second discharge region one time, while the power amplifier is configured to have multiple passes through the second discharge region.

In addition, in the specified laser system, the master oscillator contains optical components that form a resonant path for two passes through the first discharge region, while the master oscillator contains optical components that provide a resonant path for two passes through the first discharge region, and in which the power amplifier contains optical components that provide multiple passages of the beam through the second region of the discharge.

In addition, in said laser system, said first laser unit comprises an optical cavity system, and said laser system further comprises an optical table for holding a resonator optical system of the first laser unit independently of the first discharge chamber.

In this case, the optical table in the laser system has a U-shape and forms a U-shaped cavity in which the first discharge chamber is installed.

The laser system also further comprises a vertically mounted optical table with first and second discharge cameras mounted thereon.

In addition, each of the first and second laser chambers in the laser system forms a gas flow path with a gradually increasing cross section behind the electrodes, which allows to restore a large percentage of the static pressure drop that occurs in the discharge region.

Each of the first and second chambers of the laser system contains a blade structure behind the discharge region to normalize the gas velocity after the discharge region.

In a laser system, the first fan and the second fan are tangential fans and each contains a shaft driven by two brushless DC motors, the motors being water-cooled motors.

In addition, each of the motors contains a stator and each of the motors contains a magnetic rotor enclosed in a pressure cap separating the stator from the laser gas.

Moreover, the first and second fans are tangential fans, each of which contains a blade structure made on a machine made of one part of aluminum raw materials, and said blade structure has an outer diameter of about five inches, while the blade structure contains blade elements having sharp leading edges.

Engines in the laser system do not have sensors, and said laser system further comprises a master motor controller for controlling one of the motors and a slave motor controller for controlling the other motor.

In addition, in the laser system, each tangential fan contains blades located at an angle to the shaft, and each heat exchanger system is cooled by water, with each heat exchanger system containing at least four separate water-cooled heat exchangers.

Moreover, each heat exchanger system contains at least one heat exchanger having a tubular channel for the flow of water, in the path of which at least one turbulator is located.

In addition, in the system, each of the four heat exchangers contains a tubular channel for the flow of water with a turbulator located in it.

In the laser system of the invention, the pulsed power system comprises water-cooled electrical elements, wherein at least one of the water-cooled elements operates at high voltages above 12,000 volts, the high voltage being isolated from earth by means of an inductor through which cooling water flows.

In the laser system of the invention, the pulse power system comprises a first battery of charging capacitors and a first pulse compression circuit for supplying electrical pulses to the first pair of electrodes, and a second battery of charging capacitors and a second pulse compression circuit for supplying electrical pulses to the second pair of electrodes, and a resonant charging system for charging parallel to the first and second batteries of charging capacitors to a precisely controlled voltage, while the resonant charging system contains a De-Qing circuit, chrome Moreover, the resonant charging system may comprise a bleed circuit or a De-Qing circuit and a bleed circuit.

In addition, in the laser system, the pulsed power system contains a charging system consisting of at least three power sources arranged in parallel, and the selection unit is located after the master oscillator, while the line selection unit contains many prisms, and many prisms consists of five prisms.

A plurality of prisms can be arranged in a loop to allow the laser beams to rotate 360 ° from the first laser unit 360 ° before entering the second laser unit.

In addition, the laser system further comprises a visible light laser-sighting device, and the line selection unit contains a Lyo filter, in addition, the first discharge chamber and the second discharge chamber contain windows arranged so that all the angles of incidence of the laser beams on said windows are greater than the Brewster angle .

In the laser system, a beam control means is additionally provided for controlling the laser beams generated in the first laser unit, the control means comprising means for rotating the optical element, and the beam control means further comprising means for adjusting the pressure in the line selection unit.

The laser system also includes a prismatic output coupler, partially forming a resonator for the first laser unit, the prismatic output coupler containing two surfaces, the first of which is oriented at an angle with low loss for p-polarization, and the second is located orthogonal to the laser beams from the first laser unit, and further comprises A) a first temperature monitor for monitoring the temperature of the gas in the first discharge chamber, B) a first gas temperature control system comprising s algorithm for adjusting gas temperature to avoid adverse acoustic effects resulting from reflected acoustic waves.

In addition, the laser system further comprises A) a second temperature monitor for monitoring the gas temperature in the second discharge chamber, B) a second gas temperature control system containing a control algorithm for adjusting the gas temperature to eliminate negative acoustic effects caused by reflected acoustic waves, and further contains a nitrogen filter, a nitrogen purge system containing a purge module with flow monitors, while the laser also contains tubes for transporting waste of the main purge gas from the laser, as well as a shutter unit containing an electric shutter and a power meter that can be positioned along the path of the output laser beam by the control signal, further comprises a beam shielding system containing A) at least one beam seal, containing a metal bellows; and B) a purge means for purging the beam enclosure with purge gas.

Moreover, in the laser system according to the invention, the beam shielding means comprises a flow guiding means for generating a purge stream perpendicular to the laser beams generated in the second laser unit.

In addition, in the laser system, the at least one beam seal is configured to easily replace the laser chamber, while the at least one beam seal does not contain an elastomer, provides isolation from camera vibrations, provides isolation of the beam path from atmospheric gases and allows replace the laser chamber without breaking the line selection unit, and the at least one beam seal is comparable to a vacuum.

Moreover, in the laser system, the at least one beam seal is a plurality of beam seals, which are easily sealed bellows seals made with the possibility of easy manual removal.

In the laser system, said measuring and control system comprises a primary beam splitter for separating a small percentage of each output laser pulse from the second laser unit and optical means for directing part of this small percentage to the pulse energy detector, and an insulating means for isolating the volume limited to at least partially, by the primary beam splitter and the window of the pulse energy detector from other parts of the measurement and control system, for the formation of an isolated region.

In addition, the laser system further comprises purge means for purging the insulated region with purge gas.

Moreover, the laser system according to the invention is configured to operate as a KrF laser system, or an ArF laser system, or an F ₂ laser system with minor modifications.

In the claimed laser system, substantially all of the components are enclosed in a laser housing, but the system comprises an AC / DC module physically separate from the laser housing.

In the laser system, the pulsed power system comprises a battery of charging capacitors of the master oscillator, a battery of charging capacitors of the power amplifier and a resonant charger configured to charge both batteries of charging capacitors in parallel, while the pulsed power system comprises a power source configured to supply at least 2000 In power to the resonant charger.

The claimed laser system further comprises a gas control system for controlling the concentration of F ₂ in the first laser gas to control the parameters of the beam of the master oscillator, and further comprises a gas control system for regulating the pressure of the first laser gas to control the parameters of the beam of the master oscillator, in addition further comprises a calculation controller the discharge time to start the charges in the power amplifier, so that they occur 20-60 ns after the discharges in the master oscillator, additional In fact, it contains a discharge controller programmed to cause discharges in some circumstances at a moment in time, which makes it possible to avoid significant energy of the output pulse, while the controller in some circumstances is programmed to cause a discharge in a power amplifier at least 20 ns earlier than a discharge in a master oscillator.

The laser system according to the invention further comprises a pulse multiplier unit for increasing the duration of the laser output pulses, wherein the pulse multiplier unit is adapted to receive the laser output pulse and multiply the number of laser output pulses per second by at least two times to generate one multiplied laser pulse output a beam consisting of a larger number of pulses with significantly lower intensities than the laser output pulses, moreover, the pulse multiplier unit contains (1) a first beam splitter designed to separate a part of the laser output pulse beam, the separated part forming a delayed part, and the laser output pulse beam determines the size and angular divergence of the beam in the first beam splitter, (2) the first delay path, starting and ending in the first beam splitter, the first delay path comprising at least two focusing mirrors configured to focus said delayed part in f in the first path of delay and return of the delayed part to the first beam splitter with a beam size and angular divergence equal to or approximately equal to the beam size and angular divergence of the laser output pulse beam in the first beam splitter, at least two focusing mirrors are spherical mirrors.

The laser system according to the invention further comprises a second delay path comprising at least two spherical mirrors, wherein the first delay path comprises four focusing mirrors.

Moreover, the laser system according to the invention further comprises a second delay path formed by a second beam splitter located on the first delay path.

Furthermore, said first delay path comprises a second beam splitter, the system further comprising a second delay path comprising at least two focusing mirrors arranged to focus the delayed part in focus on the first delay path and return the delayed part to the first beam splitter with the size and angular divergence of the beam equal to or approximately equal to the size and angular divergence of the laser output pulse beam in the first beam splitter, the first splitting The beam is configured to direct the laser beam in at least two directions using the optical property of the truncated internal reflection, and the first beam splitter consists of two transparent optical elements, each of which has a flat surface, both optical elements being arranged so that their first surfaces are parallel to each other and separated by a distance of less than 200 nm.

In addition, in the laser system of the invention, the first beam splitter is an uncoated optical element oriented at an angle to the laser output pulse beam to achieve the desired reflection-transmission ratio.

According to a second aspect of the invention, there is provided a system of a very narrow-band dual-chamber gas-discharge F ₂ laser with a high repetition rate, comprising

A) a first laser unit containing

1) the first discharge chamber accommodating

a) the first laser gas,

b) a first pair of elongated spaced apart electrodes forming a first discharge region,

2) the first fan to ensure sufficient movement of the first laser gas in the first region of the discharge to remove from the first region of the discharge after each gas discharge essentially all of the ions formed as a result of the discharge before the next gas discharge when operating at a repetition rate of 4,000 gas discharges per second or above,

3) a first heat exchanger system for removing heat energy from a first laser gas,

B) a line selection unit for minimizing energy outside the spectrum of one selected line,

C) a second laser unit containing

1) a second discharge chamber containing

a) a second laser gas,

b) a second pair of elongated spaced apart electrodes forming a second discharge region,

2) a second fan to ensure sufficient movement of the second laser gas in the second region of the discharge to remove from the second region of the discharge after each gas discharge essentially all of the ions formed as a result of the discharge before the next gas discharge when operating at a repetition rate of 4,000 gas discharges per second or above,

3) a second heat exchanger system for removing thermal energy from the second laser gas,

D) a pulsed power supply system configured to supply electric pulses to the first pair of electrodes and the second pair of electrodes sufficient to generate laser output pulses with a frequency of about 4000 laser output pulses per second with precisely controlled laser output pulse energies above about 5 mJ,

E) a laser beam measurement and control system for measuring the energy of the laser output pulses generated by the two-chamber laser system and for controlling the laser output pulses in a feedback control device, the output laser beams from the first laser unit being used as a seed beam to seed the second laser unit .

In the specified laser system, the first laser unit is made in the form of a master oscillator, and the second laser unit is made in the form of a power amplifier.

In a laser system, the first laser gas contains fluorine and neon, or the first laser gas contains fluorine and helium.

In a laser system, the first and second laser gas contains fluorine and a buffer gas selected from the group consisting of neon, helium, or a mixture of neon and helium.

In addition, in the laser system, the power amplifier is capable of one beam passage through the second discharge region, while the power amplifier is configured to have multiple passes through the second discharge region, and the master oscillator contains optical components that form a resonant path for making two passes through the first region discharge.

Moreover, in the laser system, the master oscillator contains optical components that provide a resonant path for two passes through the first region of the discharge, and in which the power amplifier contains optical components that provide multiple passes of the beam through the second region of the discharge.

According to a second aspect of the invention, in the laser system, said first laser unit comprises an optical resonator system and said laser system further comprises an optical table for holding the optical system of the resonator of the first laser unit independently of the first discharge chamber, wherein the optical table is U-shaped and forms a U- shaped cavity in which the first discharge chamber is installed.

In addition, the laser system further comprises a vertically mounted optical table with first and second discharge cameras mounted thereon.

In the laser system according to the invention, each of the first and second laser chambers forms a gas flow path with a gradually increasing cross section behind the electrodes, which allows to restore a large percentage of the static pressure drop occurring in the discharge region, each of the first and second chambers containing a blade structure behind the discharge region to normalize the gas velocity after the discharge region.

Moreover, in the laser system, each of the first fan and the second fan is a tangential fan and each contains a shaft driven by two brushless DC motors, and the motors are water-cooled motors, each of the motors contains a stator and each of the motors contains a magnetic rotor enclosed into the pressure cap separating the stator from the laser gas.

In addition, in the laser system, each of the first and second fans is a tangential fan containing a blade structure made on a machine made of one part of aluminum raw materials, while the blade structure has an outer diameter of about five inches and the blade structure contains blade elements having sharp leading edges.

In the laser system, the engines do not have sensors, and said laser system further comprises a master motor controller for controlling one of the engines and a slave motor controller for controlling the other motor.

In the laser system, each tangential fan contains blades located at an angle to the shaft, and each heat exchanger system is cooled by water, and each heat exchanger system contains at least four separate water-cooled heat exchangers.

In addition, in the laser system, each heat exchanger system contains at least one heat exchanger having a tubular channel for the flow of water, in the path of which at least one turbulator is located.

Moreover, each of the four heat exchangers contains a tubular channel for the flow of water with a turbulator located in it.

In a laser system, a pulsed power system contains water-cooled electrical elements, and at least one of the water-cooled elements operates at high voltages above 12,000 volts, while the high voltage is isolated from the ground by means of an inductor through which cooling water flows.

In the laser system of the invention, the pulse power system comprises a first battery of charging capacitors and a first pulse compression circuit for supplying electrical pulses to the first pair of electrodes, and a second battery of charging capacitors and a second pulse compression circuit for supplying electrical pulses to the second pair of electrodes, and a resonant charging system for charging parallel to the first and second batteries of the charging capacitors to a precisely controlled voltage, while the resonant charging system contains a De-Qing circuit.

In a laser system, a resonant charging system comprises a bleed circuit.

In addition, in the laser system, the resonant charging system includes a De-Qing circuit and a bleed circuit.

A pulsed power system contains a charging system consisting of at least three power sources arranged in parallel.

In the laser system, the selection unit is located after the master oscillator, and the line selection unit contains many prisms, with many prisms consisting of five prisms, and many prisms can be arranged in a loop to ensure that the laser beams rotate 360 ° from the first laser unit to the entrance into the second laser unit.

The laser system further comprises a visible light laser-sight.

In addition, in the laser system, the line selection unit contains a Lio filter, and the first discharge chamber and the second discharge chamber contain windows arranged so that all the angles of incidence of the laser beams on said windows are greater than the Brewster angle.

The laser system further comprises a beam control means for controlling laser beams generated in the first laser unit, wherein the control means comprises means for rotating the optical element.

In addition, the beam control means comprises means for adjusting the pressure in the line selection unit.

The laser system comprises a prismatic output coupler partially forming a resonator for the first laser unit, the prismatic output coupler containing two surfaces, the first of which is oriented at an angle with low loss for p-polarization, and the second is located orthogonally to the laser beams from the first laser unit .

In addition, the laser system further comprises A) a first temperature monitor for monitoring the temperature of the gas in the first discharge chamber, B) a first gas temperature control system comprising a control algorithm for adjusting the gas temperature to eliminate negative acoustic effects caused by reflected acoustic waves, and further comprises A) a second temperature monitor for monitoring the temperature of the gas in the second discharge chamber, B) a second gas temperature control system, comprising a control algorithm for adjusting the temperature of the gas in order to eliminate negative acoustic effects caused by reflected acoustic waves, and the nitrogen filter further comprises a nitrogen purge system comprising a purge module with flow monitors, the laser also comprising tubes for conveying the exhaust purge gas from the laser, and the laser system further comprises a shutter unit comprising an electric shutter and a power meter that can be positioned along the exit path laser beam by a control signal.

The laser system further comprises a beam shielding system comprising A) at least one beam seal containing a metal bellows, and B) purge means for purging the beam shielding with the purge gas, the beam shielding means comprising flow guiding means for generating a purge stream perpendicular to the laser beams generated in the second laser unit.

In the laser system, said at least one beam seal is capable of easily replacing a laser chamber, wherein said at least one beam seal does not contain elastomers, provides isolation from camera vibrations, provides isolation of the beam path from atmospheric gases, and allows the laser to be easily replaced the chamber without disturbing the line selection unit, wherein said at least one beam seal is comparable to a vacuum, while said at least one beam seal may represent a lot of beam seals, which use easily sealed bellows seals made with the possibility of easy removal by hand.

In the laser system, said measurement and control system comprises a primary beam splitter for separating a small percentage of each output laser pulse from the second laser unit, and optical means for directing a portion of this small percentage to the pulse energy detector, and an insulating means for isolating a volume limited to at least partially, by the primary beam splitter and the window of the pulse energy detector from other parts of the measurement and control system, to form an isolated region.

The laser system according to the invention further comprises purge means for purging the insulated region with purge gas.

The laser system is configured to operate as a KrF laser system, or an ArF laser system, or an F ₂ laser system with minor modifications.

Moreover, in the laser system, essentially all components are enclosed in the laser housing, but the system contains an AC / DC module physically separate from the laser housing.

In addition, in the laser system, the pulsed power system comprises a battery of charging capacitors of the master oscillator, a battery of charging capacitors of the power amplifier and a resonant charger configured to charge both batteries of charging capacitors in parallel, while the pulsed power system comprises a power source configured to supply at least 2000 V power to a resonant charger.

The laser system further comprises a gas control system for controlling the concentration of F ₂ in the first laser gas to control the parameters of the master oscillator beam, and further comprises a gas control system for regulating the pressure of the first laser gas to control the parameters of the master oscillator beam, a discharge time calculation controller for triggering charges in the power amplifier, so that they occur in 20-60 ns after the discharges in the master oscillator, and also additionally contains a discharge controller, programmed to cause in some circumstances discharges at a point in time, avoiding significant energy output pulse.

Moreover, in the laser system, the controller in some circumstances is programmed to cause a discharge in the power amplifier at least 20 ns earlier than the discharge in the master oscillator.

In addition, the laser system further comprises a pulse multiplier unit for increasing the duration of the laser output pulses, wherein the pulse multiplier unit is configured to receive the laser output pulse beam and multiply the number of pulses per second by at least two times to generate one multiplied laser pulse output a beam consisting of a larger number of laser output pulses with significantly lower values of intensity than laser output pulses, moreover, the pulse multiplier unit contains (1) a first beam splitter designed to separate a part of the laser output pulse beam, the separated part forming a delayed part, and the laser output pulse beam determines the size and angular divergence of the beam in the first beam splitter, (2) the first delay path, starting and ending in the first beam splitter, the first delay path comprising at least two focusing mirrors configured to focus said delayed part in piece on the first path of delay and return of the delayed part to the first beam splitter with a beam size and angular divergence equal to or approximately equal to the beam size and angular divergence of the laser output pulse beam in the first beam splitter.

Moreover, in the laser system, at least two focusing mirrors are spherical mirrors.

The laser system according to the second aspect of the invention further comprises a second delay path comprising at least two spherical mirrors, wherein the first delay path contains four focusing mirrors.

The laser system further comprises a second delay path formed by a second beam splitter located on the first delay path, said first delay path comprising a second beam splitter, wherein the laser further comprises a second delay path comprising at least two focusing mirrors arranged to focusing the delayed part in focus on the first delay path and returning the delayed part to the first beam splitter with a beam size and angular divergence equal to or approximately equal to the size and angular divergence of the laser output pulse beam in the first beam splitter.

The first beam splitter in the laser system according to the invention is configured to direct the laser beam in at least two directions using the optical property of truncated internal reflection, wherein the first beam splitter consists of two transparent optical elements, each of which has a flat surface, both optical elements are arranged so that their first flat surfaces are parallel to each other and separated by a distance of less than 200 nm, and the first beam splitter can be in made in the form of an optical element without coating, oriented at an angle to the laser output pulse beam, to achieve the desired reflection-transmission ratio.

Brief Description of the Drawings

Figure 1 depicts a perspective view of a preferred embodiment of the present invention,

figa and 1B - U-shaped optical table,

figs and 1C1 - the second preferred embodiment of the invention,

fig.1D is a third preferred embodiment of the invention,

figure 2 and 3 - details of the camera,

4 and 4A-E are details of a preferred switching power system,

5A, 5B, 5C1, 5C2, 5C3 and 5D are additional details of a switching power supply,

6 - 12V - details of the switching power supply,

13 is a method of minimizing oscillation problems,

figa1-13a6 - details of the preferred design for current return,

Fig and 15 - details of the electrode,

figa-16E is a method of selecting a line,

Fig - prism used as an output coupler,

Fig.18 - electric fan,

figa - the preferred design of the fan blades,

Fig.19 and 19A-19D5 - details of the purge system,

20, 20A and 20B are details of a preferred shutter,

Fig.21 and 21A - details of the heat exchanger,

figa-22D - details of the block of the pulse multiplier,

Fig.23, 23A and 23B are methods of spatial filtering of the seed beam.

Detailed Description of Preferred Options

First Preferred Option

Three wavelength platform

First general equipment layout

1 is a perspective view of a first preferred embodiment of the present invention. This embodiment is a narrow-band F ₂ laser injection injection system having the configuration of a ZGUM laser system. It is especially designed to be used as a light source for lithography integrated circuits. The main improvements of the present invention by the example of this embodiment compared with the known lithographic lasers consist in converting it into a light source with a wavelength of 157.63 nm based on F ₂ laser and in the use of injection seeds, and especially in the configuration of the master oscillator-power amplifier "(ZGUM) with two separate discharge chambers.

A first preferred embodiment of the invention is a F ₂ laser system, however, this system uses a modular platform configured to use fluorine (F ₂ ), krypton fluoride (KrF) or argon fluoride (ArF) components for the components. The design of this platform allows you to use the same main body and many modules of the laser system and components of any of these three types of lasers. The authors called this platform a “platform for three wavelengths,” because these three laser designs generate laser beams with wavelengths of about 248 nm for KrF, about 193 nm for ArF, and about 157.63 nm for F ₂ . This platform is also equipped with interface elements that ensure the compatibility of laser systems at each of the three wavelengths with modern lithographic equipment from major manufacturers. Preferred parameters of the F ₂ laser are as follows:

Repetition rate Pulse energy Pulse duration 4 kHz 7 mJ 24 ns 4 kHz 7 mJ 40 ns 4 kHz 10 mJ 24 ns 4 kHz 12 mJ 12 ns

The main components of this preferred laser system 2 are shown in FIG. They include the following:

(1) The frame 4 of the laser system is configured to accommodate all laser modules except for the module of the DC / AC power source,

(2) Module 6 of a high voltage AC / DC power supply,

(3) Module 7 of a resonant charger for charging two batteries of charging capacitors up to about 1000 volts with a frequency of 4000 charges per second,

(4) two switch modules 8A and 8B, each of which contains one of the capacitor banks mentioned above and a switch circuit for generating very short high voltage electrical pulses of about 16,000 volts with a duration of about 1 μs from the energy stored in the batteries of the charging capacitors,

(5) two modules of discharge chambers mounted one above the other in the frame 4 and consisting of module 10 of the master oscillator and module 12 of the power amplifier. Each module has a discharge chamber 10A and 12A and a compression head 10B and 12B mounted on top of the camera. The compression head compresses (in time) the electrical pulses from the switch module having a duration of about 1 μs to about 50 ns with a corresponding increase in current,

(6) the optical system of the master oscillator, which includes a rear mirror 100 and a line selection unit, BVL, 10C, which contains an output coupler and a five-prism line selector,

(7) a wave process unit 14, comprising an optical system and devices for generating and directing a seed beam to a power amplifier and for controlling the output power of the ZG,

(8) a beam stabilizer module 16 comprising monitors for wavelength, bandwidth, and energy,

(9) shutter module 18,

(10) an auxiliary housing in which the gas control module 20, the cooling water supply module 22, and the air ventilation module 24 are located,

(11) a user interface module 26,

(12) a laser control unit 28 and

(13) status lamp 30.

For many applications, the laser system will preferably have a pulse expansion unit (not shown) to extend the pulse duration to more than 12 ns.

U-shaped optical table

In the embodiment depicted in FIG. 1, the optical system of both 3D and PA is located on the U-shaped optical table shown in FIGS. 1A and 1B. The U-shaped optical table is kinematically coupled to the base of the laser, as described in US Pat. No. 5,863,017, referred to in the Description for reference. Both UM and ZG cameras are not mounted on the table, and each is supported by three wheels (two on one side and one on the other) on rails mounted from the lower frame of chamber 2. (Wheels and rails are preferably located as described in US Pat. No. 6,109,574, referenced for reference). This arrangement provides isolation of the optical system from vibrations caused by cameras.

Second general arrangement of equipment

The second general arrangement of equipment depicted in FIG. 1C is similar to the first described above, but has the following distinguishing features:

(1) Two cameras and a laser optical system are mounted on a vertical optical table 11, which is kinematically mounted (as will be described in the next section) in the laser housing 4. The cameras rest on rigid consoles bolted to the optical table. In this design, the driving generator 10 is mounted above the power amplifier 12.

(2) The 6V high-voltage power supply (VVIP) is located in the laser housing 4. This two-chamber F ₂ laser with a frequency of 4000 Hz requires two power supplies with a voltage of 1200 volts. However, space is provided in the laser housing for additional high voltage power supplies that may be required for the F ₂ laser system at a frequency of 6000 Hz.

(3) Each of the two laser chambers and switching power supplies for these cameras are very similar to the camera and switching power supply used in a single-chamber ArF laser system with a frequency of 4000 Hz, described in US patent application No. 09/854097, referred to here for reference ,

(4) The pulse multiplier module 13 located at the rear of the optical table 11 is included in this embodiment to extend the duration of the pulse exiting the power amplifier.

(5) The optical system for outputting the beam from the master oscillator in the line selection unit 10C directs the output beam from the MH to the input optics 14C of the power amplifier for one passage through the power amplifier 12. A short pulse (about 12 ns) from the output of the power amplifier 12 expands in the block 13 of the expansion pulse located at the back of the optical table 11. The entire path of the beam through the laser system, including the pulse expander, is hermetically sealed by shells (not shown) through which nitrogen or helium is blown.

Third overall equipment layout

Parts of a third general arrangement are shown in FIG. This arrangement represents an embodiment of the present invention using laser cameras in which the length of the discharge region is approximately half the length of the discharge region in the first two embodiments of the invention. That is, the length of the discharge region is about 26.5 cm, in contrast to the usual length of about 53 cm. In this case, the resonator of the master oscillator 10 (1) is formed by two passes through the discharge region between the maximum reflecting mirror 10E and the output coupler, which is located together with the five-prism line selector in BVL 10C. In this scheme, the beam makes four passes through the power amplifier 12 (1). The first pass after reflection from the mirror 15A goes through the lower half of the discharge region at an angle to the axis of the electrodes (for example, in the lower half from left to right at an angle of about 10 milliradians). The second passage after reflection from the mirrors 15B goes through the upper half at an angle from right to left of about 4 degrees. The third passage after reflection from two mirrors 15C is aligned with the electrodes through the upper half of the discharge region, and the last passage after reflection from mirrors 15D is aligned with the electrodes through the lower half of the discharge region. This last passage forms the output beam of the power amplifier. It bypasses the mirrors 15C and is guided by mirrors (not shown) to the pulse multiplier block (also not shown).

In each of the equipment configurations described above, measures are preferably provided to allow the output beam to exit from the left or right side of the laser casing in order to take into account the user's preferences without making significant changes to the design.

You can make some improvements to the characteristics of each of the three layouts by combining the switch and compression head into one module. In the past, authors rejected such a union because the failure of an element would require the replacement of the entire module. However, experience has shown that these elements are extremely reliable and can be combined in one module. In practice, one of the small number of failures of the switching power supply was caused by damage to the electrical cable connecting the two modules. In a federated module, this cable is not required.

In the future, the construction and operation of the preferred laser systems and modules described above will be described in more detail.

Master oscillator

The master oscillator 10 shown in FIGS. 1 and 1C is in many respects similar to the known ArF lasers described in the '884 patent and US 6,128,323, and the ArF laser described in US patent application 09/854097, except that The cameras and the optical system of the resonator are made specifically for the operation of the F ₂ laser in the spectral range of 157.63 nm. In addition, the energy of the output pulse is about 0.1-1.0 mJ instead of about 5 mJ. However, the main improvements compared to the laser according to the '323 patent allow operation at a frequency of 4000 Hz and higher. The master oscillator can be optimized for spectral characteristics, including line width control. The master oscillator comprises a discharge chamber 10A shown in FIG. 1, FIG. 2 and FIG. 3, in which two elongated electrodes 10A2 and 10A3 are located, each about 50 cm long, spaced about 12 mm apart. Anode 10A4 is mounted on the anode beam 10A6 forming a flow. Four separate finned water-cooled heat exchange units 10A-8 are provided. The tangential fan 10A10 is driven by two engines (not shown) to supply a laser gas stream at a speed of 80 m / s between the electrodes. The camera has window units (not shown) with CaF ₂ windows located at an angle of 47 ° relative to the laser beam. An electrostatic filter unit with an inlet in the center of the chamber filters out a small portion of the gas stream, as shown at 11 in FIG. 2, and the cleaned gas is directed to the window units as described in US Pat. No. 5,359,620 (referred to as reference) to carry discharge products from the windows. . The amplification region in the master oscillator is created by discharges between the electrodes through a laser gas, which in this embodiment contains about 0.5% F ₂ , the rest is neon, helium, or a combination of helium and neon. The gas stream cleans the products of each discharge from the discharge region before the next pulse. The resonator is formed on the output side by an output coupler located in the BVL 10C. The output coupler consists of uncoated CaF ₂ reflective optics mounted perpendicular to the beam direction to reflect about 5% of the laser light with a wavelength of about 157 nm and transmit about 95% of the light with a wavelength of 157 nm. The opposite boundary of the cavity is formed by a maximum reflective mirror 100, as shown in FIG. The line selection unit 10C will be described in more detail below with reference to FIG.

In preferred embodiments of the invention, the main batteries of the charging capacitors for the master oscillator and the power amplifier are charged in parallel to reduce the oscillation problem. This is preferred since the pulse compression time in the pulse compression circuits of two pulse power systems depends on the charge level of the charging capacitors. Preferably, the pulse energy at the output of the power amplifier is controlled sequentially for each pulse by adjusting the charging voltage. This to some extent limits the use of voltage to control the parameters of the beam of the master oscillator. However, the pressure of the laser gas and the concentration of F ₂ can be easily adjusted to obtain the required beam parameters for a wide range of pulse energy. The bandwidth decreases with decreasing F ₂ concentration and laser gas pressure. For a master oscillator, the time between discharge and luminescence depends on the concentration of F ₂ (1 ns / kPa), therefore, the concentration of F ₂ can also be changed to change the calculation of time, but this may be undesirable, since it may complicate other aspects of controlling the laser beam.

Amplifier

The power amplifier in each of the three embodiments of the invention consists of a laser chamber, which is very similar to the corresponding discharge chamber of the master oscillator. The presence of two separate chambers makes it possible to control to a large extent the pulse energy and the integrated pulse sequence energy (called the dose) separately from the wavelength and bandwidth. This improves the stability of the dose. All camera elements are the same and interchangeable during the manufacturing process. However, during operation, the gas pressure in the exhaust gas is preferably significantly lower than in the CM. The power amplifier compression head 12B shown in FIG. 1 is also substantially identical in this embodiment to the compression head 10B, and the elements of the compression head are also interchangeable in manufacturing. One difference is that the capacitors of the batteries of the condensers of the compression head are placed more widely for the master oscillator, in order to obtain a significantly higher inductance compared to the PA. In addition, the distance between the electrodes in the power amplifier is preferably about 8 mm (compared with a distance of about 12 mm in the 3G). This similarity of chambers and electrical elements of pulsed power supply systems ensures that the characteristics of the calculation of time for pulse shaping circuits will be the same or almost the same, and this will minimize the problems of oscillation.

The power amplifier is capable of one passage through the discharge region of the discharge chamber of the power amplifier in the variants of the invention according to FIG. 1 and FIG. 1C and four passes in the embodiment according to FIG. 1D, as described above. In the embodiment of FIG. 1, a line is selected in the BWL 10C using the five prism line selector shown in FIG. 16. The line selection is preferably carried out after the active medium of the MO, since the gain in F ₂ lasers is very high compared to KrF and ArF lasers. A seed beam with a selected line is reflected upward by mirror 14A and is reflected horizontally for one pass through the power amplifier, as discussed above. Charging voltages are preferably selected sequentially for each pulse in order to maintain the required pulse energies and doses. The concentration of F ₂ and the pressure of the laser gas can be adjusted in the power amplifier to provide the required operating range of the charging voltage. This desired range can be chosen so as to obtain the desired value of dE / dV, since the change in energy with a change in voltage depends on the concentration of F ₂ and the pressure of the laser gas. The calculation of the injection time is preferably based on the charging voltage. The injection frequency should preferably be high to maintain relatively constant conditions in the laser chamber and may be continuous or semi-continuous. Some users of these options may prefer longer durations (e.g. 2 hours) between F ₂ injections.

Switching power supply

In the preferred embodiment of the invention shown in FIGS. 1, 1C and 1D, the main switching power supply circuits are similar to the switching power supply circuits of known excimer laser light sources intended for lithography. However, separate impulse power circuits are provided after the charging capacitors for each discharge chamber. Preferably, one resonant charger charges two batteries of charging capacitors connected in parallel to ensure that both batteries of charging capacitors are charged with exactly the same voltage. Important improvements are also provided for adjusting the temperature of elements of switching power circuits. In preferred embodiments, the temperatures of the magnetic cores of the saturable inductance coils are controlled, and the temperature signals are used in the feedback circuit to adjust the relative discharge time in the two chambers. FIGS. 5A and 5B show important elements of a preferred main switching power supply circuit that is used for MH. The same basic scheme is used for PA.

Resonant Charger

A preferred resonant charger system is shown in FIG. 5 B. The main elements of this circuit are:

I1 is a three-phase power supply 300 with a constant DC output.

C-1 is the original capacitor 302, which is an order of magnitude or much greater than the existing C ₀ capacitor 42.

Q1, Q2 and Q3 are switches for controlling the flow of current for charging and maintaining an adjustable voltage at C ₀ .

D1, D2 and D3 provide current flow in one direction.

R1 and R2 provide voltage feedback to the control circuit.

R3 allows you to quickly discharge the voltage at C ₀ in the case of a small excess charge.

L1 is a resonant inductor between the capacitor C-1 302 and the batteries 42 of the capacitors C ₀ to limit the current flow and set the time of charge transfer.

The control board 304 gives commands to open and close Q1, Q2, Q3 based on the feedback circuit parameters.

This circuit contains a switch Q2 and a diode D3, which are collectively known as a De-Qing switch. This switch improves circuit adjustment by allowing the control unit to short-circuit the inductor during the resonant charging process. This “de-qing" prevents the transfer of the additional energy stored in the current of the charging inductance coil, L1, to the capacitor C ₀ .

Before the need for a laser pulse arises, the voltage at C-1 is charged to 600-800 volts, and the switches Q1-Q3 open. After a command from the laser, Q1 closes. At this time, current will flow from C-1 to C ₀ through the charging inductor L1. As described in the previous section, the computer on the control panel estimates the voltage at C ₀ and the current flowing in L1, relative to the voltage value set by the laser command. Q1 opens when the voltage across the capacitor banks CO plus the equivalent energy stored in the inductor L1 is equal to the required voltage according to the command. It is calculated as

V _f = [V _C0s ² + ((L ₁ · I _L1s ² ) / C ₀ )] ^0.5

where V _f = voltage at C ₀ after Q1 opens and the current in L1 tends to zero.

V _C0s = voltage at C ₀ when Q1 opens.

I _L1s = current flowing through L1 when Q1 opens.

After opening Q1, the energy stored in L1 begins to be transferred to the CO capacitor banks through D2 until the voltage across the CO capacitor banks becomes approximately equal to the voltage specified by the command. At this time, Q2 closes, and the current stops flowing in CO and is directed through D3. In addition to the de-jing circuit, Q3 and R3 from the tap circuit allow for additional fine-tuning of the CO voltage.

The switch Q3 of the bleed circuit 216 will receive a short circuit command from the control panel, while the current flowing through the inductor L1 will stop and the voltage across the CO will be etched until the desired control voltage is reached, and then the switch Q3 will open. The time constant of the capacitor C ₀ and the resistor R3 must be small enough to enable the capacitor C ₀ to be pitted to the command voltage, without taking up a significant part of the total charging cycle.

As a result, the resonant charger can be configured with three levels of control. Somewhat coarse adjustment is carried out by the energy calculator and the opening of switch Q1 during the charging cycle. When the voltage across the CO capacitor banks approaches the target value, the de-qing switch closes, stopping the resonant charging when the voltage at C ₀ is equal to or slightly higher than the target value. In a preferred embodiment, the Q1 switch and the de-jing switch are used to provide adjustment with an accuracy higher than +/- 0.1%. If additional adjustment is required, a third voltage adjustment control can be used. This is a bleed circuit formed by switch Q3 and R3 (shown as 216 in FIG. 5B) to discharge the CO to the exact target value.

Improvements after CO

As indicated above, each of the pulse power supply systems ZG and UM according to the invention has the same basic structure (figa), as used in known systems. However, some significant improvements to this basic design were required in order to increase the heat load by approximately 3 times, caused by a significantly increased repetition rate. These enhancements will be discussed below.

Detailed description of the switch and compression head

This section will describe the design details of the switch and compression head.

Semiconductor switch

The semiconductor switch 46 is a P / N switch CM 800 HA-34H IGBT manufactured by Powerex, Inc., Youngwood, PA. In a preferred embodiment, two such switches are used in parallel.

Inductors

Inductors 48.54 and 64 are saturable inductors, similar to those used in known systems described in US patent No. 5448580 and No. 5315611. 6 shows the preferred design of the inductor L ₀ 48. In this inductor, four conductors from two IGBT switches 46B pass through sixteen ferrite ring cores 49, forming a portion 8A of a hollow cylinder 8 inches long of a highly permeable material with an inner diameter (ID) about 1 inch and an outer diameter (OD) of about 1.5 inches. Each of the four conductors is then wrapped twice around the insulating toroidal core, forming a 48 V. Part. Next, the four conductors are connected to a plate, which, in turn, is connected to the high voltage side of the battery 52 of capacitors C ₁ .

Fig. 8 shows a drawing of a preferred saturable inductor 54. In this case, the inductor has a single-turn geometry in which the upper and lower assembly covers 541 and 542 and the central core 543, all under high voltage, form one turn through the magnetic cores of the inductor . The outer casing 545 has ground potential. The magnetic cores are made of 0.0005 inch thick tape of 50-50% Ni-Fe alloy supplied by Magnetics, Butler, PA or National Arnold, Adelanto, CA. The fins 546 on the inductor housing facilitate the transfer of the heat dissipated inside to forced air cooling. In addition, a ceramic disk (not shown) is mounted under the bottom cover of the reactor to facilitate heat transfer from the central part of this assembly to the main plate of the modular chassis. On Fig also shows high-voltage connections with one capacitor of the battery 52 of the capacitors C ₁ and with a high-voltage wire on one of the induction blocks of the pulse transformer 56 with a boost factor of 1:25. Case 545 is connected to the ground wire of block 56.

On figa and 9B, respectively, shows a view of the saturated inductor 64 from above and in section. In the inductors of this embodiment, flow-extinguishing metal parts 301, 302, 303 and 304 are added, as shown in FIG. 9B, to reduce leakage flow in the inductors. These parts excluding the flux significantly reduce the area into which the magnetic flux can penetrate, and thereby contribute to the reduction of the saturated inductance of the inductor. The current makes five loops through the vertical conductive rods in the assembly of the inductor around the magnetic core 307. The current enters at point 305, passes down the large diameter conductor in the center, designated as "1", and up through six small conductors on the circle, also indicated as “1” as shown in FIG. 9A. Then, the current flows down two conductors, designated as 2, on the inside, then up six conductors, designated as 2 on the outside, then downstream to prevent metal flow on the inside and upward through six conductors, indicated as 3, on the outside side, then down two conductors labeled 3 on the inside, then up six conductors labeled 4 on the outside, then down a conductor labeled 4 on the inside. A voltage equal to half the total pulse voltage across the conductor is maintained in the metal-excluding metal elements, which reduces the safe distance between the metal-excluding metal parts and the metal rods of other turns. The magnetic core 307 is made up of three coils 307A, B, and C formed by winding a 0.0005 inch thick tape of 80-20% Ni-Fe alloy supplied by Magnetics, Butler, PA or National Arnold, Adelanto, CA. It should be noted that nanocrystalline materials such as VITROPERM manufactured by VACUUM SCHITELZE GmbH, Germany and FINEMET manufactured by Hitachi Metals, Japan, can be used for inductors 54 and 64. In known switching power systems, a potential problem may be oil leakage from electrical elements. In this preferred embodiment, oil-insulated elements are limited only by saturable inductor coils. Moreover, the saturable inductor 64 shown in FIG. 9B is housed in an oil housing in the form of a container in which all sealing connections are located above the oil level to substantially eliminate the possibility of oil leakage. For example, the lowest seal in inductor 64 is shown as 308 in FIG. Since the normal oil level is located under the top cover of the housing 306, there is no possibility of oil flowing out of the assembly to the outside while the housing is held upright.

Capacitors

All capacitor banks 42, 52, 62, and 82 (i.e., C ₀ , C ₁ , C _p-1, and C _p ) shown in FIG. 5A are comprised of a commercially available capacitor bank connected in parallel. Capacitors 42 and 52 are film-type capacitors supplied by Vishay Roederstein, Statesville, North Carolina, or Wima, Germany. The authors preferred soldering their capacitors and inductors to the positive and negative terminals on a special printed circuit board having copper wires with a thick nickel coating, as described in US patent No. 5448580. Batteries 62 and 64 capacitors traditionally consist of a parallel matrix of high voltage ceramic capacitors from suppliers such as Murata or TDK, Japan. In a preferred embodiment, to use this ArF laser, the capacitor bank 82 (i.e., C _p ) is a thirty-three 0.3 nF capacitor bank with a total capacitance of 9.9 nF; C _p-1 is a battery of twenty-four 0.40 nF capacitors with a total capacitance of 9.6 nF; C ₁ is a 5.7 uF capacitor bank and C ₀ is a 5.3 uF battery.

Pulse transformer

Pulse transformer 56 is also similar to the pulse transformer described in US patent No. 5448580 and No. 5313481, however, pulse transformers in this embodiment have only one turn in the secondary winding and 24 induction elements equivalent to 1/24 of one primary turn for an equivalent gain of 1:24 . 10 is a drawing of a pulse transformer 56. Each of the 24 induction blocks contains an aluminum coil 56A having two flanges (each with a flat edge with threaded holes for bolts) that are bolted to the positive and negative terminals on the printed circuit board 56B, as shown along the bottom edge of FIG. 10. (The negative terminals are the high voltage terminals of the twenty-four primary windings). Insulators 56C separate the positive terminal of each coil from the negative terminal of an adjacent coil. Between the flanges of the coil there is a hollow cylinder with a length of 1/16 inch with a diameter of 0.875 and a wall thickness of about 1/32 inch. The coil is wrapped with Metglas 2605 S3A material one inch wide, 0.7 mils thick, and 1 mil thick Mylar film until the ND reaches 2.24 inches of the Metglas insulation wrap. A perspective view of one wrapped coil forming one primary winding is shown in FIG. 10A.

The secondary winding of the transformer is a single rod with stainless steel ND mounted in a tightly fitting PTFE (Teflon) insulating tube. The winding consists of four sections, as shown in Fig.10. The low voltage end of the stainless steel secondary winding, shown as 56D in FIG. 10, is connected to the primary high voltage wire on the 56B circuit board at 56E, wherein the high voltage terminal is shown as 56F. As a result, the transformer has an autotransformer configuration with a boost factor of 1:25 instead of 1:24. Thus, a pulse of approximately -1400 volts between the + and - terminals of the induction nodes generates a pulse of approximately -35000 volts at terminal 56F on the secondary side. This single-turn secondary winding design provides a very low leakage inductance, resulting in a very fast rise time.

Detailed description of the electrical elements of the laser camera

Capacitor Cp 82 is a battery of thirty-three 0.3 nF capacitors mounted on top of the pressure chamber. (Usually, an ArF laser works with a laser gas consisting of 3.5% argon, 0.1% fluorine, the rest is neon). The electrodes are about 28 inches long and spaced about 0.5-1.0 inches apart, preferably about 5/8 inches apart. Preferred electrodes will be described below. In this embodiment, the upper electrode is called the cathode, and the lower electrode is grounded, as shown in FIG. 5A, and is called the anode.

Discharge time calculation

In ArF, KrF, and F ₂ discharge lasers, the electrical discharge lasts only about 50 ns (i.e., 50 billionths of a second). This discharge creates the population inversion necessary for amplification, but the inversion exists only during the discharge. Therefore, an important requirement for the injection seed of an ArF, KrF, and F ₂ laser is to ensure that the seed beam from the master oscillator passes through the discharge region of the power amplifier during these approximately 50 billionths of a second when the population is inverted in the laser gas so that amplification can occur seed beam. An important obstacle to accurate calculation of the discharge time is the fact that there is a delay of approximately 5 microseconds between the start of the time switch 46 (as shown in FIG. 5A) for closing and the start of the discharge, which lasts only about 40-50 ns. About 5 microseconds of time are spent for the pulse to oscillate through the circuit between C ₀ and the electrodes. This time interval varies significantly depending on the magnitude of the charging voltage and the temperature of the inductors in the circuit.

Nevertheless, in the described preferred embodiment, the authors have developed electrical pulse power circuits that provide control over the calculation of the discharge time of two discharge chambers within a relative accuracy of less than about 2 ns (i.e., 2 billionths of a second). Structural diagrams of two circuits are shown in FIG.

The authors conducted tests that showed that the calculation of time changes with a change in the charging voltage by approximately 5-10 ns / volt. This imposes a strict requirement on the accuracy and repeatability of a high voltage power supply that charges charging capacitors. For example, if you want to control the calculation of time by 5 ns with a sensitivity shift of 10 ns per volt, then the resolution accuracy will be 0.5 volts. For a nominal charging voltage of 1000 volts, this will require a charging accuracy of 0.05%, which is very difficult to achieve, especially when capacitors need to be charged to these specific values 4000 times per second.

The authors preferred to solve this problem by charging the charging capacitor of both the MH and the PA in parallel from one resonant charger 7, as shown in FIGS. 1 and 4 and described above. It is also important to design two pulse compression / gain circuits for these two systems so that the time delay versus charge voltage relationship curves are consistent, as shown in FIG. 4A. This requirement is easy to implement using the same elements as possible in each circuit.

Therefore, to minimize changes in the calculation of time (called oscillations), in this preferred embodiment, the authors developed switching batteries for both discharge chambers with the same components and confirmed that the curves of the ratio of the time delay to voltage actually repeat each other, as shown in figa . The authors confirmed that in the normal operating range of the charging voltage there is a significant change in the time delay with a change in voltage, but this change is practically the same for both circuits. Therefore, when both charging capacitors are simultaneously charged, the charging voltages can be changed over a wide operating range without changing the relative discharge time.

The temperature control of electrical elements in a pulsed power circuit is also important, since temperature changes can affect the pulse compression time (especially temperature changes in saturable inductors). Thus, the designer is faced with the task of minimizing temperature changes, and the second task is to control the temperature of temperature-sensitive elements and, using feedback control, adjust the start time for compensation. Management can be carried out using a processor programmed by the learning algorithm to make adjustments based on statistical data, which determines the relationship of past changes in the calculation of time with known working statistics. These statistics are then used to predict changes in the timing calculation based on the current operation of the laser system.

Launch control

The discharge is launched for each of the two chambers separately using each starting circuit, for example, the circuit described in US Pat. No. 6,016325. These circuits add time delays to correct changes in the charging voltage and temperature in the electrical elements of the switching power supply so that the time between start-up and discharge is kept as constant as possible. As noted above, since these two schemes are basically the same, the changes after correction are almost equal (i.e., within 2 ns relative to each other).

The characteristics of this preferred embodiment are significantly improved if the discharge in the power amplifier occurs approximately 40-50 ns after the discharge in the master oscillator. This is because several nanoseconds are required for the development of a laser pulse in the master oscillator and a few more nanoseconds for the front of the laser beam from the generator to reach the amplifier, and also because the rear end of the laser pulse from the master oscillator has a much narrower bandwidth, than its front end. For this reason, for each camera, separate trigger signals are sent to the start switch 46. The actual delay is selected to achieve the desired beam quality based on the actual performance. You can, for example, notice that a narrower bandwidth and longer pulses can be obtained due to the pulse energy by increasing the delay between the ZG trigger and the UM trigger.

Other ways to manage discharge time calculation

Since the relative discharge time can have an important effect on the quality of the beam, additional measures can be justified to control the calculation of the discharge time. For example, some laser modes can cause wide fluctuations in the charging voltage or temperature of the induction coil. These wide fluctuations can complicate discharge time management.

Monitoring time

The discharge time can be monitored on each consecutive pulse, and the time difference can be used in a feedback control system to adjust the timing of the trigger signals closing switch 42. Preferably, the discharge of the MIND can be controlled using a photosensor to observe the fluorescence of the discharge (called ASE) rather than laser beam, since in the absence of the formation of a laser beam in the CM, the calculation of time will be very incorrect. For MH, either ASE or a seed laser pulse can be used.

Bias voltage adjustment

The pulse time can be increased or decreased by adjusting the bias currents through the inductors L _B1 L _B2 and L _B3 , which provide bias for the inductors 48, 54 and 64, as shown in figa. Other methods can be used to increase the time required to saturate these inductors. For example, the core material can be mechanically separated from the very responsive PZT element, which can be controlled with feedback based on the feedback signal from the pulse synchronization monitor.

Adjustable spurious load

An adjustable spurious load can be added to one or both switching power supply circuits after CO.

Additional feedback control

In addition to the control signals of the synchronization of pulses in feedback control, you can use the signals of the charging voltage and temperature of the inductor to adjust the bias voltage or mechanical separation of the core, as mentioned above, in addition to the adjustment of the calculation of the start time described above.

Work in batch mode

Feedback time management is relatively easy and efficient when the laser is in continuous operation. However, typically lithographic lasers operate in batch mode, for example, illustrated below to process 20 sections on each of a plurality of plates:

Switch off for 1 minute to move the plate into place

4000 Hz for 0.2 seconds to illuminate area 1

Turn off for 0.3 seconds to move to section 2

4000 Hz for 0.2 seconds to illuminate area 2

Turn off for 0.3 seconds to move to section 3

4000 Hz for 0.2 seconds to illuminate section 3

...

4000 Hz for 0.2 seconds to illuminate area 199

Turn off for 0.3 seconds to move to area 200

4000 Hz for 0.2 seconds to illuminate 200

Switch off for 1 minute to replace the plates

4000 Hz for 0.2 seconds to illuminate section 1 on the next plate, etc.

This process can be repeated for many hours, but from time to time it will be interrupted for periods longer than 1 minute.

The length of downtime will affect the relative timing between the pulsed power supply systems of the gas generator and the amplifier, and it may be necessary to adjust the start control to ensure that a discharge in the amplifier will occur when the seed beam from the generator is in the desired location. By controlling the discharges and the time the light exits from each camera, the laser operator can adjust the start time (with an accuracy of about 2 ns) to achieve the best performance.

Preferably, the laser control processor is programmed to automatically control the calculation of the time and quality of the beam and adjust the time to ensure the best performance. In preferred embodiments of the invention, time calculation algorithms are used that generate sets of binary values applicable to various operating modes. These algorithms in preferred embodiments are configured to switch to feedback control during continuous operation when the time values for the current pulse are a set based on feedback data collected for one or more previous pulses (e.g., immediately preceding pulse).

No output discharge

Time calculation algorithms, such as those described above, are very well suited for continuous or regularly repeated work. However, the accuracy of the time calculation may not be satisfactory enough in an unusual situation, for example, for the first pulse after turning off the laser for an abnormal period of time, for example, 5 minutes. In some situations, inaccurate timing for the first one or two bursts of a packet may not be a problem. The preferred method is to pre-program the laser in such a way that the discharges of GB and AM deliberately occurred out of sequence for one or two pulses, to amplify the seed beam from the GB was impossible. For example, a laser can be programmed to start a UM discharge 80 ns before the start of a 3G. In this case, no significant output will occur from the laser, but the laser metrological sensors will be able to determine the time parameters, so that the time parameters for the first output pulse will be accurate. Alternatively, the GB can be launched quite early in relation to the launch of the MIND, so that the GB beam passes through the MIND before the AM discharge. 40, 4D1, 4E and 4E1 illustrate possible control algorithms using these methods.

Water cooling elements

To account for large thermal loads, water cooling of pulsed batteries is provided in addition to the usual air cooling provided by cooling fans inside the laser housing to support an operating pulse frequency of 4 KHz or higher.

One drawback of water cooling is traditionally the possibility of leaks near electrical elements or high voltage wires. In this particular embodiment, this potential problem is essentially eliminated by the use of one solid piece of cooling pipe, which is laid in the module to cool those elements that usually dissipate most of the heat in the module. Since there are no connections inside the module chamber and the cooling pipe is a continuous piece of solid metal (for example, copper, stainless steel, etc.), the leakage potential is significantly reduced. Therefore, modular connections with cooling water are made outside the assembly housing made of sheet metal, where the cooling pipe articulates with a quick disconnect connector.

Saturable Inductor

A water-cooled saturable inductor 54A is provided in the switch module, as shown in FIG. 11, which is similar to the inductor 54 shown in FIG. 8, except that the fins 54 are replaced by a water-cooled jacket 54A1, as shown in FIG. 11. The cooling line 54A2 is laid in the module around the jacket 54A1 and through the aluminum base plate, where IGBT switches and serial diodes are installed. These three components provide most of the power dissipation inside the module. Other elements that also dissipate heat (damping diodes and resistors, capacitors, etc.) are cooled by the forced air flow generated by two fans at the rear of the module.

Since jacket 54A1 is held at ground potential, there is no problem of isolation from voltage when the cooling pipe is directly connected to the reactor vessel. This is done by pressing the pipe into the trapezoidal groove cut in the outside of the casing, as shown at 54A3, and using a thermally conductive joint that contributes to good thermal contact between the cooling pipe and the casing.

Cooling high voltage elements

Although IGBT switches “float” at high voltage, they are mounted on an aluminum base that is electrically isolated from the switches with an 1/16 inch thick aluminum plate. The aluminum base plate, which functions as a heat sink and with earth potential, is much easier to cool since high-voltage insulation is not required in the cooling circuit.

A drawing of a water-cooled aluminum plate is shown in Fig.7. In this case, the cooling pipe is pressed into the groove in the aluminum base on which the IGBTs are mounted. As with the inductor 54a, a heat-conducting connection is used to improve the overall connection between the pipe and the base plate.

Series diodes also “float” at high potential during normal operation. In this case, the diode case, traditionally used in this design, does not provide isolation from high voltage. To ensure this, a package of diodes in the form of a “hockey puck” is clamped in the heat sink assembly, which is then mounted on top of a ceramic base, and it, in turn, is mounted on a water-cooled aluminum base plate. The ceramic base only has a sufficient thickness to provide the necessary electrical insulation, but it is not so thick as to introduce a thermal impedance higher than necessary. 1/16 inch thick alumina is used as ceramic in this design, although other, more exotic, materials such as beryllium oxide can also be used to further reduce the thermal impedance between the diode compound and cooling water.

The second version of the water-cooled switch uses one assembly with a cold plate, which is attached to the base of the chassis for IGBT and diodes. A cold plate can be made by brazing one solid nickel pipe to two aluminum "upper" and "lower" plates. As described above, the IGBTs and diodes are configured to transfer their heat to the cold plate by using the above-mentioned ceramic disks located under the assembly. In a preferred embodiment of the invention, a cold plate cooling method is also used to cool IGBTs and diodes in a resonant charger. Thermally conductive rods or a heat pipe are also used to transfer heat from the outside of the chassis to the chassis plate.

Detailed description of the compression head

The water-cooled compression head has a design similar in part to the circuitry known to the prior art air-cooled version (uses the same type of ceramic capacitors and similar material in the reactor structures). The main differences in this case are that the module must operate at higher frequencies and therefore with a higher average power. In the case of the compression head module, most of the heat is dissipated inside the modified saturable inductor 64A. The cooling of this part is not the only problem, since the entire body operates with short pulses of very high voltage. A solution to this problem is shown in FIGS. 12, 12A and 12B for inductively decoupling the case from ground potential. This inductance is provided by wrapping the cooling pipe around two cylindrical shapes that contain a ferrite magnetic core. Both the inlet and outlet cooling lines are wound around the cylindrical parts of the ferrite core formed of two cylindrical parts and two ferrite blocks, as shown in FIGS. 12, 12A and 12B.

Ferrite elements are made of CN-20 material manufactured by Ceramic Magnetics Inc., Ferfield, New Jersey. One part of the copper pipe (0.187 inch in diameter) is press fit and wound around a winding shape around body 64A1 of inductor 64A and around a second winding shape. Sufficient length is left at the ends allowing it to pass through fittings in the cover of the compression head made of sheet metal, so that there are no cooling pipe connections inside the chassis.

The inductor 64A comprises a trapezoidal groove, as shown at 64A2, similar to that used in the water-cooled housing of the first stage of the switch reactor. This case is almost the same as previous air-cooled versions except for the trapezoidal groove. A copper water-cooled tube is press-fit into this groove to obtain a good thermal connection between the body and the water-cooled tube. A heat-conducting compound is also added to minimize thermal impedance.

The electrical circuit of the inductor 64A is slightly changed compared to 64 shown in FIGS. 9A and 9B. The inductor 64A forms only two loops (instead of five) around the magnetic core 64A3, which consists of four turns of tape (instead of three).

As a result of this water-cooled tubing from the output potential to the ground, the bias current circuit in this case is somewhat different. As before, the bias current is supplied by the DC / DC converter to the switch via a cable in the compression head. The current passes through the "positive" bias inductor L _B2 and is connected to the voltage node Ср-1. Then the current is separated, and one part is returned to the switch via a high-voltage cable (passing through the secondary winding of the transformer to the ground and back to the DC-DC converter). The other part passes through the reactor Lp-1 of the compression head (to bias the magnetic switch) and then through the "negative" bias inductor L _{B3 of the} water-cooled pipe back to earth and to the DC / DC converter. By balancing the resistance in each branch, the designer will be able to ensure that there is sufficient current for both the compression head reactor and the switch transformer.

A "positive" bias inductor L _B2 is similar to a "negative" bias inductor L _B2 . In this case, the same ferrite rods and blocks are used as a magnetic core. However, two 0.125-inch-thick plastic spacers are used to create an air gap in the magnetic circuit so that the cores are not saturated with direct current. Instead of winding an inductor with water-cooled tubing, 18AWG Teflon wire is wound around the molds.

Quick connections

In this preferred embodiment of the invention, the three modules of switching power supplies use reactive conjugated electrical connections, so that all electrical connections to parts of the laser system are made simply by sliding the module in place in the laser housing. These modules include an AC distribution module, a power supply module, and a resonant charging module. In each case, the plug on the module mates with the counterpart on the back of the case. In each case, two approximately 3-inch taper pins on the module guide the module to its exact position so that the electrical connectors fit correctly. Reactive mating connectors, such as the AMP Model No. 194242-1, are commercially available from AMP, Inc., Harrisburg, PA. In this embodiment, the connectors are designed for various power circuits, for example, 208 volts of alternating current, 400 volts of alternating current, 1000 volts of direct current (output of the power source and input of resonant charges) and several signal voltages. These reactive mates allow you to remove the modules for maintenance and replace them within seconds or minutes. In this embodiment, reactive coupled connections are not used for the switch module, the output voltage of which is in the range of 20-30000 volts. Instead, a conventional high voltage connector is used.

Bit elements

13 and 13A (1) show details of an improved bit configuration used in preferred embodiments of the present invention. This configuration includes the configuration of the electrode, which the authors called a blade-dielectric electrode. In this design, the anode 540 contains an electrode 542 in the form of a blunt-ended blade with dielectric spacers 544 mounted on both sides of the anode, as shown in the drawing, to improve the passage of gas flow in the discharge region. The spacers are attached to the support beam 546 of the anode with screws at each end of the spacers under the discharge region. These screws allow thermal expansion between the spacers and the beam. The anode has a length of 26.4 inches and a height of 0.439 inches. Its width is 0.284 inches at the bottom and 0.141 inches at the top. It is connected to the support beam 546 by forming anode flow with screws through sockets that allow differential thermal expansion of the electrode from its central position. The anode consists of an alloy based on copper, preferably C36000, C95400 or C19400. The cathode 541 has the cross-sectional shape shown in FIG. Preferably, C36000 is used as the cathode material. Additional details of this blade-dielectric configuration are described in US patent application No. 09/768753, incorporated herein by reference. The current return path 548 in this configuration consists of a whalebone part with 27 ribs evenly spaced along the length of the electrode 542, the cross section of which is shown in FIG. 13A (1). As noted above, the current return path is made of sheet metal, and the fins in the form of a whalebone (each with a cross-sectional size of about 0.15 × 0.09 inches) are bent so that a long dimension of each edge is located in the direction of the current flow.

An alternative dielectric spacer structure for the anode, designed to further improve flow, is shown in FIG. 13A2. In this case, the spacers more accurately mate with the support beam forming the flow of the anode to ensure better passage of the gas stream. The authors called this design a "quick return" blade-dielectric anode.

Alternative switching power supply

A second preferred switching power supply circuit is shown in FIGS. 5C1, 5C2 and 5C3. This circuit is similar to that described above, but it uses a higher voltage power supply to charge C _° to a higher value. As in the options described above, a high-voltage switching power supply operating from an industrial power of 230 or 460 volts AC is a power source for the fast-charging resonant charger described above, and is designed to accurately charge two 2.27 mF with a frequency of 4000 -6000 Hz to voltages in the range 1100-2250 V. The electric elements in the switch and the compression head for the master oscillator are, as far as possible, identical to the corresponding elements in the power amplifier. This is done in order to maintain the same temporal characteristics as possible in two circuits. Switches 46 are parallel groups of two IGBT switches, each rated at 3300 V. The batteries of 42 C ₀ capacitors consist of 128 capacitors at 0.068 mF, 1600 V located in 64 parallel branches to provide a 2.17 mF Co battery. Batteries 52 of capacitors C ₁ consist of 136 capacitors of 0.068 mF, 1600 V located in 68 parallel branches, to ensure a battery capacity of 2.33 mF. The capacitor banks C _p-1 and C _p are the same as described above with reference to FIG. 5A. 54 saturable inductors are single-coil inductors providing a saturated inductance of about 3.3 nG with five cores made of 50% -50% Ni-Fe alloy 0.5-inch thick with 4.9-inch ID and 3.8-inch ID. 64 saturable inductors are double-coil inductors providing a saturated inductance of about 38 nG, each of which consists of 5 0.5-inch-thick cores made from an 80% -20% Ni-Fe alloy with a 5-inch diameter and 2.28-inch high-pressure impellers. Trigger circuits are provided for closing the IGBT 46 with two nanosecond timing accuracy. The master oscillator typically starts 40 ns before the IGBT 46 starts for the power amplifier. However, the accurate timing is preferably determined by feedback signals from sensors that measure the time it takes to exit the master oscillator and discharge in the power amplifier.

Impulse length

The length of the output pulse measured in the experiments conducted by the inventors on these F ₂ lasers is about 12 ns and to some extent depends on the relative time of the two discharges. A large pulse length (ceteris paribus) can increase the service life of the optical elements of lithographic equipment.

The authors identified several methods for increasing the pulse length. As noted above, the relative time between discharges can be optimized for the pulse length. The pulse power supply circuits of ZG and UM can be optimized to provide longer pulses using the methods described in US patent application No. 09/451995, referred to here by reference. An optical pulse multiplier system, as described in US Pat. No. 6,067,311, incorporated herein by reference, may be added after the CM to reduce the intensity of individual pulses. A preferred pulse multiplier unit (also called a pulse expander) will be described in the next section. This pulse multiplier can be part of the beam path to the lens elements of the lithographic device. The camera can be made more extended, and the electrodes can be made with the possibility of issuing discharges of the traveling wave, designed to obtain a longer pulse length.

Pulse Multiplier

A preferred pulse multiplier is shown in FIG. 22A. A beam of light 20 from a laser 50 hits a beam splitter 22. The splitter has a reflectivity of about 40%. About 40% of the light is reflected as the first part of the output beam 30. The rest of the incoming beam is transmitted through the splitter 22 as a beam 24. This beam is reflected back at a very small angle by a mirror 26, which is a spherical mirror with a focal length equal to the distance from the beam splitter 22 to mirrors. So, the beam is focused at a point 27 near the beam splitter 22, slightly not falling on it. This beam propagates further and is now reflected by mirror 28, which is also a spherical mirror with a focal length equal to the distance from this mirror to point 27. Mirror 28 reflects the beam back at a small angle and also collimates the reflected beam. This reflected beam 32 propagates to the right and is reflected by the mirror 29 to the beam splitter 22, where about 60% of the beam is transmitted through the beam splitter 22 to combine and become the second part of the output beam 30. Part (about 40%) of the beam 34 is reflected by the beam splitter 22 in the direction beam 24 to repeat the movement of the beam 32. As a result, the short input pulse is divided into several parts, so that the total duration of the beam increases, and its peak intensity decreases. Mirrors 26 and 28 form a transmission system that reproduces parts of the outgoing beam on top of each other. Due to this reproduction, each part of the output beam is essentially the same. (If mirrors 26 and 28 were flat, the beam divergence would propagate the beam at each subsequent repetition, so that the size of the beam would be different for each repetition). The total length of the optical path from the beam splitter 22 to the mirror 26, mirror 28, mirror 27, and finally to the beam splitter 22 determines the time delay between repetitions. 22 B1 shows a profile of a typical pulse generated by an ArF laser. (These results can be applied to an F ₂ laser, except that the non-multiplied laser pulse 12 F ₀ is about 12 ns instead of 18 ns for an ArF laser). FIG. 22B2 shows a simulated output profile of such an ArF laser pulse after being expanded in a pulse expander configured in accordance with FIG. 6. In this example, the T _is pulse was increased from 18.16 ns to 46.78 ns. (T _{is the} pulse duration is a measure of the duration used to describe laser pulses, meaning the square of an integer number of pulse durations).

On figs shows a layout similar to the layout in figa, but with an additional delay path. In this case, the first beam splitter 22A is configured to reflect 25 percent, and the second beam splitter 22B is configured to reflect 40 percent. The beam shape obtained by computer simulation is shown in FIG. 22D. T _is for this extended pulse is about 73.2 ns. In the embodiment of FIG. 22C, the parts of the beam transmitted through the splitter 22B rotate the orientation when they return and combine into the output beam 30, which causes a significant decrease in the spatial coherence of the beam.

The pulse extension unit can be mounted behind the vertical optical table 11, as suggested above, either on top of the table or even inside it.

Pulse and dose energy management

The control of the pulse energy and the dose energy is preferably carried out using a feedback control system and an algorithm, for example, described above. The pulse energy monitor can be located on the laser as close as possible to the plate in the lithographic device. Using this method, charging voltages are selected to obtain the desired pulse energy. In the preferred embodiment described above, the AM and MH receive the same charging voltage, since the COs are charged in parallel.

As noted above, the authors found that this method works very well and significantly minimizes the problems of time fluctuations. However, it does not reduce them to such an extent that the laser operator can control the ZG independently of the CM. Nevertheless, there are a number of operational parameters of the ZG and UM, which can be controlled separately, to optimize the operation of each unit. These other parameters include laser gas pressure, F ₂ concentration, and laser gas temperature. Preferably, the control of these parameters is independent in each of the two cameras and their adjustment is carried out in a device controlled by the processor based on feedback.

Further improvement in optical quality

The present invention provides a laser system, which allows to obtain a much higher pulse energy and output power than known from the prior art single chamber gas discharge lasers with a high repetition rate. In this system, the master oscillator determines to a large extent the wavelength and bandwidth, and the power amplifier mainly controls the pulse energy. The pulse energy needed to effectively seed the power amplifier can be as small as a fraction of mJ. Since a master-type laser can easily generate 5 mJ pulses, it has unused energy. This additional pulse energy makes it possible to use special methods to improve the quality of the beam, which are not particularly effective in terms of energy.

These methods include:

- The pulse trim described in US patent No. 5852621, referred to here by reference.

The energy of the pulse is controlled, the pulse is delayed, and part of the delayed pulse is cut off using a very fast optical switch, such as a Pockels cell.

- Use of a line narrowing module with very large beam expansion and small apertures, as will be described later in this application.

- wave technology

Intraresonator wavefront correction can be added to or after the master oscillator. To do this, you can use a diffraction grating with one or more bends, as described in application for US patent No. 09/703317, referred to here by reference; static correction of the wavefront by a deformable mirror, for example, a non-planar mirror facet of a prism, made with the possibility of correction of known wavefront distortion.

- beam filtering

To reduce the bandwidth, you can add filters for the beam, such as the spatial filter described in application for US patent No. 09/309478, referred to here by reference, and indicated by 11 in Fig.23. The filter for the beam can be located inside the cavity of the ЗГ or between the ЗГ and the PA. It can also be installed after UM. A preferred spatial filter that does not require beam propagation through the focus is a common internal spatial filter, which is described in the next section.

- Coherence Management

Laser beam coherence can be a problem for integrated circuit manufacturers. Gas discharge lasers typically generate a laser beam having low coherence. However, since the bandwidth is made very narrow, this can increase the coherence of the output beam. For this reason, some forced spatial incoherence may be preferable. Preferably, optical elements are introduced to reduce the coherence either into the resonator of the MIND or between the MG and the AM. Several optical elements are known to reduce coherence, for example, movable phase plates or acoustic-optical devices.

- Aperture

The quality of the seed beam can also be improved by narrowing the beam aperture.

Common internal reflective spatial filter

Spatial filtering is effective in reducing integrated 95 percent bandwidth. However, all the methods of direct spatial filtration proposed earlier required at least the concentration of the beam and, in most cases, the actual focusing of the beam. In addition, all known designs require the use of multiple optical elements. A simple compact spatial filter that does not require a focused beam would be easier to fit for installation inside a laser resonator if spatial filtering is required.

A preferred filter is a single prism approximately 2 inches long. The input and output faces of the prism are parallel to each other and perpendicular to the incident beam. The other two faces are parallel to each other, but oriented at an angle equal to the critical angle relative to the input and output faces. At a wavelength of about 157 nm, the critical angle for CaF ₂ is 38.89 degrees. The only necessary coating is an antireflection coating for the normal incidence angle at the input and output faces of the prism.

This spatial filter works as follows. The beam enters with a normal angle of incidence on the input face of the prism. Then the beam propagates to the edge of the prism with a critical angle. If the beam were collimated, all the rays would fall on this second face at a critical angle. However, if the beam diverges or converges, some rays will fall on this surface at angles greater than or less than the critical angle. All rays incident on this face at a critical angle or more will be reflected 100%. Rays incident on this face at an angle less than critical will be reflected with values less than 100% and weaken. All reflected rays will fall on the opposite surface of the prism at the same angle and will also be weakened to the same degree. In the proposed design, there are only six reflections for each passage. The reflectance for p-polarized light at an angle of 1 mrad less than the critical angle is about 71%. Consequently, all rays with incidence angles different from the critical angle by 1 mrad or more will be transmitted to the output face with an intensity of less than 13% of their initial intensity.

However, one pass through this filter will be only one-way. All rays that fall at angles greater than the critical angle are reflected 100%. After exiting the prism of the spatial filter, the beam will fall on the mirror. Inside the laser cavity, this mirror may be an output coupler. After reflection from the mirror, the rays will again enter the prism of the spatial filter, but with one important difference.

All rays that come out of the spatial filter at angles greater than critical will be turned after reflection from the mirror. Now these rays will again enter the prism with values less than the critical angle and will weaken. It is the second pass through the prism that changes the transmitting function of the prism from a one-way filter to a true band-pass filter.

On figv shows the design of the spatial filter. The inlet and outlet faces of the prism are 1/2 inch thick. The edges of the critical angle are about 2 inches. The width of the input beam is 2.6 mm and represents the width of the beam along the short axis. This prism has a height of 1 inch in the plane of the drawing. The figure depicts three groups of rays. The first group of rays is collimated and falls on the surface at a critical angle. These are green rays. The second group of rays falls on the surface at an angle less than critical and breaks off at the first reflection. These are blue rays. These rays are more visible in an enlarged section. They are rays that are attenuated during the first pass. The last group of rays falls at an angle greater than critical. These rays pass the entire first pass, but break off at the first reflection of the second pass. They are rays that are attenuated in the second pass.

Telescope between cameras

In preferred embodiments, a cylindrical refracting telescope is provided between the output of the master oscillator and the input of the power amplifier. It controls the horizontal size of the beam entering the power amplifier. This telescope can also be performed using known methods for controlling horizontal divergence.

Metrology

In preferred embodiments of the present invention, the energy of each serial pulse is controlled using feedback from a high-speed photodiode energy monitor. In many applications, control of the wavelength and bandwidth of each successive pulse is not provided, since the natural wavelength and bandwidth of the main line F _{2 are} relatively constant. However, if necessary, it is possible to control the wavelength and bandwidth essentially the same as in the known excimer lasers, but at a wavelength range of 157 nm.

Preferably, at the output of the master oscillator, after power amplification and after amplification of the pulse, power monitors (p-cells) should be provided. Preferably, a p-cell should also be provided to control any back reflections to the master oscillator. Such back reflections can be amplified in the generator and damage the optical elements of the master oscillator. The back reflection signal from the back reflection monitor is used to turn off the laser if a hazardous threshold is exceeded. Also, the system should be made with the possibility of eliminating flicker in the path of the beam, which could cause any significant back reflection.

Next, measurement of the beam parameters and control of this laser will be described. The wavemeter used in the preferred embodiment is similar to the wavemeter described in US patent No. 5978394, and part of the description below is taken from this patent. At a wavelength of about 157 nm, elements for metrology of the wavelength and bandwidth are exposed to radiation, therefore, the authors recommend that these measurements be carried out periodically, and not at each pulse. For example, the wavelength and bandwidth can be controlled for every 30 pulses every 10 minutes. At this frequency, the metrological elements in F ₂ lasers should have a service life of at least comparable to the service life of KrF and ArF lasers. For this purpose, a shutter should be provided for the wave meter, to block the access of the beam to the elements for metrology of the wavelength and bandwidth.

The optical equipment in these nodes measures the pulse energy, wavelength and bandwidth. These measurements are used with feedback circuits to preserve the pulse energy and wavelength within specified limits.

A small part of the laser beam is reflected to an energy detector, which contains a very high-speed photodiode, capable of measuring the energy of individual pulses arising at a frequency of 4000 pulses per second. The pulse energy is about 10 mJ, and the output of the detector 69 is supplied to a computer controller that uses a special algorithm to adjust the charging voltage of the laser to precisely control the energy of future pulses based on the stored pulse energy data in order to limit changes in the energy of individual pulses and the integrated packet energy pulses.

Based on the energy measurement of each pulse described above, the energy of subsequent pulses is controlled to maintain the required pulse energies, as well as the required total integrated dose of a given number of pulses, as described in US Pat. as a reference.

Line selection

Prism Line Selector

In preferred embodiments, the strongest of the intrinsic resonance lines F ₂ is selected using the five-prism line selector shown in FIGS. 16A and B. These five prisms 112A-E are precision mounted on one prism plate (not shown) located in the BVL 10C shown in figure 1. BVL is located after the master oscillator at a small distance from the output coupler of the master oscillator. Each of the five prisms is a 65-degree prism (apex angle) and is horizontal, as shown in FIG. 16A. The angles of incidence for prisms in this particular embodiment are as follows: for prisms 112A-E, respectively: 79.6 °, 61.4 °, 47.7 ° and 42.1 °. (Many other prism configurations may provide similar results.)

The mirror 114B is arranged to reflect the beam upward to the mirror 114C, which reflects the beam through an aperture (not shown) in the discharge region of the power amplifier, as shown at 12A in FIG. 1. The five prism line selector creates an angular divergence of 10.56 milliradians between the F ₂ lines _of 157.63 nm and 157.52 nm, which provides a spatial separation of about 5.5 mm at a distance of about 0.5 meters behind the aperture in front of the power amplifier. This discrepancy is quite enough to highlight the 157.52 nm line.

Ring line selector

An alternative prism line selection unit is shown in FIG. It has a ring configuration. It can be introduced into the path of the beam without violating the direction of the beam. In this embodiment, the ring is formed by four prisms with an angle at the apex of 45 ° and four prisms with an angle at the apex of 65 °.

Lio Filter

An alternative to the prism line selector shown in FIG. 16A is a Lyo filter. This filter uses a birefringence dispersion of a non-isotropic crystalline material, such as MgF ₂ , to rotate the polarization of light depending on the wavelength. By appropriate selection of the crystal thickness, the total angle of rotation of the polarization of two wavelengths VUV (vacuum ultraviolet) can be substantially changed. These rotated waves can be discriminated by using polarization-dependent optical elements, such as Brewster windows in the gas discharge chamber of an F ₂ laser. One Brewster window made of CaF ₂ material provides a ratio of transmission intensity between p-polarized and s-polarized waves, respectively, equal to 1: 0.7. Losses on s-polarized waves are caused by reflections on surfaces. Since the camera has two windows, it should be considered that these values with the fourth power will give the correct ratio for the full path, equal to 1: 0.24. Optimum discrimination of one of the lines is achieved if the total angle of rotation of the polarization during double passage through the crystal is exactly 90 degrees. This can be achieved by adjusting the thickness of the crystal to the thickness of the characteristic quarter-wave plate. However, the wavelength of the other line should not be subjected to such a rotation; in fact, the total polarization rotation at this wavelength should be a full multiple of 180 degrees (half-wave plate) so that the discriminating elements do not affect the transmission of this wave. Therefore, the combination of a dispersive birefringent crystal, polarizing elements (Brewster windows) and a rear mirror (for the second pass back through the crystal) suppresses one of the wavelengths, while the other remains unaffected.

The advantages of this scheme are its inherent stability, ease of adjustment, fewer optical elements and the ability to use only antireflection coatings on the crystal for small (close to zero) angles of incidence.

FIG. 16C1 is a schematic view of a line-select laser system F ₂ comprising an Lio intracavity filter. The resonator is configured from a highly reflective mirror 116A, a birefringent dispersion crystal 116B, Brewster windows (in the chamber) 116C and 116D, and an output branch mirror 116E, which is partially reflective. Optical amplification is generated in a gas discharge in chamber 116F.

On figs shows another design in which one or more additional elements of the Brewster 116G is used to increase discrimination between p - and s-polarization.

On figs3 shows an alternative design of a highly reflective mirror and crystal. In fact, both elements can be combined by applying a dielectric reflective coating 116H directly to the back side of the crystal 116B to reduce the number of required optical elements.

Line selection on the back of the ZG

The embodiment of the invention depicted in figure 1, provides for the selection of the line after the master oscillator without selecting a line in the resonator of the master oscillator. Alternative options may include selecting a line in the cavity, for example, on the rear side of the laser chamber 10A. This line selector may be provided in addition to or instead of the line selector in FIG.

When using a prism line selector inside the cavity, it is more important that the optical loss in the selected polarization is minimized. A preferred design of such a line selector contains five CaF or MgF prisms oriented so that the entry and exit of the prism are at a Brewster angle (about 57.3 degrees). This allows the use of prisms without antireflection coatings. A drawing illustrating such a decision for selecting a line is shown in FIG. This device contains five prisms 118A with apex angles of 2X (90 ° -θ _in ), where θ _in is the Brewster angle. Reflective optical system 118B is implemented as a half prism with a maximum reflective coating 118C on the rear side.

Here, a new feature is the elimination of the need for an antireflection coating on the surface of the angle of incidence due to the use of the Brewster angle (providing a zero reflection coefficient for a correctly polarized laser) and the application of a reflective coating directly on the back surface of the optical system.

Prismatic output coupler

The output coupler of gas-discharge lasers made as generators is usually a partially reflecting mirror, which is usually a wedge-shaped optical element with one surface oriented transversely to the beam path and having a coating to reflect the desired part of the beam and transmit the rest. The other surface is often coated with an antireflection coating, and it can be oriented at an angle different from the transverse to the path of the beam, so that any reflections from this surface will not return to the amplification region.

Coated surfaces sometimes cause problems throughout their life if they are used with a high intensity UV range. On Fig shows a solution to this problem. In this case, the output coupler 120 is in the shape of a prism. The front surface (closest to the gain region) is oriented at the angle of least loss (for p-polarization), and the second surface is orthogonal to the refracted laser beam to provide a reflected beam for amplification. This design eliminates the need for an antireflection coating and also provides some additional spectral separation as a result of dispersions. In this F ₂ application, the prism consists of CaF ₂ with an apex angle of 32.7 degrees and a dip angle of 57.2 degrees. In this preferred embodiment, there is no coating on the second surface and approximately 4.7 percent of the Fresnel reflection provides sufficient reflection for the master oscillator.

Beam control using an optical element

Despite efforts to maintain constant conditions along the beam path, many laser operations, such as burst operation, which was described above, create transient conditions, which in some cases require significant transient control of the outgoing laser beam. This transient control can be corrected using an active beam direction control system, which includes a beam direction monitor and a beam direction control mechanism. In a preferred embodiment, the beam direction monitor is a bifurcated detector, also known as a bi-cell detector or segmented detector. This type of detector has two separate photosensitive elements, separated by a small gap. The ratio of the outputs of the two elements is a measure of the direction of the beam. The beam direction control mechanism may be a swivel mirror, preferably in the line selection unit 10C in FIG. 1. Alternatively, one of the prisms in the line selector unit shown in FIG. 16A may be rotatable. If all the prisms in the selector block are mounted on a prism plate, then the plate itself can be rotated. The drive providing this rotation is preferably a piezoelectric drive, or it can be a linear electric drive coil, or a stepper motor drive unit, or any similar drive unit. The control of the direction control mechanism should preferably comprise a processor programmed with an appropriate feedback algorithm, as well as additional electronic control and software allowing the operator to control the beam direction.

Beam control compensation using purge pressure

As indicated above, a small degree of beam control along its path can be provided by operations such as batch mode operation, which is usually used for lithography of integrated circuits. Even very slight changes in beam direction can be extremely undesirable. As noted above, methods for eliminating the causes of beam control can be used. In addition, unwanted changes in beam directions can be corrected by rotating optical elements, such as prisms or mirrors. Another approach is to correct for unwanted changes in beam direction by adjusting the purge gas pressure in parts of the beam path. In a preferred embodiment, the purge pressure in the line selection unit is controlled to compensate for changes in beam direction. The authors developed this method for the five-prism line selector shown in figa. The direction of the output beam for this five-prism configuration is related to the purge gas pressure within about 1 atm by the following coefficient: Δφ = 15 milliradians per atmosphere. Preferably, the beam direction is controlled by the bifurcated detector described above, and the feedback signal controls the pressure in the LEL by controlling the purge gas flow valve. Another solution is to use a temperature sensor to provide a feedback signal.

Gas management

In a preferred embodiment of the present invention, there is a gas control module shown in FIG. 1, which is configured to fill each chamber with an appropriate amount of laser gas. Preferably, appropriate controllers and processors are provided to maintain a constant gas flow into each chamber, to maintain a constant or approximately constant concentration of laser gas at the required levels. This can be accomplished using the methods described in US patent No. 6028880, or 6151349, or 6240117 (referred to here as a reference).

Another method for ensuring a constant flow of laser gas into the chambers, which the authors called the binary filling method, is to provide several (for example, 5) filling lines, with each consecutive line having openings allowing to double the flow of the previous line, and each line has a shut-off valve. The lowest flow line has openings allowing a minimum equilibrium gas flow to be obtained. By selecting the appropriate valve combinations for opening, almost any desired flow rate can be obtained. Preferably, an intermediate reservoir is provided between the orifice lines, and the laser gas source is maintained at a pressure approximately twice that of the laser chambers.

Vertical optical table

In preferred embodiments, two cameras and a laser optical system are mounted on a vertically oriented optical table. This table is preferably fixed to the laser frame using a three-point kinematic mount. One preferred embodiment of this arrangement is shown in FIG. 1C1. Metal strips are provided on the table 11 in sections A, B and C, where the table is mounted on the laser frame 4 (not shown in FIG. 1C1). A pivot hinge is provided in area A, which secures the table but allows it to rotate. At section B, a ball and a V-groove are provided that resist rotation in the plane of the lower surface of the table and rotation in the plane of the front surface of the table. A ball and a groove are provided in section C to prevent rotation around axis AB.

Laser cameras

4 kHz operation

Preferred embodiments of the invention are designed to operate at a pulse rate of 4000 pulses per second. Cleaning the discharge region from the gas subjected to the discharge between pulses requires a gas flow between electrodes 18A and 20A at a speed of up to about 67 m / s. To achieve this speed, the diameter of the tangential fan block was determined to be 5 inches (the length of the blade structure was 26 inches), and the rotation speed was increased to 3500 rpm. To achieve such characteristics in this embodiment, two motors are used, which together transmit up to about 4 kW of drive power of the fan blade structure. At a pulse frequency of 4000 Hz, the discharge will add about 12 kW of thermal energy to the laser gas. To remove the heat created by the discharge together with the heat added by the fan, four separate water-cooled finned heat exchangers 58A are provided. Engines and heat exchangers will be described in detail below.

In a preferred embodiment of the present invention, four finned water-cooled heat exchangers 58A are used, shown generally in FIG. 4. Each of these heat exchangers is somewhat similar to one heat exchanger, shown as 58 in FIG. 1, but has significant improvements.

Components of a heat exchanger

A cross-sectional drawing of one of the heat exchangers is shown in FIG. The middle section of the heat exchanger is cut out and both ends are shown. On figa shown on an enlarged scale the end of the heat exchanger, which takes into account thermal expansion and contraction.

The heat exchanger elements include a ribbed structure 302, which is made on a solid copper machine (CU 11000) and contains twelve fins 303 per inch. The flow of water passes through an axial channel having an inner diameter of 0.33 inches. A plastic turbulizer 306 located in the axial channel prevents the stratification of water in the channel and the formation of a hot boundary layer on the inner surface of the channel. Flexible flange block 304 is a welded block consisting of an inner flange 304A, a bellows 304B, and an outer flange 304C. The heat exchanger block contains three c-shaped seals 308, isolating the water flowing into the heat exchanger from the laser gas. Bellows 304B allows expansion and contraction of the heat exchanger relative to the chamber. A two-hole nut 400 connects the heat exchanger duct to a standard 5/16 inch positional angle fitting, which in turn is connected to a water source. O-ring 402 provides a seal between nut 400 and ribbed structure 302. In preferred embodiments, the direction of the cooling flow in two blocks is opposite to that in the other two to minimize axial temperature gradients.

Turbulator

In a preferred embodiment, the turbulator consists of four commercially available, long mixing elements that are typically used to mix epoxy components and are supplied by 3M Corporation (Static mixer. Part No. 06-D1229-00). These in-line mixers are shown at 306 in FIGS. 21 and 21A. They allow water to flow almost in a spiral path that crosses its direction clockwise at approximately every step distance (equal to 0.3 inches). The turbulator significantly improves the operation of the heat exchanger. Tests conducted by the authors showed that the addition of a turbulizer reduces the required water flow by about 5 times, while maintaining comparable gas temperature conditions.

Flow path and acoustic effects

In this preferred embodiment, the gas flow to and from the discharge region has been significantly improved compared to known laser chambers. The region above the discharge and near the outlet of the fan has such a shape that a smooth transition is formed from a large cross section to a small cross section of the discharge. The cross section of the region immediately after the discharge gradually increases for a small discharge to a much larger value before the gas is forced to rotate 90 ° into the heat exchangers. This arrangement minimizes the pressure drop and associated turbulence caused by the high-speed flow through sharp steps. Providing this smoothly and gradually expanding flow path away from the laser also reduces the negative acoustic effects caused by acoustic waves from a pulse reflected back into the discharge region during the next pulse. Methods of reducing these effects are described in US Pat. No. 6,212,211 and US Pat. No. 6,317,447, incorporated by reference. The time required to return the acoustic wave to the discharge region is largely dependent. As a result, reflection from a specific surface can be a problem only with a certain combination of repetition rate and gas temperature. If the reflective surface is difficult to exclude, the alternative may be a failure to work with a problematic combination of temperature-repetition rate. One solution could be to program the laser controller to automatically change the gas temperature if necessary, to avoid working with a problematic combination.

Blower Motors and Large Blower

In a first preferred embodiment of the present invention, there is provided a large tangential fan driven by two engines to circulate the laser gas. In a preferred arrangement, a gas flow is formed between the electrodes at a speed of 67 m / s, which is sufficient to clean a space of about 1.7 cm in the discharge region between pulses with a frequency of 4000 Hz.

A perspective view of the fan is shown in FIG. 18A. The blade structure has a diameter of 5 inches and is made on a machine made of a solid bar material of aluminum alloy 6061-T7. A separate blade in each section is slightly shifted relative to the adjacent section, as shown in FIG. 18A. This shift is preferably made uneven to prevent the formation of a compression wave front. Alternatively, individual blades may form a small angle relative to their axis (also to avoid the formation of compression wave fronts). The blades have very sharp leading edges to reduce acoustic reflections from the edge of the blade facing the discharge region.

In the embodiment of FIG. 18, two three-phase brushless DC motors are used, each with a magnetic rotor enclosed in a metal pressure cap that separates the stator portion of the motors from the laser gas medium, as described in US Pat. No. 4,950,840. In this embodiment, the pressure cap is made of 0.16-inch thin-walled nickel alloy 400, which acts as a barrier to the laser gas. Two motors 530 and 532 drive the same shaft and are programmed to rotate in opposite directions. Both motors do not contain sensors (i.e. they operate without position sensors). The right-hand motor controller 534 controlling the right-hand motor 530 acts as a master controller controlling the slave motor controller 536 via analog and digital signals, and issues commands for starting / stopping, current, current feedback, and the like. Communication with the controller 24A of the laser is carried out through the serial port RS-232 in the master controller 534.

Purge system

In a first embodiment of the present invention, an ultrapure N ₂ purge system is provided that significantly improves performance and substantially increases the life of the elements.

19 is a structural diagram of a purge system illustrating the main features of the first preferred embodiment of the invention. The five excimer laser components through which nitrogen gas is blown, according to this embodiment of the present invention, include the following components: BVL 2P, high-voltage elements 4P mounted on the laser chamber 6P, high-voltage cable 8P connecting high-voltage elements 4P with impulse elements located in front of them power supply 10P, output coupler 12P and wave meter 14P. Each of the components 2P, 4P, 8P, 12P and 14P is assembled in an airtight container or chamber, each of which has only two windows - a window for inlet N ₂ and a window for outlet N ₂ . The 16P N ₂ source is typically a large N ₂ tank (typically contained at liquid nitrogen temperature) in an integrated integrated circuit plant, but it can also be a relatively small N ₂ bottle. The source gas N ₂ exits the source N ₂ 16P, passes into the purge module N ₂ 17P and through the filter N ₂ 18P to the distribution panel 20P, containing valves for controlling the flow of N ₂ supplied to the blown elements. For each element, the purge flow is directed back to the module 17P to the flow monitor unit 22P, which controls the flow returning from each purge unit, and if the monitored flow is less than a predetermined value, an alarm (not shown) is activated.

On figa presents a linear diagram illustrating the specific elements of this preferred variant of implementation, having some additional features N ₂ not specifically associated with the elements of the purge according to the present invention.

Filter N ₂

An important feature of the present invention is the inclusion of an N ₂ filter 18. In the past, manufacturers of excimer lasers for integrated circuit lithography have assumed that a N ₂ purge gas filter is not necessary, since commercial N ₂ characteristics are almost always high enough, and gas corresponding to These characteristics are reasonably clean. However, the authors found that sometimes the source gas may not meet specifications, or the N ₂ lines leading to the purge system may contain contaminants. In addition, these lines may become contaminated during repairs or maintenance. The authors determined that the cost of such a filter is a good enough guarantee to exclude even a low probability of damage caused by pollution.

A preferred N ₂ filter is an inert gas purifier, Model 500K, supplied by Aeronex, Inc., San Diego, California. This filter removes H ₂ O, O ₂ , CO, CO ₂ , H ₂ and non-methane hydrocarbons to levels of parts per billion. It removes 99.9999999 percent of all particles 0.003 microns or larger.

Flow monitors

In block 22, a flow monitor is provided for each of the five blown elements. There are commercially available units that provide for the issuance of an alarm at low flow.

Pipelines

Preferably, all pipelines are made of stainless steel (316SST) with an electropolished inner side. You can also use some types of plastic pipes made of PFA 400 or ultra-pure Teflon.

Recycling and Cleaning

All or part of the purge gas may be recycled as shown in FIG. In this case, a blower and a water-cooled heat exchanger are added to the purge module. For example, the purge stream from the optical elements can be recycled, and the purge stream from the electrical components can be released into the atmosphere, or a part of the combined stream can be released. In addition, an ozone decontamination element may be added to remove ozone from the closed beam path. This may be a filter made of any material that reacts with O _3.

BVL purge with helium

In preferred embodiments of the invention, helium is blown through a line selection unit, and nitrogen is blown through the rest of the beam path. Helium has a much lower refractive index than nitrogen, so when using helium, the thermal effects in LEL are minimized. However, helium is approximately 1000 times more expensive than nitrogen. In addition, when using helium, it becomes difficult to control the control of the beam using the purge pressure.

Advanced seals

Preferred beam path fencing methods are described in US Patent Application No. 10/000991, filed November 14, 2001 for the invention, "An Advanced Beam Passage Discharge Laser," referred to herein by reference. 19F1, 2, 3, 4, and 5 show an easy-to-seal bellows seal that provides seals between the laser modules but allows quick disconnection of the modules to replace them.

Easy to seal bellows seal

The authors have developed an easy-to-seal bellows seal, which allows one to quickly isolate the beam path to a state comparable to vacuum when reinstalling laser modules in the beam path. It should be noted that, although these seals form vacuum-compatible seals of the corresponding parts of the beam path, the path itself does not work like a vacuum, but traditionally has pressures slightly higher than atmospheric.

Rapid sealing is essential, as there is a great need for the ability to replace these modules within minutes. The basic design of an easily sealed bellows seal is shown in FIGS. 8A-E. The easy-to-seal bellows seal consists of four parts. These four parts include the following: (1) the bellows portion 93A in FIG. 8A, the flange portion 93A shown in FIGS. 8A and 8B, the metal c-ring 93C shown in FIG. 8A, and the first extruded ring clip 93D shown in figs. An alternative second compression ring clamp is shown in FIG. An easy-to-seal bellows seal is shown assembled in FIG. 80. Two additional metal c-shaped seals can be used to seal the flange portion 93B with the first laser portion 93E and the bellows portion 93A with the second laser portion 93F. These additional seals are placed in the grooves 102 and 104. The flange part 93B is sealed with the first laser part by screws passing through the counter recessed holes 106 and tightened with a universal wrench through the holes 108.

Flange portion 93B comprises a tapering flange 120. This flange has a taper of 20 °, as shown in FIG. 8A. Flange 114 also has a constriction of 20 °. The compression clamp 93D is then opened by unscrewing the finger bolt 118 and placed around the conical flanges 120 and 114. The compression clamp 93D has a hinge section 122 and a bolt section 124. It has a conical inner surface with grooves consistent with the type of flanges 114 and 120. The diameter of the groove with bolt 118 fully inserted, slightly smaller than the aligned inclined surfaces of the flanges 114 and 120, so that when the bolt 118 is tightened, the two flanges are pressed against each other, clamping the seal 93C between them, and a vacuum-compatible seal. The authors determined that it is preferable to apply compression force of 400 pounds to guarantee the required vacuum seal. To do this, apply a torque of about 40 inches per pound to the handle of the bolt 118 of the first extruded ring clamp. In this preferred embodiment, the handle is only 1 inch long, so most mechanics will need a high-speed wrench (or similar tool) to provide 40 inches per pound. With a two-inch grip, the seal can be provided with finger force. The second easy-to-seal annular compression clamp shown in FIG. 8E presses the two conical flanges against each other as the curved arm of the lever 119 is pressed into place against the circumference of the ring. The clamp is opened by turning the lever away from the circumference of the ring, after which the two halves of the ring clamp can be separated. The authors estimated that a force of 40 pounds applied to the end of the shoulder of the lever 119 provides a compressive force of about 400 pounds on the c-seal 93C. This design of the clamp is based on the construction of commercially available clamps, known as knee-type pull-pull clamps.

This sealing system has the following important advantages:

(1) the seal can be formed in a short time (about 1-2 minutes);

(2) an excellent vacuum seal is obtained;

(3) substantial vibrational coupling between the camera and optical components is eliminated;

(4) sealing is relatively cheap compared to most other vacuum sealing methods.

The seal between the flange portion 93B and 93A is performed using a metal c-seal 93C located between the two parts, as shown in FIG. 8A, using the extruded ring clamp 93D shown in FIG. 80. The metal seal is seated in a groove 110. The seals in these embodiments are slightly oval to fit a circular groove 110. The longer the diameter of the c-ring is 1.946 inches and the shorter diameter is 1.84 inches. The spring force of the oval c-shaped o-ring presses it against the edge of the groove 110, which prevents the c-shaped o-ring from falling out during assembly. The bellows portion 93A includes an annular protrusion 112 that protects the seal ring 93C from scratches caused by part 93B when these two parts slip relative to each other during assembly.

Crossflow purge gas

The authors found experimentally that F ₂ lasers at high repetition frequencies, such as 4000 Hz, and several millijoules per pulsed beam will undergo transient divergence and reflection processes caused by the interaction of the beam with the purge gas. The authors also found that these effects can be minimized by forming purge flows transversely to the beam path. The authors found several ways to implement this mode. Four such methods are shown in FIGS. 19C1, 19C2, 19C3 and 19C4. On figs C1 and 2 shows the deflectors in the path of the beam, which provide a transverse purge flow. Fig. 19C3 shows a fan on the purge line for recirculating the purge gas, and in Fig. 19C4, the purge flow is directed by the purge nozzles transversely to the beam path.

Laser sight

Providing a fenced vacuum-type purge path complicates the alignment of the laser optical system. In known systems, the purge path needs to be broken in order to introduce a small visible laser laser. In preferred embodiments, the small visible laser can be turned on as a constant part of the beam path, which is very useful during maintenance operations. Preferably, a helium and neon laser or a small diode laser mounted on the back side of the highly reflective 10D mirror, which in this design should consist of a CaF plate having a dielectric reflective coating designed to reflect a very high part of 157 nm ultraviolet, is preferably used as a laser light, but transmitting a high portion of visible light. The laser-sight can be used to align the entire path of the beam through the 3G, PA and pulse expander without interrupting the path of the beam. (It is clear that any optical system for selecting a line will change the direction of the laser-sight. The line optics can be removed to align the rest of the system, or you can adjust the beam direction of the alignment laser-sight to take into account the optics of line selection).

Advantages of a beam path with vacuum quality

The vacuum-quality purge system described above represents a major improvement to ensure long-term operation of an excimer laser, especially F ₂ lasers. It basically solved the problems of contamination, which leads to a significant increase in the service life of the components and an improvement in the quality of the beam. In addition, since leakage is excluded, with the exception of the exhaust windows, the flow can be adjusted to the required values, which reduces the need for N _{2 by} almost 50 percent.

Sealed shutter unit with power meter

In a first preferred embodiment of the invention, there is provided a sealed shutter 500 with an integrated power meter, as shown in FIGS. 20, 20A and 20B. In this important improvement, the shutter has two functions: firstly, it acts as a shutter blocking the laser beam, and secondly, as a full beam power meter to control beam power when measurement is required.

On Fig presents a top view showing the main elements of the shutter block. These include a gate 502, a beam dump 504, and a power meter 506. The output path of the laser beam with the shutter in the closed position is shown as 510 in FIG. An open beam path is shown as 512. The active gate surface of the beam stop element 516 is located at an angle of 45 ° to the direction of the beam exiting the chamber, and when the shutter is closed, the beam is absorbed by the gate surface and is reflected to the beam dump 504. Both the active surface of the beam dump and the active surface of the shutter are coated with chrome for high absorption of the beam. In this embodiment, the beam stop element 516 is mounted on a flexible spring steel lever 518. The shutter opens when current is supplied to the coil 514, as shown in FIG. 20B, which pulls the flexible lever 518 and the beam stop element 516 to the coil, retracting the beam stop element 516 with paths of the output laser beam. The shutter closes when the current flow through the coil 514 ceases, which allows the permanent magnets 520 to pull the beam stop element 516 and the flexible arm 518 back to the closed position. In a preferred embodiment, the current flow is carefully selected to allow easy movement of these element and lever between open and closed positions.

The power meter 506 operates in a similar manner by placing a pyroelectric photodetector in the path of the output laser beam, as shown in FIGS. 20 and 20A. In this case, the coil 520 and magnets 522 attract the detector unit 524 and its flexible arm 526 to and from the beam path to measure the output power. This power meter can operate with both an open and a closed shutter. Current is supplied to the coil in the same way as in the case of the shutter, which is controlled to ensure easy transition of block 524 to and from the beam path.

Enhanced polarization with a camera window angle greater than the Brewster angle

Known camera windows are often positioned at the Brewster angle to provide about 100% transmission in the s-polarization direction with about 58% in the p-polarization direction. In other known constructions, windows are arranged at an angle of about 45 degrees, in this case, the s-polarization transmission is slightly less, and the p-polarization is slightly larger than the indicated values.

For F ₂ lasers, the authors determined that when the camera windows are located at an angle slightly less than the Brewster angle, there is significant competition in the gain region between s and p polarizations. This is because usually in the discharge region of F ₂ lasers there is a very large gain compared to KrF and ArF lasers. This competition is not desirable, since in most applications s-polarized light is not useful and is usually lost as unwanted heat. Therefore, there is a need to minimize the s-polarization generated in the laser.

The preferred easy-to-implement method is to significantly increase the angle of incidence for camera windows compared to the Brewster angle. For example, at the Brewster angle, about 100% p-polarization and about 83% s-polarization are transmitted for the F ₂ 157 nm beam.

If the angle of incidence is increased to 64 °, the p-polarization transmission will decrease to about 99%, but only 76% s-polarization will be transmitted. Since the output light from the amplification region makes about two passes through each of the two windows in the master oscillator (for 4 passes through the windows and 2 surfaces for each window), the ratio of these two polarizations in the output beam (assuming that the angles of the windows are 64 ° ) is:

In tests on windows with an angle of 47 °, the authors received about 72% of the light with p-polarization and 28% with s-polarization.

When changing the window angle by 64 ° and adding an additional window with an angle of 64 ° in front of the highly reflective mirror 10D according to Fig. 1, the percentage of s-polarization in the output beam decreases to about 4%, and the remaining 96% of the light has p-polarization.

Various modifications may be made to the present invention without altering its scope. For specialists, many other possible options will be apparent. For example, the switching power supply may be a conventional circuit before the output of the switching transformer 56 of FIG. 5A. This solution can further reduce fluctuations, as explained in application for US patent No. 09/848043, referred to here by reference. Fig.3B of this application from the prior art with the image of the input and output of a pulse transformer is included in the present application as Fig.13 for the convenience of the reader. Other heat exchanger configurations may be obvious modifications to the configuration shown here. For example, all four nodes can be combined into one. Significant advantages can be achieved by using fins on a heat exchanger of a much larger size to reduce the effects of rapid changes in gas temperature that occur as a result of laser operation in batch mode. It is clear that at extremely high pulse frequencies, feedback control on the pulse energy does not have to be fast enough to control the energy of a particular pulse using the immediately preceding pulse. For example, control methods may be provided in which the measured energy of a particular pulse is used to control a second or third subsequent pulse. You can use many other layouts other than the layout shown in figure 1. For example, cameras can be installed side by side, or arrange the MIND from below. In addition, the second laser unit can be performed as a slave generator by including an output coupler, for example a partially reflecting mirror. Other options are possible. In addition to tangential fans, other fans can also be used. This may be required with a repetition frequency far exceeding 4 kHz. Fans and a heat exchanger can be located outside the discharge chambers. You can also use the methods for calculating the pulse time described in application for US patent No. 09/837035 (referred to here as a reference). Line selection methods other than the five prism design described above may be used. For example, a strong line can be selected using 3, 4 or 6 prisms and using the construction methods described above. It may be preferable to make accurate measurements of the bandwidth. This can be done using a standard having a smaller free spectral range than the standards described above. Other well-known methods can be adapted to accurately measure the width of the strip. Thus, the above description should not be construed as limiting, and the scope of the invention should be determined on the basis of the attached claims and legal equivalents.

Claims (154)

1. The system is a very narrow-band double-chamber gas discharge F ₂ laser with a high repetition rate, containing A) a first laser unit containing 1) the first discharge chamber accommodating a) the first laser gas, b) a first pair of elongated spaced apart electrodes forming a first discharge region, 2) the first fan to ensure sufficient speeds of the first laser gas in the first region of the discharge to remove from the first region of the discharge after each pulse essentially all of the ions formed as a result of the discharge before the next pulse when operating with a repetition rate of 4000 pulses per second or higher, 3) a first heat exchanger system configured to remove at least 16 kW of thermal energy from the first laser gas, B) a line selection unit for minimizing energy outside the spectrum of one selected line, C) a second laser unit containing 1) a second discharge chamber containing a) a second laser gas, b) a second pair of elongated spaced apart electrodes forming a second discharge region, 2) a second fan to ensure sufficient speeds of the second laser gas in the second region of the discharge to remove from the second region of the discharge after each pulse essentially all of the ions formed as a result of the discharge before the next pulse when operating at a repetition rate of 4000 pulses per second or higher, 3) a second heat exchanger system, configured to remove at least 16 kW of thermal energy from the second laser gas, D) a pulsed power system configured to supply electric pulses to the first pair of electrodes and the second pair of electrodes, sufficient to generate laser pulses with a frequency of about 4000 pulses per second with precisely controlled pulse energy above about 5 mJ, E) a laser beam measurement and control system for measuring the energy of the output laser pulses generated by the two-chamber laser system and for controlling the output laser pulses in a feedback control device, the output laser beams from the first laser unit being used as a seed beam to seed the second laser unit . 2. The laser system according to claim 1, in which the first laser unit is made in the form of a master oscillator, and the second laser unit is made in the form of a power amplifier. 3. The laser system according to claim 2, in which the first laser gas contains fluorine and neon. 4. The laser system according to claim 2, in which the first laser gas contains fluorine and helium. 5. The laser system according to claim 2, in which the first and second laser gas contains fluorine and a buffer gas selected from the group consisting of neon, helium, or a mixture of neon and helium. 6. The laser system according to claim 2, in which the power amplifier is configured to one passage of the beam through the second region of the discharge. 7. The laser system according to claim 2, in which the power amplifier is configured to have multiple passes through the second region of the discharge. 8. The laser system according to claim 2, in which the master oscillator contains optical components that form a resonant path for making two passes through the first region of the discharge. 9. The laser system according to claim 2, in which the master oscillator contains optical components that provide a resonant path for two passes through the first region of the discharge, and the power amplifier contains optical components that provide multiple passes of the beam through the second region of the discharge. 10. The laser system according to claim 1, wherein said first laser unit comprises an optical resonator system and said laser system further comprises an optical table for holding the optical system of the resonator of the first laser unit independently of the first discharge chamber. 11. The laser system of claim 10, in which the optical table is U-shaped and forms a U-shaped cavity in which the first discharge chamber is installed, 12. The laser system according to claim 1, which further comprises a vertically mounted optical table with first and second discharge cameras mounted on it. 13. The laser system according to claim 1, in which each of the first and second laser chambers forms a gas flow path with a gradually increasing cross section behind the electrodes, allowing you to restore a large percentage of the static pressure drop occurring in the discharge region. 14. The laser system according to claim 2, in which each of the first and second chambers contains a blade structure behind the discharge region to normalize the gas velocity after the discharge region. 15. The laser system according to claim 1, in which the first fan and the second fan are tangential fans and each contains a shaft driven by two brushless DC motors. 16. The laser system according to clause 15, in which the engines are water-cooled engines. 17. The laser system according to clause 15, in which each of the engines contains a stator and each of the engines contains a magnetic rotor enclosed in a pressure cap that separates the stator from the laser gas. 18. The laser system according to claim 1, in which the first and second fans are tangential fans, each of which contains a blade structure made on the machine from one part of aluminum raw materials. 19. The laser system of claim 18, wherein said blade structure has an outer diameter of about five inches. 20. The laser system according to claim 19, in which the blade structure contains blade elements having sharp leading edges. 21. The laser system according to clause 15, in which the engines do not have sensors and which further comprises a master motor controller for controlling one of the engines and a slave motor controller for controlling the other engine. 22. The laser system according to clause 15, in which each tangential fan contains blades located at an angle to the shaft. 23. The laser system according to claim 1, in which each heat exchanger system is cooled by water. 24. The laser system of claim 23, wherein each heat exchanger system comprises at least four separate water-cooled heat exchangers. 25. The laser system according to item 23, in which each heat exchanger system contains at least one heat exchanger having a tubular channel for the flow of water, in the path of which at least one turbulator is located. 26. The laser system according to paragraph 24, in which each of the four heat exchangers contains a tubular channel for the flow of water with a turbulator located in it. 27. The laser system of claim 1, wherein the pulsed power system comprises water-cooled electrical elements. 28. The laser system according to item 27, in which at least one of the water-cooled elements operates at high voltages above 12000 V. 29. The laser system of claim 28, wherein the high voltage is isolated from the earth using an inductor through which cooling water flows. 30. The laser system of claim 1, wherein the pulsed power system comprises a first battery of charging capacitors and a first pulse compression circuit for supplying electrical pulses to the first pair of electrodes, and a second battery of charging capacitors and a second pulse compression circuit for supplying electrical pulses to the second pair electrodes, and a resonant charging system for charging in parallel the first and second batteries of the charging capacitors to precisely regulated voltage. 31. The laser system of claim 30, wherein the resonant charging system comprises a De-Qing circuit. 32. The laser system of claim 30, wherein the resonant charging system comprises a bleed circuit. 33. The laser system of claim 30, wherein the resonant charging system comprises a De-Qing circuit and a bleed circuit. 34. The laser system according to claim 1, in which the pulsed power system comprises a charging system consisting of at least three power sources arranged in parallel. 35. The laser system according to claim 1, in which the selection unit is located after the master oscillator. 36. The laser system according to clause 35, in which the line selection unit contains many prisms. 37. The laser system of claim 36, wherein the plurality of prisms consists of five prisms. 38. The laser system of claim 36, wherein the plurality of prisms are arranged in a loop to allow the laser beams to rotate 360 ° from the first laser unit 360 ° before entering the second laser unit. 39. The laser system according to claim 1, which further comprises a laser-visor of visible light. 40. The laser system according to claim 1, in which the line selection unit contains a Lyo filter. 41. The laser system of claim 30, wherein the first discharge chamber and the second discharge chamber comprise windows arranged such that all angles of incidence of the laser beams on said windows are greater than the Brewster angle. 42. The laser system according to claim 1, which further comprises a beam control means for controlling laser beams generated in the first laser unit. 43. The laser system of claim 42, wherein the control means comprises means for rotating the optical element. 44. The laser system of claim 42, wherein the beam control means comprises means for adjusting pressure in the line selection unit. 45. The laser system according to claim 1, in which the laser system comprises a prismatic output coupler, partially forming a resonator for the first laser unit, the prismatic output coupler containing two surfaces, the first of which is oriented at an angle with low loss for p-polarization, and the second located orthogonally to the laser beams from the first laser unit. 46. The laser system according to claim 1, which further comprises A) a first temperature monitor for monitoring the temperature of the gas in the first discharge chamber, B) a first gas temperature control system containing a control algorithm for adjusting the gas temperature to prevent negative acoustic effects caused by reflected acoustic waves. 47. The laser system according to claim 1, which further comprises A) a second temperature monitor for monitoring the gas temperature in the second discharge chamber, B) a second gas temperature control system containing a control algorithm for adjusting the gas temperature in order to eliminate negative acoustic effects caused by reflected acoustic waves. 48. The laser system according to claim 1, which further comprises a nitrogen filter. 49. The laser system according to claim 1, which further comprises a nitrogen purge system comprising a purge module with flow monitors, the laser also comprising tubes for transporting exhaust purge gas from the laser. 50. The laser system according to claim 1, which further comprises a shutter unit containing an electric shutter, and a power meter that can be positioned along the path of the output laser beam by the control signal. 51. The laser system according to claim 1, which further comprises a beam enclosure system comprising A) at least one beam seal containing a metal bellows, and B) purge means for purging the beam enclosure with purge gas. 52. The laser system of claim 51, wherein the beam shielding means comprises a flow guiding means for generating a purge stream perpendicular to the laser beams generated in the second laser unit. 53. The laser system of claim 51, wherein said at least one beam seal is configured to easily replace the laser camera. 54. The laser system of claim 51, wherein said at least one beam seal does not contain an elastomer, provides isolation from camera vibrations, isolates the beam path from atmospheric gases, and allows the laser camera to be easily replaced without disturbing the line selection unit. 55. The laser system of claim 51, wherein said at least one beam seal is comparable to a vacuum. 56. The laser system of claim 55, wherein said at least one beam seal is a plurality of beam seals, which use easily sealed bellows seals that are easily removable manually. 57. The laser system of claim 1, wherein said measuring and control system comprises a primary beam splitter for separating a small percentage of each output laser pulse from the second laser unit, and optical means for directing a portion of this small percentage to the pulse energy detector, and an isolation means for isolating the volume limited, at least in part, by the primary beam splitter and the window of the pulse energy detector from other parts of the measurement and control system, for forming in isolation area. 58. The laser system of claim 57, further comprising purge means for purging the insulated region with purge gas. 59. The laser system according to claim 1, which is configured to operate as a KrF laser system, or an ArF laser system, or an F ₂ laser system with minor modifications. 60. The laser system according to claim 1, in which essentially all the components are enclosed in the laser housing, but the system contains an AC / DC module physically separate from the laser housing. 61. The laser system according to claim 1, in which the pulsed power system comprises a battery of charging capacitors of the master oscillator, a battery of charging capacitors of a power amplifier and a resonant charger configured to charge both batteries of charging capacitors in parallel. 62. The laser system of claim 61, wherein the pulsed power system comprises a power source configured to supply at least 2000 V of power to the resonant charger. 63. The laser system according to claim 1, which further comprises a gas control system for controlling the concentration of F ₂ in the first laser gas to control the parameters of the beam of the master oscillator. 64. The laser system according to claim 1, which further comprises a gas control system for regulating the pressure of the first laser gas to control the parameters of the beam of the master oscillator. 65. The laser system according to claim 2, which further comprises a controller for calculating the discharge time for starting charges in the power amplifier, so that they occur 20-60 ns after the discharges in the master oscillator. 66. The laser system according to claim 2, which further comprises a discharge controller programmed to cause discharges in some circumstances at a point in time, avoiding significant energy of the output pulse. 67. The laser system of claim 66, wherein said controller is programmed in some circumstances to cause a discharge in the power amplifier at least 20 ns earlier than the discharge in the master oscillator. 68. The laser system according to claim 1, which further comprises a pulse multiplier unit for increasing the duration of the laser output pulses. 69. The laser system of claim 68, wherein the pulse multiplier unit is configured to receive the laser output pulse and multiply the number of laser output pulses per second by at least two times to generate one multiplied laser output pulse beam consisting of a larger number of pulses with significantly lower intensities than the laser output pulses, and said pulse multiplier unit contains (1) a first beam splitter designed to separate part of the laser output of the pulsed beam, the separated part forming the delayed part, and the laser output pulsed beam determines the size and angular divergence of the beam in the first beam splitter, (2) the first delay path starting and ending in the first beam splitter, the first delay path containing at least two focusing mirrors configured to focus said delayed part in focus on the first delay path and return the delayed part to the first beam splitter with a size and angular divergence it beam equal to or approximately equal to the beam size and angular divergence of the laser output pulse beam in the first beam splitter. 70. The laser system of claim 69, wherein the at least two focusing mirrors are spherical mirrors. 71. The laser system of claim 69, further comprising a second delay path comprising at least two spherical mirrors. 72. The laser system of claim 69, wherein the first delay path comprises four focusing mirrors. 73. The laser system of claim 72, further comprising a second delay path formed by a second beam splitter located on the first delay path. 74. The laser system of claim 69, wherein said first delay path comprises a second beam splitter, the system further comprising a second delay path comprising at least two focusing mirrors arranged to focus the delayed portion in focus on the first delay path and returning the delayed part to the first beam splitter with a size and angular divergence of the beam equal to or approximately equal to the size and angular divergence of the laser output pulse beam in the first beam splitter. 75. The laser system of claim 69, wherein the first beam splitter is configured to direct the laser beam in at least two directions using the optical property of the truncated internal reflection. 76. The laser system of claim 69, wherein the first beam splitter consists of two transparent optical elements, each of which has a flat surface, both optical elements being arranged so that their first surfaces are parallel to each other and separated by a distance of less than 200 nm. 77. The laser system of claim 69, wherein the first beam splitter is an uncoated optical element oriented at an angle to the laser output pulse beam to achieve the desired reflection-transmission ratio. 78. The system is very narrow-band double-chamber gas discharge F ₂ laser with a high repetition rate containing A) a first laser unit containing 1) the first discharge chamber accommodating a) the first laser gas, b) a first pair of elongated spaced apart electrodes forming a first discharge region, 2) the first fan to ensure sufficient movement of the first laser gas in the first region of the discharge to remove from the first region of the discharge after each gas discharge essentially all of the ions formed as a result of the discharge before the next gas discharge when operating at a repetition rate of 4,000 gas discharges per second or above, 3) a first heat exchanger system for removing heat energy from a first laser gas, B) a line selection unit for minimizing energy outside the spectrum of one selected line, C) a second laser unit containing 1) a second discharge chamber containing a) a second laser gas, b) a second pair of elongated spaced apart electrodes forming a second discharge region, 2) a second fan to ensure sufficient movement of the second laser gas in the second region of the discharge to remove from the second region of the discharge after each gas discharge essentially all of the ions formed as a result of the discharge before the next gas discharge when operating at a repetition rate of 4,000 gas discharges per second or above, 3) a second heat exchanger system for removing thermal energy from the second laser gas, D) a pulsed power supply system configured to supply electric pulses to the first pair of electrodes and the second pair of electrodes sufficient to generate laser output pulses with a frequency of about 4000 laser output pulses per second with precisely controlled laser output pulse energies above about 5 mJ, E) a laser beam measurement and control system for measuring the energy of the laser output pulses generated by the two-chamber laser system and for controlling the laser output pulses in a feedback control device, the output laser beams from the first laser unit being used as a seed beam to seed the second laser unit . 79. The laser system according to p, in which the first laser unit is made in the form of a master oscillator, and the second laser unit is made in the form of a power amplifier. 80. The laser system of claim 79, wherein the first laser gas contains fluorine and neon. 81. The laser system of claim 79, wherein the first laser gas contains fluorine and helium. 82. The laser system of claim 80, wherein the first and second laser gas contains fluorine and a buffer gas selected from the group consisting of neon, helium, or a mixture of neon and helium. 83. The laser system according to claim 80, in which the power amplifier is configured to one passage of the beam through the second region of the discharge. 84. The laser system of claim 79, wherein the power amplifier is configured to have multiple passes through a second discharge region. 85. The laser system according to p, in which the master oscillator contains optical components that form a resonant path for making two passes through the first region of the discharge. 86. The laser system of claim 79, wherein the master oscillator comprises optical components that provide a resonant path for making two passes through the first discharge region, and the power amplifier contains optical components that provide multiple beam passes through the second discharge region. 87. The laser system of claim 78, wherein said first laser unit comprises a cavity optical system and said laser system further comprises an optical table for holding a cavity optical system of a first laser block independently of the first discharge chamber. 88. The laser system of claim 87, wherein the optical table is U-shaped and forms a U-shaped cavity in which the first discharge chamber is mounted. 89. The laser system of claim 78, further comprising a vertically mounted optical table with first and second discharge cameras mounted thereon. 90. The laser system of claim 78, wherein each of the first and second laser chambers forms a gas flow path with a gradually increasing cross section behind the electrodes, allowing a large percentage of the drop in static pressure that occurs in the discharge region to be restored. 91. The laser system of claim 79, wherein each of the first and second chambers comprises a blade structure behind the discharge region to normalize the gas velocity after the discharge region. 92. The laser system of claim 78, wherein each of the first fan and the second fan is a tangential fan and each comprises a shaft driven by two brushless DC motors. 93. The laser system of claim 92, wherein the engines are water cooled engines. 94. The laser system according to paragraph 92, in which each of the engines contains a stator and each of the engines contains a magnetic rotor enclosed in a pressure cap that separates the stator from the laser gas. 95. The laser system according to p, in which each of the first and second fans is a tangential fan containing a blade structure made on the machine from one part of aluminum raw materials. 96. The laser system of claim 95, wherein said blade structure has an outer diameter of about five inches. 97. The laser system of claim 96, wherein the blade structure comprises blade elements having sharp leading edges. 98. The laser system according to paragraph 92, in which the engines do not have sensors and which further comprises a master motor controller for controlling one of the engines and a slave motor controller for controlling the other motor. 99. The laser system according to paragraph 92, in which each tangential fan contains blades located at an angle to the shaft. 100. The laser system of claim 78, wherein each heat exchanger system is water cooled. 101. The laser system of claim 100, wherein each heat exchanger system comprises at least four separate water-cooled heat exchangers. 102. The laser system according to claim 100, wherein each heat exchanger system comprises at least one heat exchanger having a tubular channel for the flow of water, at least one turbulator is located in its path. 103. The laser system of claim 101, wherein each of the four heat exchangers comprises a tubular channel for the flow of water with a turbulator located therein. 104. The laser system of claim 78, wherein the pulsed power system comprises water-cooled electrical elements. 105. The laser system of claim 104, wherein at least one of the water-cooled elements operates at high voltages above 12,000 V. 106. The laser system of claim 105, wherein the high voltage is isolated from the earth using an inductor through which cooling water flows. 107. The laser system of claim 78, wherein the pulsed power system comprises a first battery of charging capacitors and a first pulse compression circuit for supplying electrical pulses to the first pair of electrodes, and a second battery of charging capacitors and a second pulse compression circuit for supplying electrical pulses to the second pair electrodes, and a resonant charging system for charging in parallel the first and second batteries of the charging capacitors to precisely regulated voltage. 108. The laser system of claim 107, wherein the resonant charging system comprises a De-Qing circuit. 109. The laser system of claim 107, wherein the resonant charging system comprises an etching circuit. 110. The laser system of claim 107, wherein the resonant charging system comprises a De-Qing circuit and a bleed circuit. 111. The laser system of claim 78, wherein the pulsed power system comprises a charging system consisting of at least three power sources arranged in parallel. 112. The laser system of claim 78, wherein the selection unit is located after the master oscillator. 113. The laser system of claim 112, wherein the line selection unit comprises a plurality of prisms. 114. The laser system of claim 113, wherein the plurality of prisms consists of five prisms. 115. The laser system of claim 113, wherein the plurality of prisms are arranged in a loop to allow the laser beams to rotate 360 ° from the first laser unit to enter the second laser unit. 116. The laser system of claim 78, further comprising a laser-sighting laser. 117. The laser system of claim 78, wherein the line selection unit comprises a Lyo filter. 118. The laser system of claim 78, wherein the first discharge chamber and the second discharge chamber comprise windows arranged such that all angles of incidence of the laser beams on said windows are greater than the Brewster angle. 119. The laser system of claim 78, further comprising a beam control means for controlling laser beams generated in the first laser unit. 120. The laser system according p, in which the control means comprises means for rotating the optical element. 121. The laser system according p, in which the beam control means comprises means for adjusting the pressure in the line selection unit. 122. The laser system of claim 78, wherein the laser system comprises a prismatic output coupler partially forming a resonator for the first laser unit, the prismatic output coupler containing two surfaces, the first of which is oriented at an angle with low loss for p-polarization, and the second located orthogonally to the laser beams from the first laser unit. 123. The laser system of claim 78, which further comprises A) a first temperature monitor for monitoring the gas temperature in the first discharge chamber, B) a first gas temperature control system comprising a control algorithm for adjusting the gas temperature to eliminate negative acoustic effects caused by reflected acoustic waves. 124. The laser system of claim 78, which further comprises A) a second temperature monitor for monitoring the gas temperature in the second discharge chamber, B) a second gas temperature control system containing a control algorithm for adjusting the gas temperature to eliminate negative acoustic effects caused by reflected acoustic waves. 125. The laser system of claim 78, further comprising a nitrogen filter. 126. The laser system of claim 78, further comprising a nitrogen purge system comprising a purge module with flow monitors, the laser also comprising tubes for transporting exhaust purge gas from the laser. 127. The laser system of claim 78, further comprising a shutter unit comprising an electric shutter and a power meter that can be positioned along the path of the output laser beam from the control signal. 128. The laser system of claim 78, further comprising a beam guarding system comprising A) at least one beam seal comprising a metal bellows, and B) purge means for purging the beam enclosure with purge gas. 129. The laser system of claim 128, wherein the beam shielding means comprises flow guiding means for generating a purge stream perpendicular to the laser beams generated in the second laser unit. 130. The laser system of claim 128, wherein said at least one beam seal is configured to easily replace the laser camera. 131. The laser system of claim 128, wherein said at least one beam seal does not contain an elastomer, provides isolation from camera vibrations, isolates the beam path from atmospheric gases, and allows the laser camera to be easily replaced without disturbing the line selection unit. 132. The laser system of claim 128, wherein said at least one beam seal is comparable to a vacuum. 133. The laser system of claim 132, wherein said at least one beam seal is a plurality of beam seals, which utilize easily sealed bellows seals that are easily removable by hand. 134. The laser system of claim 78, wherein said measuring and control system comprises a primary beam splitter for separating a small percentage of each output laser pulse from the second laser unit, and optical means for directing a portion of this small percentage to the pulse energy detector, and an isolation means for isolating the volume limited, at least in part, by the primary beam splitter and the window of the pulse energy detector from other parts of the measurement and control system, to form an isolated oh area. 135. The laser system of claim 134, further comprising purge means for purging the insulated region with purge gas. 136. The laser system of claim 78, which is configured to operate as a KrF laser system, or an ArF laser system, or an F ₂ laser system with minor modifications. 137. The laser system of claim 78, wherein substantially all components are enclosed in the laser housing, but the system comprises an AC / DC module physically separate from the laser housing. 138. The laser system of claim 78, wherein the pulse power system comprises a charging capacitor bank of a master oscillator, a battery of charging capacitors of a power amplifier and a resonant charger configured to charge both charging capacitor banks in parallel. 139. The laser system according to p, in which the pulsed power system contains a power source configured to supply at least 2000 V power to a resonant charger. 140. The laser system of claim 78, further comprising a gas control system for controlling a concentration of F ₂ in the first laser gas to control beam parameters of the master oscillator. 141. The laser system of claim 78, further comprising a gas control system for controlling a pressure of the first laser gas for controlling beam parameters of a master oscillator. 142. The laser system of claim 79, further comprising a controller for calculating the discharge time for triggering the charges in the power amplifier so that they occur 20-60 ns after the discharges in the master oscillator. 143. The laser system of claim 79, further comprising a discharge controller programmed to cause discharges in some circumstances at a point in time to avoid significant output pulse energy. 144. The laser system of claim 143, wherein said controller in some circumstances is programmed to cause a discharge in the power amplifier at least 20 ns earlier than the discharge in the master oscillator. 145. The laser system of claim 78, further comprising a pulse multiplier unit for increasing the duration of the laser output pulses. 146. The laser system of claim 145, wherein the pulse multiplier unit is configured to receive the laser output pulse beam and multiply the number of pulses per second by at least two times to generate one multiplied laser output pulse beam consisting of a larger number of laser outputs pulses with significantly lower intensities than the laser output pulses, wherein said pulse multiplier unit contains (1) a first beam splitter designed to separate a part of the laser of the output pulse beam, wherein the separated part forms the delayed part, and the laser output pulse beam determines the size and angular divergence of the beam in the first beam splitter, (2) the first delay path starting and ending in the first beam splitter, the first delay path containing at least two focusing mirrors configured to focus said delayed part in focus on the first delay path and return the delayed part to the first beam splitter with a size and angular a beam divergence equal to or approximately equal to the beam size and the angular divergence of the laser output pulse beam in the first beam splitter. 147. The laser system of claim 146, wherein the at least two focusing mirrors are spherical mirrors. 148. The laser system of claim 144, further comprising a second delay path comprising at least two spherical mirrors. 149. The laser system of claim 126, wherein the first delay path comprises four focusing mirrors. 150. The laser system of claim 149, further comprising a second delay path formed by a second beam splitter located on the first delay path. 151. The laser system of claim 146, wherein said first delay path comprises a second beam splitter, wherein the laser further comprises a second delay path comprising at least two focusing mirrors arranged to focus the delayed portion in focus on the first delay path and returning the delayed part to the first beam splitter with a size and angular divergence of the beam equal to or approximately equal to the size and angular divergence of the laser output pulse beam in the first beam splitter. 152. The laser system of claim 146, wherein the first beam splitter is configured to direct the laser beam in at least two directions using the optical property of the truncated internal reflection. 153. The laser system of claim 146, wherein the first beam splitter consists of two transparent optical elements, each of which has a flat surface, both optical elements being arranged so that their first flat surfaces are parallel to each other and separated by a distance of less than 200 nm. 154. The laser system of claim 146, wherein the first beam splitter is an uncoated optical element oriented at an angle to the laser output pulse beam to achieve the desired reflection-transmission ratio.

Images