JP5068596B2 - Pulse shaping device, pulse shaping method, and electron gun - Google Patents

Description

  The present invention relates to a pulse shaping device that shapes pulsed light, a pulse shaping method, and an electron gun using the shaped pulsed light.

  An electron gun as an electron source such as an X-ray free electron laser (XFEL), a femtosecond X-ray pulse light source by inverse Compton scattering, a femtosecond time-resolved electron microscope, an ultrashort pulse electron beam drawing apparatus, an energy recovery linac (ERL) Is used. Usually, a high-intensity electron gun or an ultrashort pulse electron gun is used for these applications.

  The electron guns are classified into three types: a hot cathode electron gun, a photocathode electron gun, and a field emission electron gun. The hot cathode electron gun emits electrons by thermal energy. The photocathode electron gun emits electrons by the photoelectric effect. The field emission electron gun emits electrons by applying a high electric field (1 GV / m or more) to the cathode. Further, when the electron gun is classified from the electron acceleration method, it is divided into an RF electron gun and a DC electron gun. An RF electron gun accelerates electrons by microwaves (RF) with the help of a resonator cavity. A DC electron gun accelerates electrons in a DC manner (impulse manner) by applying a voltage between opposing electrodes. When the beam current is large, a combination of a photocathode and an RF acceleration method and a combination of a thermal cathode and a DC acceleration method are common. This is because the properties of both can be matched and the compatibility is good. Moreover, emittance, luminance, etc. are mentioned as important characteristics of an electron beam. Therefore, development of an electron gun that can stably generate a high-quality electron beam with low emittance and high brightness is desired.

  The photocathode RF electron gun has (1) low emittance compared to the hot cathode DC electron gun and field emission electron gun, and (2) an electron beam easily by controlling the intensity of the light source. Therefore, it has advantages such as excellent controllability. Currently, the world's brightest pulsed electron beam is obtained by a photocathode electron gun. In the photocathode electron gun, the photocathode is irradiated with laser light to generate electrons.

  In the case of using a photocathode electron gun, the electron bunch can be formed into a desired shape by shaping the pulse of the laser beam. In other words, the quality of the electron beam can be improved by shaping the pulse shape. For this reason, it is desired to shape the laser light pulse into a desired shape. For example, a technique for shaping laser light into a cylindrical shape using a pulse stacker is disclosed (Non-Patent Document 1). Here, the pulse stacker is composed of a wave plate and a polarizing beam splitter (Non-Patent Document 2). And it is branched into each polarization component (S polarization and P polarization) by the polarization beam splitter cube. Then, an optical path length difference is given to the branched pulse light, and it is combined by the combining polarization beam splitter. In the above document, since a stack of three stages is used, eight branched pulse lights are superimposed. In such a pulse stacker, for example, pulse shaping in the vertical direction (direction parallel to the optical axis) can be performed. That is, the pulse can be shaped in the vertical direction by adjusting the optical path length given to the branched pulse laser beam. Furthermore, in the processing method using pulsed laser light as shown in Patent Document 1, it is desired to shape the pulsed light into a desired shape in order to control the processing shape.

  When shaping the pulse, adjustments are made based on the actually measured pulse shape. When measuring the pulse shape of laser light, the pulse shape in the lateral direction (direction perpendicular to the optical axis) can be measured by, for example, a beam profiler or a CCD camera. Further, the pulse shape in the vertical direction can be measured by, for example, a streak camera.

Hiromitsu Serizawa “Aiming for a Laser Light Source That Can Meet Accelerator Requirements—Development of a Laser Light Source for Photocathode RF Electron Guns” Accelerators Vol3, No3, 2006 (251-262) Luminex website [Searched on June 19, 2007], Internet <http://www.luminex.co.jp/Newcatalog/Pulse_Stacker_Kit/Pulse%20Stacker%20Kit.pdf> JP 2002-205179 A

However, with the conventional pulse shaping method, it is difficult to shape three-dimensionally.
That is, in the beam profiler and the CCD camera, the pulse shape is measured by integrating over time, so that information in the vertical direction cannot be obtained. In addition, the streak camera cannot obtain lateral information. Therefore, there is a problem that it is difficult to shape the pulsed light into a desired shape.

  The present invention has been made in the background of such circumstances, and an object of the present invention is to provide a pulse shaping device, a pulse shaping method, and an electron gun using the same that can easily shape pulsed light. Is to provide.

  The pulse shaping device according to the first aspect of the present invention includes an optical branching unit that splits pulsed light into two light beams, a light combining unit that combines the two light beams branched by the optical branching unit, A delay unit that shifts the timing by giving an optical path length difference to the pulsed light of the two light beams incident on the light combining unit from the light branching unit; and an adjusting unit that adjusts the pulse shape of the pulsed light from the light combining unit; A light blocking member capable of blocking one light beam incident on the light combining unit between the first light branching unit and the light combining unit; and the light blocking unit for adjusting the pulse shape by the adjusting unit. A measuring device that performs measurement according to the other light beam in a state where the one light beam is shielded by a member. Thereby, since it can measure with a sufficient precision, a pulsed light can be shaped with a simple structure.

  A pulse shaping device according to a second aspect of the present invention is the above-described pulse shaping device, wherein a λ / 2 plate rotatably provided is disposed in front of the light branching unit, and the light combining unit and the light The branching means is a polarization beam splitter that branches light according to the polarization state. Thereby, the intensity of the branched light beam can be changed.

  A pulse shaping device according to a third aspect of the present invention is the above-described pulse shaping device, wherein the measuring device measures the pulse shape of the other light beam, and the other light measured by the measuring device. Based on the pulse shape of the beam, the adjusting means adjusts the pulse shape. Thereby, the pulse shape can be accurately shaped.

  A pulse shaping device according to a fourth aspect of the present invention is the above-described pulse shaping device, characterized in that the optical path length difference provided by the delay means is variable. Thereby, the pulse shape in the vertical direction can be easily adjusted.

  A pulse shaping device according to a fifth aspect of the present invention is the above-described pulse shaping device, wherein the light synthesis unit synthesizes the one light beam and the other light beam, so that the measurement light is transmitted from the light synthesis unit. And the utilization light are branched and emitted, and the pulse shape of the utilization light is adjusted based on the measurement result of the pulse shape of the measurement light. Thereby, it can prevent that the efficiency of the light actually utilized reduces.

  A pulse shaping device according to a sixth aspect of the present invention includes a birefringent element that gives a time delay between orthogonally polarized components of pulsed light, and an adjusting unit that adjusts the pulse shape of the pulsed light from the birefringent element; Means for taking out one of the orthogonal polarization components of the pulsed light emitted from the birefringent element from the other polarization component, and the one polarization component taken out by the means for adjusting the pulse shape by the adjustment means And a measuring device that performs measurement according to the above. Thereby, since it can measure more accurately, pulsed light can be shaped easily.

  A pulse shaping device according to a seventh aspect of the present invention is the pulse shaping device described above, wherein means for extracting one of the orthogonal polarization components of the pulsed light from the other polarization component is inserted in the optical path of the pulsed light. It is provided so as to be removable. Thereby, it can mask simply.

  A pulse shaping device according to an eighth aspect of the present invention combines a first stack unit that emits a first light beam by synthesizing two pieces of pulsed light that are shifted in timing, and two pulsed lights that are shifted in timing. A second stack unit that emits a second light beam, and a rear-stage side light combiner that combines the first and second light beams at a subsequent stage of the first and second stack units, And a profile changing unit that is disposed between the rear-stage light combining unit and the stack unit and changes a profile of the first light beam. Thereby, pulsed light can be shaped with a simple configuration.

  A pulse shaping device according to a ninth aspect of the present invention is the pulse shaping device described above, wherein each of the first and second stack units branches the main pulse light into two light beams, and An optical branching unit that generates two pulse lights, an optical combining unit that combines two optical beams branched by the optical branching unit, and a pulse of two light beams incident on the optical combining unit from the optical branching unit A delay unit that shifts timing by giving an optical path length difference to the light, and a light blocking member that can block one light beam incident on the light combining unit between the first light branching unit and the light combining unit; And adjusting the pulse shape based on a measurement result of a measuring device that performs measurement according to the other light beam in a state where the one light beam is shielded by the light shielding member. Thereby, since it can measure accurately, pulsed light can be shaped simply.

  A pulse shaping device according to a tenth aspect of the present invention is the above-described pulse shaping device, wherein the latter-stage light combining unit combines the first and second light beams, so that the latter-stage light combining unit The measurement light and the utilization light are branched and emitted, and the pulse shape of the utilization light is adjusted based on the measurement result of the pulse shape of the measurement light. It is possible to prevent the light efficiency actually used from being reduced.

  A pulse shaping device according to an eleventh aspect of the present invention is the pulse shaping device described above, wherein the means for generating probe light whose wavelength changes with time, the measurement light and the probe light are synchronized. And a spectroscope that measures the spectrum of light emitted from the non-linear optical element, and based on the measurement result of the spectroscope, the pulse shape in the traveling direction of the pulsed light is determined. To be adjusted. Thereby, since the pulse shape in the vertical direction can be accurately measured, accurate shaping becomes possible.

  A pulse shaping device according to a twelfth aspect of the present invention is the pulse shaping device described above, wherein the post-stage side light combining means is a non-polarizing beam splitter.

  An electron gun according to a thirteenth aspect of the present invention includes the pulse shaping device described above and a photocathode on which the pulsed light shaped by the pulse shaping device is incident. Thereby, a high-quality electron beam can be obtained.

  An electron gun according to a fourteenth aspect of the present invention is the electron gun described above, wherein a bunch shape of an electron beam generated at the photocathode is measured, and the pulse shape is determined based on the measurement result of the bunch shape. To be adjusted. Thereby, a suitable bunch-shaped electron beam can be obtained.

  An electron gun according to a fifteenth aspect of the present invention is the above electron gun, wherein the electrons in the bunch of the electron beam are spatially dispersed according to energy, and then the spatial distribution of the electrons in the bunch is measured. The pulse shape in the traveling direction of the pulsed light is adjusted based on the measurement result of the spatial distribution. Thereby, since the vertical bunch shape can be measured accurately, it can shape more correctly.

  A pulse shaping method according to a sixteenth aspect of the present invention includes an optical branching unit that splits pulsed light into two light beams, a light combining unit that combines the two light beams branched by the optical branching unit, A pulse shaping method using a pulse shaping device comprising: delay means that shifts the timing by giving an optical path length difference to pulse light of two light beams incident on the light combining means from the light branching means, A step of shielding one of the two light beams branched by the light branching means, a step of performing a measurement according to the other light beam in a state where the one light beam is shielded, Adjusting a pulse shape based on a measurement result in the step of performing the measurement. Thereby, since it can measure accurately, pulsed light can be shaped simply.

  A pulse shaping method according to a seventeenth aspect of the present invention is the above-described pulse shaping method, comprising: shielding the other of the two light beams branched by the light branching unit; A step of performing measurement according to one light beam in a state where the other light beam is shielded, and the step of adjusting the pulse shape includes the step of performing measurement according to the one light beam. The pulse shape is adjusted based on the measurement result. Thereby, since it can measure more accurately, pulsed light can be shaped easily. Moreover, since the space shaping technique and the time shaping technique do not affect each other, they can be controlled independently.

  A pulse shaping method according to a seventeenth aspect of the present invention includes a birefringent element that gives a time delay between orthogonally polarized components of pulsed light, and adjustment that adjusts the pulse shape of the pulsed light from the birefringent element. A step of extracting one of the orthogonally polarized components of the pulsed light emitted from the birefringent element from the other polarized component, and adjusting the pulse shape by the adjusting means And a step of performing a measurement according to the one extracted polarization component and a step of adjusting a pulse shape based on a measurement result in the step of performing the measurement. Thereby, since it can measure more accurately, pulsed light can be shaped easily.

  According to a nineteenth aspect of the present invention, there is provided a pulse shaping method comprising: combining two pulse lights having different timings to emit a first light beam; and combining two pulse lights having different timings. Combining the step of emitting the second light beam, the step of changing the profile of the first light beam, the second light beam, and the first light beam having the changed profile. , And a step of emitting. Thereby, pulse light can be shaped easily.

  A pulse shaping method according to a twentieth aspect of the present invention is the pulse shaping method described above, wherein one of the two pulse lights included in the first or the second light beam is shielded by a light shielding member, The pulse shape is adjusted based on the measurement result of the pulse shape in a state where the light is shielded by the light shielding member. Thereby, since it can measure more accurately, pulsed light can be shaped easily.

  ADVANTAGE OF THE INVENTION According to this invention, the pulse shaping apparatus which can shape pulsed light simply, the pulse shaping method, and an electron gun using the same can be provided.

  Hereinafter, embodiments to which the present invention can be applied will be described. The following description is to describe the embodiment of the present invention, and the present invention is not limited to the following embodiment. For clarity of explanation, the following description is omitted and simplified as appropriate. Further, those skilled in the art will be able to easily change, add, and convert each element of the following embodiments within the scope of the present invention. In addition, what attached | subjected the same code | symbol in each figure has shown the same element, and abbreviate | omits description suitably.

  An electron gun according to an embodiment of the present invention will be described with reference to FIG. FIG. 1 is a diagram schematically illustrating a configuration of an electron gun 100 according to the first embodiment. The electron gun 100 according to the present embodiment is a photocathode electron gun that generates electrons when laser light is incident on the cathode. The electrons generated by the electron gun 100 are accelerated by the microwave from the microwave source 53. The electron gun 100 according to the present embodiment has a transmissive photocathode. That is, the laser beam is irradiated from the side opposite to the electron beam emission side. Furthermore, the electron gun 100 has a pulse shaping device for shaping the pulsed laser light into a desired shape.

  The beam measuring apparatus according to the present embodiment includes a light source unit 10, a pulse stacker 20, a deformable mirror 40 (DM: Deformable Mirror), a mirror 41, a lens 42, a lens 43, a lens 44, a mirror 45, a beam splitter 46, and a camera. 47, a control device 48, a photocathode 51, a resonator 52, a microwave source 53, a beam monitor 55, a wedge 56, a mirror 57, and a mirror 58.

  The light source unit 10 generates pulsed light that enters the photocathode. In the light source unit 10, the pulse width of the pulse laser beam is expanded and the pulse laser beam is chirped. The pulsed light emitted from the light source unit 10 enters the pulse stacker 20 through the wedge 56. That is, four transparent wedges 56 are provided between the light source unit 10 and the pulse stacker 20. The control device 48 drives each wedge 56 to adjust the position of the pulsed light incident on the pulse stacker 20. For example, the position of the pulsed light can be adjusted by rotating each wedge 56 independently. Pointing compensation can be performed by using such a wedge 56.

  The pulse stacker 20 divides the pulsed light to generate a plurality of micro pulses. Furthermore, the pulse stacker 20 synthesizes a plurality of micro pulses while shifting them in time. Thereby, the pulsed light is shaped so that the pulse width becomes long. That is, the pulse shape in the vertical direction (direction perpendicular to the optical axis) becomes wider. The configurations of the light source unit 10 and the pulse stacker 20 will be described later.

  The pulsed light from the pulse stacker 20 is incident on the deformable mirror 40 via the two mirrors 57. That is, two mirrors 57 are provided between the pulse stacker 20 and the deformable mirror 40. Further, a lens may be provided between the pulse stacker 20 and the deformable mirror 40 to collimate the pulsed light. Thereby, the beam diameter incident on the deformable mirror 40 can be adjusted, and the beam diameter can be made equal to or smaller than the effective diameter of the deformable mirror 40. The control device 48 drives each mirror 57 to adjust the position of the pulsed light incident on the deformable mirror 40. For example, the position of the pulsed light can be adjusted by tilting each mirror 57 independently. As described above, the position of the pulsed light can be changed by the control device 48 controlling the angle of the mirror 57. Pointing compensation can be performed by using such a mirror 57.

  By performing the pointing compensation using the mirror 57 and the wedge 56, it is possible to enter the deformable mirror 40 with high accuracy. That is, the positional deviation of the pulse laser beam in the deformable mirror 40 can be prevented. Thereby, high-precision shaping becomes possible. In addition, pointing compensation can be performed using the takumi algorithm of Japanese Patent Application No. 2004-30236. Furthermore, an optimization algorithm such as a genetic algorithm or an annealing method may be used.

  The deformable mirror 40 shapes the shape of the pulsed light in the horizontal direction (direction perpendicular to the optical axis). That is, the deformable mirror 40 shapes the pulse shape in a plane perpendicular to the optical axis. Specifically, the deformable mirror 40 adjusts the wavefront shape of the laser light by changing the position of the reflecting surface. For this reason, the deformable mirror 40 is provided with a plurality of electrodes (channels) for changing the shape of the mirror surface (reflection surface), for example. And the shape of a wave front changes by adjusting the voltage applied to each electrode. By controlling the surface shape of the mirror, it is possible to correct the wavefront of the optical system and to control the direction and shape of the reflected beam. As the deformable mirror 40, for example, an electrostatic capacity type, a piezoelectric actuator type, or the like can be used. The deformable mirror 40 has an effective diameter of, for example, φ20 mm to φ50 mm.

  Pulse light from the deformable mirror 40 propagates through a propagation optical system including the mirror 41, the lens 42, the lens 43, the lens 44, and the mirror 45. That is, the pulsed light reflected by the deformable mirror 40 is reflected by the mirror 41. Then, the light is refracted by the lens 42, the lens 43, and the lens 44 and enters the mirror 45. The pulsed light incident on the mirror 45 is reflected in the direction of the beam splitter 46. In this way, the pulsed light propagates in the order of the mirror 41, the lens 42, the lens 43, the lens 44, and the mirror 45. For example, the lens 42 is a convex lens with f = 51.5 mm, the lens 43 is a concave lens with f = 100 mm, and the lens 44 is a convex lens with f = 300. The distance between the lens 42 and the lens 43 is 22 mm, and the distance between the lens 43 and the lens 44 is 108 mm. The lateral shape of the pulsed light is deformed by the lens 42, the lens 43, the lens 44, and the like. That is, the lateral shape of the pulsed light is deformed by utilizing the aberration of the transmission optical system such as a lens or window material.

  The pulsed light reflected by the mirror 45 enters the beam splitter 46. The beam splitter 46 branches the incident pulse light into two light beams. The beam splitter 46 reflects a part of the incident pulsed light in the direction of the camera 47. The camera 47 is, for example, a two-dimensional camera such as a CCD camera in which light receiving elements are arranged in an array. The camera 47 measures the profile of the pulsed light. Specifically, the camera 47 measures the pulse shape in a plane perpendicular to the optical axis. Since the time integration type camera 47 is used, the two-dimensional shape of the entire one pulse is measured. That is, one frame of the camera 47 is sufficiently longer than the pulse length. The camera 47 measures the lateral profile of the pulsed light. The camera 47 then outputs a signal based on the measurement result corresponding to the pulse shape in the horizontal direction to the control device 48. As described above, the camera 47 is a measuring device that performs measurement according to the light beam in order to adjust the pulse shape by the deformable mirror 40. Therefore, when the light beam changes, the signal output from the camera 47 changes.

  Further, part of the pulsed light incident on the beam splitter 46 passes through the beam splitter 46. Then, the pulsed light that has passed through the beam splitter 46 enters the photocathode 51. The photocathode 51 is a transmissive photocathode. The photocathode 51 is not limited to the transmissive type, and may be a reflective type photocathode. The photocathode 51 generates electrons according to the incident pulsed light. That is, electrons are generated from the photocathode 51 by the photoelectric effect. Electrons generated from the photocathode 51 enter the resonator 52. The microwave generated by the microwave source 53 is incident on the resonator 52. The resonator 52 is a cavity resonator, and generates a standing wave corresponding to the input microwave. That is, an electric field for accelerating electrons generated at the photocathode 51 is generated in the resonator 52 which is an RF resonator. The accelerating electric field in the resonator 52 changes with time. Electrons generated at the photocathode 51 are accelerated by an electric field in the resonator 52. That is, it is emitted from the resonator 52 as an electron beam having a predetermined energy. Here, the timing of the pulsed light is adjusted in accordance with the standing wave generated in the resonator 52. That is, the microwave from the microwave source 53 and the pulse of the laser beam are synchronized. Thereby, electrons are generated from the photocathode 51 at the timing when an accelerating electric field is generated in the resonator 52. Accordingly, the electron beam is accelerated efficiently.

  In this way, an electron beam bunch is emitted. The electron beam passes through a predetermined path, and is an X-ray free electron laser (XFEL), a femtosecond X-ray pulse light source by inverse Compton scattering, a femtosecond time-resolved electron microscope, an ultrashort pulse electron beam drawing apparatus, an energy recovery type Used for linac (ERL). Further, the electron beam enters the beam monitor 55. The beam monitor 55 monitors the electron beam. Information regarding the bunch can be measured by the beam monitor 55. Specifically, an electron beam profile, emittance, energy, or the like is measured. A signal corresponding to the monitoring result of the beam monitor 55 is input to the control device 48.

  The profile of the generated electron beam reflects the shape of the pulsed light incident on the photocathode 51. That is. An electron beam having a profile corresponding to the shape of the pulsed light is generated. Specifically, the lateral shape of the bunch reflects the pulse shape in a plane perpendicular to the optical axis. The vertical shape of the bunch reflects the pulse shape in a plane parallel to the optical axis. That is, the electron beam has a bunch length corresponding to the pulse width of the laser light. Thus, the profile of the electron beam changes according to the profile of the laser beam. Therefore, the beam monitor 55 becomes a measuring device that performs measurement according to the light beam in order to adjust the pulse shape by the deformable mirror 40. In other words, when the pulse shape of the light beam changes, the signal output from the beam monitor 55 also changes.

  The control device 48 is a computer having a storage area such as a CPU or a memory. For example, the control device 48 has a CPU (Central Processing Unit) that is an arithmetic processing unit, a ROM (Read Only Memory) or RAM (Random Access Memory) that is a storage area, a communication interface, and the like, and measures an asserted shape. Execute the processing necessary to For example, the ROM stores an arithmetic processing program for performing arithmetic processing, various setting data, and the like. Then, the CPU reads out the arithmetic processing program stored in the ROM and develops it in the RAM. Then, the program is executed according to the setting data and the output from the beam monitor 55 or the camera 47. Furthermore, the control device 48 has a monitor for displaying the calculation processing result. The control device 48 is an information processing device such as a personal computer, for example, and performs predetermined arithmetic processing on the measurement result and the monitor result. Further, the control device 48 performs control for adjusting the profile of the laser beam. The control device 48 is not limited to a physical unit configuration.

For example, the control device 48 controls the deformable mirror 40 based on the measurement result of the electron beam and the measurement result of the pulse shape. The control device 48 outputs a control signal to the deformable mirror 40 to change the shape of the reflecting surface. Thereby, the pulse shape of the laser beam is adjusted. Therefore, it is possible to obtain an optimum bunch shape for using an electron beam. For example, a low emittance electron beam can be obtained.
Furthermore, by performing feedback control based on the measurement result, the bunch shape can be adjusted in real time. Thus, the control device 48 controls the deformable mirror 40 to adjust the shape of the pulsed light. As a result, the profile of the pulsed light incident on the photocathode 51 changes. Then, the pulse shape is optimized based on the monitoring result of the electron beam and the measurement result of the pulse shape. Thereby, quality improvement of an electron beam can be achieved. Furthermore, a bunch-shaped electron beam suitable for the purpose of use can be generated.

  The distance from the beam splitter 46 to the photocathode 51 is equal to the distance from the beam splitter 46 to the camera 47. Therefore, the pulse shape on the photocathode 51 and the pulse shape on the camera 47 are substantially equal. Therefore, even when the control is performed based on the detection result of the camera 47, the electron beam bunch can be accurately shaped.

  Next, the configuration of the light source unit 10 will be described with reference to FIG. FIG. 2 is a diagram illustrating a configuration example of the light source unit 10. The light source unit 10 includes a laser oscillator 11, a pulse stretcher 12, an optical modulator 13, a regenerative amplifier 14, a multipath amplifier 15, a pulse compressor 16, and a wavelength converter 17.

The laser oscillator 11 is a mode-locked titanium sapphire laser oscillator, for example, and emits pulsed laser light. The laser oscillator 11 uses titanium sapphire as a crystal medium. Sapphire, which is the crystal base material, is a transparent material having good heat conductivity next to diamond, and is an excellent material that becomes heat conduction equivalent to movement when cooled to a nitrogen temperature. Therefore, there is little heat accumulation. A titanium sapphire crystal in which Ti ³⁺ is doped into sapphire as a base material has a broad fluorescence spectrum bandwidth of 400 nm (700 to 1100 nm), and is suitable for ultrashort (Fourier transform limit) pulse generation. In the first place, a femtosecond ultrashort pulse is a Fourier transform limit laser pulse that aligns the relative phase between spectral components of a broadband laser pulse exceeding 10 nm. By controlling the difference in the optical path length of each wavelength of the broad spectrum with a diffraction grating pair of a pulse stretcher 12 and a pulse compressor 16 described later, the pulse can be extended or shortened. The laser oscillator 11 oscillates a pulse laser beam having a repetition frequency of 89.25 MHz, a center wavelength of 790 nm, and a pulse width (half-value width) of 20 fsec. The repetition frequency of the laser oscillator 11 is the S-band frequency division of 32, which is widely used for electron acceleration in the electron accelerator.

  The pulse laser beam generated by the laser oscillator 11 enters the pulse stretcher 12. The pulse stretcher 12 has a diffraction grating pair for extending the pulse width. For example, the pulse stretcher 12 widens the pulse width (half width) to 150 psec. The pulsed laser light whose pulse width is widened by the pulse stretcher 12 is incident on the optical modulator 13.

  As the optical modulator 13, for example, an acousto-optic (AO) element or a liquid crystal element can be used. More specifically, an acousto-optic spatial phase control filter (AOPDF) and a liquid crystal spatial light modulator (SLM: Spatial Light Modulator) can be used as the light modulator 13. The optical modulator 13 modulates the spectral phase. As the optical modulator 13, for example, DAZZLER (registered trademark) manufactured by FASTLITE can be used. As a result, it is possible to modulate pulsed laser light having a broad spectrum. The optical modulator 13 is controlled based on the above-described electron beam monitoring result and pulse shape measurement result. Thereby, the pulse waveform can be shaped. That is, a pulse waveform suitable for improving the quality of the electron beam can be obtained.

  Further, the pulsed laser beam may be chirped by the optical modulator 13. That is, the optical modulator 13 may not only shape the pulse but also chirp it. The spectral phase or intensity of the pulsed laser light is modulated so that the wavelength changes with time. By positively chirping the pulse laser beam, the wavelength gradually decreases from the beginning of the pulse toward the rear side. The wavelength at the front end of the pulse is the longest wavelength, and the wavelength at the rear end of the pulse is the shortest wavelength. The wavelength in the laser pulse monotonously decreases toward the rear side. Furthermore, higher order dispersion is compensated so that the relationship between wavelength and time is linear, and the increase or decrease of the second order dispersion is controlled. For example, the secondary dispersion of the optical modulator 13 itself can be canceled by the secondary dispersion at the set value of the optical modulator 13, and if it is set to the Fourier limit at the cancellation, the positive chirp can be obtained. It can also be a negative chirp. Therefore, the wavelength decreases linearly with time. Thus, in the laser pulse light emitted from the optical modulator 13, the wavelength is changed linearly with time. That is, the wavelength of the pulsed laser light changes linearly with time. The spectrum of the pulse laser beam is modulated in accordance with a modulation signal (Acoustic Wave) input to the optical modulator 13. And the optical path difference according to the wavelength can be freely given to the pulsed laser light passing through the crystal by the chirped sound wave. The spectrum of the pulse laser beam is shaped into the same shape as the spectrum of the sound wave. Alternatively, the spectrum of the pulse laser beam may be shaped into the same shape as the pulse shape of the sound wave. Further, it is possible to perform chirping by an optical element other than the optical modulator 13. For example, pulse light can be positively chirped by using a transparent positive dispersion medium having a different refractive index depending on the wavelength.

  The pulsed laser light modulated by the optical modulator 13 is amplified by the regenerative amplifier 14 and the multipath amplifier 15. That is, the pulse laser beam is incident on the regenerative amplifier 14 at the previous stage. The regenerative amplifier 14 amplifies the pulse laser beam and outputs the amplified laser beam to the subsequent multipath amplifier 15. Then, the pulsed laser light amplified by the multipass amplifier 15 enters the pulse compressor 16. The regenerative amplifier 14 and the multi-pass amplifier 15 branch off the flash lamp pumped YAG laser light and use it as a pumping light source. That is, the regenerative amplifier 14 and the multipath amplifier 15 share a common excitation light source.

In the regenerative amplifier 14, a pulse train is cut out at 10 Hz by a Pockels cell in the resonator and selectively amplified. Here, for example, the pulse laser beam is amplified up to 2 mJ. This laser pulse is reciprocated four times through the same crystal by the next multi-pass amplifier 15. For example, the multipath amplifier 15 amplifies up to 30 mJ / pulse. In the regenerative amplifier 14 and the multipath amplifier 15, the center wavelength and the pulse width (half-value width) remain at 790 nm and 150 psec. In this way, the laser oscillator 11 for producing seed light and the MOPA (Master Oscillator Power Amplifier) system including the external light after the seed light is employed. The external amplifier has a two-stage configuration of a regenerative amplifier 14 and a multipath amplifier 15.

  The pulsed laser light from the multipass amplifier 15 enters the pulse compressor 16. The pulse compressor 16 has a diffraction grating pair for shortening the pulse length. Here, for example, the pulse length (half width) is reduced from 150 psec to 1 psec. Further, the shortest pulse in the spectral band of the laser, that is, a Fourier limit pulse can be obtained. Thus, the regenerative amplifier 14 and the multipath amplifier are arranged between the pulse stretcher 12 and the pulse compressor 16. Thereby, chirped pulse amplification (chirped pulse amplification) which amplifies while suppressing peak power can be performed. From the pulse compressor 16, for example, a pulse laser beam having a center wavelength of 790 nm and a spectrum bandwidth (half-value width) of 20 nm is emitted.

  The pulsed laser light from the pulse compressor 16 enters the wavelength converter 17. Here, the wavelength converter 17 generates the third harmonic (center wavelength 263 nm) of the pulse laser beam having the center wavelength 790 nm emitted from the pulse compressor 16. Specifically, the wavelength converter 17 has two nonlinear optical crystals. Here, a BBO crystal can be used as the nonlinear optical crystal. First, a fundamental wave (center wavelength 790 nm) from the pulse compressor 16 is incident on the first crystal to generate a second harmonic (center wavelength 395 nm). Then, a second harmonic and a fundamental wave are incident on the second crystal to generate a third harmonic.

The basis of the short wavelength generation of the solid-state laser is the result of the interaction in the BBO crystal using the second-order nonlinear optical effect for the two incident coherent lights (wavelengths λ ₁ and λ ₂ ). It is the generation of the wavelength λ ₃ in accordance with an energy conservation law, that is, sum frequency mixing (SFM). Second harmonic generation (SHG) corresponds to the case where the wavelengths λ ₁ and λ _{2 of the} two incident lights of the SFM are equal. Third harmonic generation (THG) is realized by SFM of the second harmonic (λ ₂ ) generated by normal SHG and the remaining fundamental wave (λ ₁ ). The conversion efficiency from the fundamental wave to the 3rd harmonic is about 10%, and the conversion efficiency from the 2nd harmonic is about 50%. Of course, the third harmonic may be generated by a nonlinear optical element other than the BBO crystal.

As described above, the light source unit 10 is subjected to pulse width expansion, amplification, compression, and wavelength conversion to ultraviolet light. Note that the pulse width may be extended by passing a quartz rod after the wavelength converter and positively chirping the ultraviolet pulse. For example, by using two fused quartz rods having a length of 45 cm, the pulse width can be increased from 80 fsec to 10 psec. (H. Tomizawa, T. Asaka, H. Dewa, H. Hanaki, T. Kobayashi, A. Mizuno, S. Suzuki, T. Taniuchi, and K. Yanagis. Sourcesc, Proceedings of the 27th International Free Electron Laser Conference, Stanford, CA, 21-26 August 2005, (2005) pp. 138-141) The light source unit 10 emits linearly polarized light. The pulsed laser light emitted from the light source unit 10 enters the pulse stacker 20 as shown in FIG.
However, the THG of an ultra-short pulse laser after amplification so that the nonlinear effect is effective can produce a refractive index distribution in the transverse direction called filamentation in quartz. This may cause a problem that the profile is deteriorated. In such a case, the quartz rod need not be used.

  Next, the configuration of the pulse stacker 20 will be described with reference to FIGS. 3 and 4. FIG. 3 is a diagram showing the configuration of the optical system of the pulse stacker 20. FIG. 4 is a timing chart showing the pulse waveforms stacked by the pulse stacker 20. In FIG. 4, it is shown that pulsed light having a pulse width of 2.5 psec is incident from the light source unit 10. In FIG. 4, the horizontal axis indicates time, and the vertical axis indicates light intensity. In FIG. 4, six pulse waveforms in the pulse stacker 20 are shown in order from the top. The six pulse waveforms are shown as A to F, respectively.

  The pulse stacker 20 includes an incident side λ / 4 plate (¼ wavelength plate) 21, an incident side λ / 2 plate (1/2 wavelength plate) 22, an incident side polarization beam splitter (hereinafter, the polarization beam splitter is referred to as PBS). ) 23, stack unit 31, stack unit 32, stack unit 33, mirror 34, and emission side λ / 2 plate 35. That is, three stack units 31 to 33 are arranged in series. Here, the stack unit 31, the stack unit 32, and the stack unit 33 have the same configuration. Each of the stack unit 31, the stack unit 32, and the stack unit 33 includes a stacking λ / 2 plate 24, a fixed mirror pair 25, a movable mirror pair 26, an actuator 27, a branching PBS 28, a combining PBS 29, and a light shielding member 37. Have. The pulse stacker 20 adjusts the power with the incident side λ / 2 plate 22 and the incident side PBS 23.

  The incident side PBS 23, the branching PBS 28, and the combining PBS 29 are, for example, polarization beam splitter cubes, and branch light according to the polarization state. For example, the incident side PBS 23, the branching PBS 28, and the combining PBS 29 pass P-polarized light and reflect S-polarized light. The incident-side λ / 4 plate 21, the incident-side λ / 2 plate 22, and the stacking λ / 2 plate 24 are zero-order quartz wavelength plates, and give a predetermined phase difference between the ordinary light component and the extraordinary light component. The polarizing beam splitter cube is preferably an optical contact type. This makes it possible to apply even to high-intensity pulsed light. When the laser intensity is high, the air gap method is preferable. In addition, there is a wave plate with optical contact, but there is less emission when using an adhesive-free one that does not use an adhesive. Further, when an adhesive is used, an aerospace adhesive is preferable.

  Pulse light from the light source unit 10 is incident on the pulse stacker 20 shown in FIG. The incident side λ / 4 plate 21 converts the elliptical pulsed light into linearly polarized light. For example, the incident side λ / 4 plate 21 is rotatably provided. Then, the rotation angle is adjusted according to the polarization state of the incident pulsed light. Thereby, the pulsed light becomes a substantially perfect straight line. Of course, when the pulsed light emitted from the light source unit 10 is linearly polarized light, the incident side λ / 4 is unnecessary. The pulsed light that has passed through the λ / 4 plate 21 enters the λ / 2 plate 22.

  The incident side λ / 2 plate 22 rotates the polarization axis of the incident linearly polarized light. The pulsed light that has passed through the incident side λ / 2 plate 22 enters the incident side PBS 23. The incident side PBS 23 allows incident light to pass according to the polarization state of incident light. The pulsed light that has passed through the incident side PBS 23 enters the stack unit 31. For example, the incident side PBS 23 transmits P-polarized light and reflects S-polarized light. Further, the incident side λ / 2 plate 22 is rotatably provided in front of the incident side PBS 23. That is, the optical axis of the incident side λ / 2 plate 22 rotates in a plane perpendicular to the optical axis. By adjusting the rotation angle of the incident side λ / 2 plate 22, the polarization plane is rotated. Thereby, the power of the pulsed light incident on the stack unit 31 can be adjusted. That is, since the polarization direction of the linearly polarized light changes depending on the rotation angle of the incident side λ / 2 plate 22, the P-polarized light component can be increased or decreased.

  The pulsed light that has passed through the incident side PBS 23 enters the first stack unit 31. The stack unit 31 branches the incident main pulse light and splits it into two light beams. Thereby, two micro pulses are generated. The two branched light beams propagate along different optical paths. The stack unit 31 gives an optical path length difference to the pulsed light of the two branched light beams, and then combines the two light beams. The two combined light beams propagate along a common optical path. As a result, two micro pulses with different timings are emitted from the stack unit 31. The configuration of the stack unit 31 having the above function will be described in detail below.

  The pulsed light incident on the stack unit 31 is incident on the stacking λ / 2 plate 24. The stacking λ / 2 plate 24 rotates the polarization plane of linearly polarized light by 45 °. Here, it becomes + 45 ° linearly polarized light as shown in FIG. As a result, the P-polarized component and the S-polarized component become substantially the same. The pulsed light emitted from the stacking λ / 2 plate 24 enters the branching PBS 28.

  The branching PBS 28 branches the incident pulsed light according to the polarization state. Here, the P-polarized component passes through the branching PBS 28, and the S-polarized component is reflected by the front-side PBS. Since the P-polarized component and the S-polarized component are substantially the same by the stacking λ / 2 plate 24, the pulsed light is divided into two equal parts. Thus, the branching PBS 28 splits the pulsed light to generate two micro pulses. Here, the P-polarized component micropulse that has passed through the branching PBS 28 is referred to as a P-polarized micropulse, and the S-polarized component micropulse reflected by the branching PBS 28 is referred to as an S-polarized micropulse. S-polarized micropulses and P-polarized micropulses propagate along different optical paths.

  The S-polarized micropulse reflected by the branching PBS 28 enters the fixed mirror pair 25. The fixed mirror pair 25 is fixed to, for example, a stage. The fixed mirror pair 25 has two mirrors 30. The mirror 30 is a plane mirror, for example, and reflects most of the incident light. Then, the S-polarized micropulse is reflected by the two mirrors 30 and enters the combining PBS 29.

  On the other hand, the P-polarized micropulse that has passed through the branching PBS 28 enters the movable mirror pair 26. Similar to the fixed mirror pair 25, the movable mirror pair 26 has two mirrors 30. Then, the P-polarized micropulse is reflected by the two mirrors 30 and is incident on the combining PBS 29. Further, the movable mirror pair 26 is provided so as to be movable with respect to the stage or the like. That is, the movable mirror pair 26 is attached to the actuator 27. Then, by driving the actuator 27, the movable mirror pair 26 moves in the direction of the dotted line arrow in FIG. The actuator 27 is controlled by the control device 48.

  As a result, the distance from the movable mirror pair 26 to the combining PBS 29 changes, and the distance from the movable mirror pair 26 to the branching PBS 28 changes. Therefore, between the branching PBS 28 and the combining PBS 29, the optical path length changes between the P-polarized micropulse and the S-polarized micropulse. The P-polarized micropulse is delayed with respect to the S-polarized micropulse. Here, in the pulse stacker 20, if a micro pulse of 2.5 psec is 2.5 psec and a macro pulse of 20 psec is produced, 10 psec in the first stage, 5 psec in the second stage, and 2,5 psec in the third stage. Give a delay. The P-polarized micropulse is delayed by 10 psec from the S-polarized micropulse. Although the P-polarized micropulse may be advanced even if it is delayed, it is set in the same direction at all stages. That is, the distance from the branching PBS 28 to the synthesizing PBS 29 is adjusted by the actuator 27 to set the time delay amount to 10 psec. As a result, a desired amount of time delay can be provided between the S-polarized micropulse and the P-polarized micropulse. As described above, the movable mirror pair 26 and the fixed mirror pair 25 serve as delay means for providing a time delay between the two light beams.

  The P-polarized micropulse and the S-polarized micropulse to which the optical path length difference is given are synthesized by the synthesis PBS 29. That is, the combining PBS 29 passes the P-polarized micropulse and reflects the S-polarized micropulse. Thereby, the P-polarized micropulse and the S-polarized micropulse are synthesized. That is, the optical axis of the P-polarized micropulse and the optical axis of the S-polarized micropulse merge and coincide. At this time, a time delay is given to the P-polarized micropulse and the S-polarized micropulse as shown in B of FIG.

  Then, the pulsed light emitted from the first stack unit 31 enters the second stack unit 32. The second stack unit 32 has the same configuration as the first stack unit 31. Here, the P-polarized micropulse and the S-polarized micropulse are incident on the second stack unit 32. For this reason, as shown in FIG. 4C, the stacking λ / 2 plate 24 causes the polarization axis to be + 45 ° for the S-polarized micropulse and −45 ° for the P-polarized micropulse.

  Then, the pulse light including two micro pulses is branched by the branching PBS 28 and synthesized by the synthesis PBS 29. At this time, the polarization axis is rotated by the stacking λ / 2 plate 24. Therefore, each micro pulse is further branched into two micro pulses. That is, four micro pulses are generated. Further, the optical path length difference is given by the fixed mirror pair 25 and the movable mirror pair 26. Here, for example, the actuator 27 is adjusted so that a time delay of 5 psec is given. That is, the P-polarized micropulse is delayed by 5 psec from the S-polarized micropulse. For this reason, the pulse light synthesized by the synthesizing PBS 29 includes four micro pulses as shown in FIG. 4D. The two micropulses on the leading side are generated from micropulses having a polarization axis of + 45 ° immediately after the stacking λ / 2 plate 24. Further, the two micropulses on the rear end side are generated from micropulses having a polarization axis of −45 ° immediately after the stacking λ / 2 plate 24.

  In the second stack unit 32, a delay time shorter than the delay time in the first stack unit 31 is given. For this reason, in the synthesized light beam, P-polarized micropulses and S-polarized micropulses are alternately arranged. Further, in the second stack unit 32, a delay time almost half that of the first stack unit 31 is given. For this reason, the time interval between adjacent micropulses is equal. That is, four micro pulses are arranged at intervals of 5 psec.

  The pulsed light having four micro pulses emitted from the second-stage stack unit 32 enters the third-stage stack unit 33. Therefore, as shown in FIG. 4E, the stacking λ / 2 plate 24 causes the polarization axis to be + 45 ° for the S-polarized micropulse and −45 ° for the P-polarized micropulse. Here, micropulses having a polarization axis of + 45 ° and micropulses having a polarization axis of −45 ° are alternately arranged.

  Then, similarly to the second stack unit 32, the number of micro pulses is doubled. As a result, eight micro pulses are generated. Here, the delay time by the actuator 27 is set to 2.5 psec. As a result, the micro pulses are arranged as shown in FIG.

  In the third stack unit 33, a delay time shorter than the delay time in the second stack unit 32 is given. For this reason, in the synthesized light beam, P-polarized micropulses and S-polarized micropulses are alternately arranged. Further, the third stack unit 33 is given a delay time almost half that of the first stack unit 32. For this reason, the time intervals between adjacent micropulses are equal. That is, eight micro pulses are arranged at intervals of 2.5 psec.

  In this way, the three stack units 31 to 33 are arranged in series, and eight micro pulses are superimposed. Thereby, the pulse length of the macro pulse is about 20 psec. By controlling the actuator 27 provided in each of the stack units 31 to 33, the pulse shape in the vertical direction can be adjusted. In addition, the pulse length can be adjusted by controlling the actuator 27. Of course, the micropulses need not be equally spaced. The superimposed macro pulse is adjusted to a shape suitable for improving the quality of the electron beam.

  The macro pulse emitted from the third stack unit 33 enters the emission side λ / 2 plate 35 via the mirror 34. Then, a macro pulse from the emission side λ / 2 plate 35 is emitted from the pulse stacker 20. By rotating the emission side λ / 2 plate 35, the power of the passing micro pulse is adjusted. Here, the rotation angle of the emission side λ / 2 plate 35 is adjusted so that the power of light incident on the photocathode 51 is maximized.

  In the stack units 31 to 33, the optical element that gives the time delay is not limited to the movable mirror pair 26. For example, transparent members having different refractive indexes may be disposed in the optical path. Thereby, the optical distance changes between the P-polarized micropulse and the S-polarized micropulse, so that a time delay can be given. Of course, anything other than these may be used as the delay means.

  The stacking λ / 2 plate 24 may be rotatably provided. Thereby, the ratio of the P-polarized component and the S-polarized component branched by the branching PBS 28 can be changed. Therefore, the intensity of the eight micro pulses can be adjusted. The vertical shape of the pulsed light incident on the photocathode 51 can be adjusted.

Of course, the number of micropulses is not limited to eight. By increasing the number of stages of the stacked unit, it is possible to stack two ³ or more micro-pulse. That is, 2 ⁿ micropulses (n = number of stages) can be stacked. Of course, the number of stack units may be reduced so that the number of micropulses is less than eight.

  Further, each stack unit 31 to 33 is provided with a light shielding member 37. For example, the stack unit 31 is provided with one light shielding member 37. The light blocking member 37 is a shutter that selectively blocks the light beam branched by the branching PBS 28 from entering the combining PBS 29. A separate light blocking member 37 may be provided for each optical path. The light shielding member 37 is movably provided in the two optical paths branched by the branching PBS 28. That is, the light shielding member 37 is provided in the optical path so that it can be inserted and removed. When the electron beam is precisely adjusted, one light shielding member 37 is inserted into the optical path. In each of the stack units 31 to 33, the light shielding member 37 shields the P-polarized micropulse or the S-polarized micropulse. That is, the light shielding member 37 is disposed in one of the two light beams branched by the branching PBS 28. Accordingly, only one micro pulse is shielded from light in the stack units 31 to 33 of each stage.

In this case, only one micropulse out of the eight micropulses shown in FIG. 4F is emitted from the stack unit 33 at the third stage. That is, when light is shielded by the three light shielding members 37, only one micro pulse enters the photocathode 51 and the camera 47. Thereby, the shape measurement of each micro pulse becomes possible. For example, the horizontal shape of each micro pulse is measured by the camera 47. The deformable mirror 40 is set based on the individual measurement results of all the micro pulses. Specifically, one micropulse is shielded from light and the other micropulse is measured. Further, the other micropulse is shielded and one micropulse is measured. Then, the light blocking member 37 is sequentially moved with respect to each of the three stack units 31 to 33. Specifically, in the arrangement of the light shielding member 37 of the two ways ^3, it performs measurement. Thereby, the shape of each micropulse can be measured. And it adjusts based on all the measurement results.

  The shape in the horizontal direction of each micropulse can be adjusted to substantially the same shape. Thereby, the shape of the macro pulse in the horizontal direction can be optimized. That is, the change in the lateral shape according to time is reduced, and cylindrical pulsed light can be generated. The macro pulse can be shaped into a top flat shape with a simple configuration. By finely shielding one micro pulse by the stack unit 33 at each stage, fine adjustment of the pulse shape can be performed.

  Of course, the shape of the macro pulse in the horizontal direction may be adjusted according to the measurement result of the beam monitor 55. What is necessary is just to irradiate the photocathode 51 with pulsed light in a state where it is shielded by the light shielding member 37. Then, the pulse shape is adjusted based on the measurement result of the beam monitor. Thereby, further fine adjustment becomes possible. Furthermore, some micro pulses may be emitted from the stack unit 33 at the same time for adjustment. For example, two or four micro pulses may be emitted from the stack unit 33 at the same time and adjusted.

  Thus, the pulse shape in the vertical direction and the horizontal direction can be precisely shaped. That is, the three-dimensional shape of the pulsed light can be easily shaped. Since it can be measured for each micro pulse, more precise shaping is possible. Then, the photocathode 51 is irradiated with a macro pulse including eight micro pulses with an optimal setting. The photocathode can be irradiated with pulsed light having a desired shape. Thereby, the quality of the electron beam from the photocathode 51 can be improved. That is, a desired bunch-shaped electron beam can be obtained. It is possible to generate an electron beam with high brightness and low emittance. Furthermore, a bunch having an optimum shape for use can be generated. With a simple configuration, the vertical profile and the horizontal profile can be optimized. In addition, as shown in “Wefers AND Nelson, IEEE J. Quantum Electron, Vol. 32, pp. 161 (1996)”, the temporal waveform is reflected in the spatial waveform due to the influence of diffraction. In the combination of the mirror 40 and the pulse stacker 20, the influence on each other can be reduced. Therefore, a three-dimensional shape can be accurately shaped.

  After fine adjustment of each micro pulse, feedback control is performed based on the measurement result of the electron beam and the pulse shape as described above. For example, the bunch shape after being accelerated by the acceleration tube can be adjusted in real time while being monitored by the beam monitor 55. Therefore, a high-quality electron beam can be generated stably. Furthermore, a bunch shape suitable for the purpose of use can be obtained, and for example, emittance can be reduced.

  Since acceleration is performed using microwaves, the energy distribution of the electron beam corresponds to the longitudinal distribution of the bunch. That is, when the phase in the microwave is shifted, the acceleration energy of the electron beam changes. For example, when the pulse width is wide, the energy dispersion of electrons becomes wide. It is also possible to adjust the vertical distribution of the pulsed light according to the measurement result of the energy distribution of the electron beam. When measuring the energy distribution of the electron beam, the electron beam is bent by a bending magnet or the like. The bending magnet generates a constant magnetic field in a direction perpendicular to the incident direction of electrons. Thereby, the electrons in the bunch can be spatially dispersed according to the energy. Then, the spatial distribution of the electrons in the bunch after being bent is measured. Thereby, the pulse shape in the vertical direction (time direction) can be measured. Based on the measurement result of the spatial distribution, the pulse shape in the traveling direction of the pulsed light is adjusted. In this manner, the secondary dispersion can be controlled by the optical modulator 13 such as DAZZLER, and the vertical pulse shape can be adjusted based on the energy dispersion of the electrons in the bunch.

The pulse shape in the vertical direction can also be adjusted by the optical modulator 13. The optical modulator 13 is controlled according to the energy distribution of the bunch. Thereby, the pulse shape in the vertical direction and the bunch shape can be optimized. Furthermore, the pulse shape can be adjusted using the optical modulator 13 so as to compensate for the loss of light in the optical path to the photocathode 51. Even when adjustment is performed in this way, desired bunch shapes and pulse shapes can be easily obtained by making adjustments based on the measurement results of the beam monitor 55 and the camera 47. Furthermore, the pulse shape in the vertical direction can also be adjusted by the pulse compressor 16. That is, the pulse shape in the vertical direction can be optimized by changing the compressor length in the pulse compressor 16. Further, as described above, the secondary dispersion can be canceled by changing the set value of the optical modulator 13. For example, when the optical modulator 13 itself has a second-order dispersion of 13066 fs ² , the secondary dispersion setting value of the optical modulator 13 is previously set to −13066 fs ² to cancel the second-order dispersion. In this state, the Fourier limit pulse can be obtained by changing the compressor length. Furthermore, by changing the set value of the optical modulator 13, compensation is made up to higher-order dispersion, so that it can be brought closer to an ideal Fourier limit pulse. By using DAZZLER (registered trademark) as the optical modulator 13 and providing a recess in the center of the spectrum, the spectrum band can be kept wide after amplification. This makes it possible to further shorten the pulse. By changing the set value of the optical modulator 13, the pulse shape in the vertical direction can be adjusted.

Note that the pulsed light incident on the pulse stacker 20 is not particularly limited. That is, any pulse light having an arbitrary pulse width may be used. Furthermore, the pulsed light emitted from the pulse stacker 20 can be used for purposes other than the generation of electrons. For example, pulsed light can be used for laser processing. In this case, the processing target is irradiated with pulsed light from the pulse stacker 20. By using the laser processing, finer and more precise processing can be performed. Also, pulsed light can be used for calibration of streak cameras and high-speed cameras. As described above, the pulse stacker 20 can easily shape the pulse shape by shielding each micro pulse from light.
When the amount of time delay decreases, adjacent micropulses overlap in the vertical direction. That is, a part of the micro pulse overlaps with the adjacent micro pulse. In such a case, interference between micropulses can be prevented by chirping the pulsed light. That is, since the wavelength of the portion overlapping with the adjacent micro pulse is different, there is no interference. Furthermore, interference between micropulses can be prevented by alternately arranging P-polarized micropulses and S-polarized micropulses. In this way, S-polarized micropulses and P-polarized micropulses are alternately arranged so as to reduce optical interference.

  When an accelerated electron beam is used for oscillation of a free electron laser (FEL), the bunch shape can be monitored while monitoring the oscillated free electron laser. For example, the pulse shape is adjusted in real time so that the oscillation intensity of the oscillated FEL is maximized. Furthermore, when using pulsed light for laser processing, the pulse may be shaped according to the shape of the workpiece. Thus, the shape of the electronic bunch can be optimized according to the application.

  Note that an evaluation function for evaluating whether the bunch has a desired shape may be used. In this case, the pulse light can be easily made flat-topped by adjusting using a plurality of parameters. . For adjusting the deformable mirror 40, it is preferable to use a genetic algorithm. Thereby, it can adjust simply. Alternatively, the pulsed light can be converted into a Gaussian. The pulse shape may be adjusted automatically using the control device 48 or manually by an operator. Further, the adjustment of the deformable mirror 40 may be performed semi-automatically. In addition, by saving past parameters, it is possible to quickly shape a desired profile.

  In the pulse stacker 20, the incident side λ / 2 plate 22, the stacking λ / 2 plate 24, and the emission side λ / 2 plate 35 are rotatably provided. The rotation axes of the incident side λ / 2 plate 22, the stacking λ / 2 plate 24, and the emission side λ / 2 plate 35 are parallel to the optical axis. The light intensity of the pulsed light can be adjusted by adjusting the rotation angles of the incident side λ / 2 plate 22, the stacking λ / 2 plate 24, and the emission side λ / 2 plate 35. Furthermore, the light intensity of each micro pulse can be adjusted by independently adjusting the rotation angles of the three stacking λ / 2 plates 24. Therefore, it becomes possible to shape the pulse shape in the vertical direction.

  In the above description, the deformable mirror 40 is used as the adjusting means for adjusting the two-dimensional shape of the pulsed light, but the adjusting means is not limited to this. For example, a micromirror array in which micromirrors are arranged in an array can be used as the adjusting means. As the micromirror array, it is preferable to use one from Texas Instruments. Of course, other adjustment means may be used.

  It is also possible to use a polarizer disposed rotatably as the light shielding member 37. That is, instead of the movable shutter, a polarizing plate that can be rotated can be used. For example, polarizing plates are respectively arranged in the optical paths of two light beams branched by the branching PBS 28. Then, each polarizing plate is rotated to shield the P-polarized micropulse or the S-polarized micropulse. For example, one polarizing plate shields P-polarized micropulses and the other polarizing plate transmits S-polarized micropulses. In this case, it arrange | positions so that the polarization plane of a P polarization | polarized-light micropulse and the absorption axis of a polarizing plate may correspond. The polarizing plane of the S-polarized micropulse and the absorption axis of the polarizing plate are arranged so as to be orthogonal. Even in this way, it is possible to block the incidence of micropulses on the combining PBS 29. Further, it may be shielded from light using a liquid crystal panel or the like. The light shielding member 37 may not completely shield the micro pulse.

  The time interval between micro pulses is not limited to 2.5 psec. Of course, all the micro pulses need not be equally spaced. In this case, the amount of time delay in the stack unit at each stage should not be an integral multiple.

  Furthermore, the pulse width incident on the pulse stacker 20 may be set in consideration of wavelength dispersion in optical elements such as lenses and mirrors arranged up to the photocathode 51. For example, in the case of a transmissive optical element, the dispersion Will widen the pulse width. In such a case, the pulse width of the pulsed light incident on the photocathode 51 is wider than the pulse width of the pulsed light incident on the pulse stacker 20. Accordingly, the pulse width of the pulsed light incident on the pulse stacker 20 is set so as to compensate for the extension of the pulse width due to chromatic dispersion. By measuring the electron bunch length and the pulse width of the pulse light branched by the beam splitter 46, the pulse light incident on the photocathode 51 can be set to a desired pulse width. Note that the pulse width may be measured using a streak camera or the like.

  Further, the non-polarized component of the pulsed light leaking from the combining PBS 29 may be measured. The non-polarized component present in the pulsed light incident on the stack unit 31 leaks from the combining PBS 29. For this reason, the non-polarized light component does not enter the stacking λ / 2 plate 24 or the mirror 34 from the combining PBS 29. Light that leaks without being synthesized by the synthesis PBS 29 can be detected. Thereby, light can be used effectively. Furthermore, since pulses having substantially the same shape can be measured, more accurate measurement is possible. For example, the leaking light may be detected and the lateral pulse shape may be measured. In this case, the non-polarized light component is measured with a camera or a beam profiler. Here, it is assumed that the non-polarized light component is measurement light incident on a camera or the like, and the polarized light component is light used for generating electrons in the photocathode 51. When the combining PBS 29 combines the two light beams, the measuring light and the utilization light are branched from the combining PBS 29 and emitted. Then, the pulse shape of the measurement light is measured by detecting the measurement light with the camera 47 or the like. And based on the measurement result of the pulse shape of measurement light, the pulse shape of utilization light is adjusted. Thereby, feedback control of the pulse shape of utilization light can be carried out.

  Further, in the pulse stacker 20 having the configuration shown in FIG. 3, when performing measurement according to one micropulse, the measurement can be performed without arranging an optical element in the optical path of the micropulse. Therefore, masking can be performed without breaking the system. In other words, each micropulse can be easily masked by inserting the light shielding member 37 into the optical path. Thereby, adjustment of an optical system etc. becomes unnecessary and accurate measurement can be performed simply.

  Next, a modified example of the pulse stacker 20 will be described with reference to FIG. FIG. 5 is a view showing a stacking rod 120 for generating micropulses having a configuration different from that of FIG. FIG. 5 shows a stacking rod 120 that stacks eight micro pulses from one pulsed light. Therefore, the stacking rod 120 is provided with three-stage stack units 131, 132, and 133. In addition, description is abbreviate | omitted about the content similar to the pulse stacker 20. FIG.

  Each of the stack units 131 to 133 includes a wedge 121, a wedge 122, a birefringent element 123, an extraction λ / 2 plate 124, and an extraction PBS 125. Further, the stack units 131 and 132 have a stacking λ / 2 plate 126. Each of the stack units 131 to 133 basically has the same configuration. Here, three stack units 131 to 133 are arranged in series.

  First, the pulsed light is incident on the first stack unit 131. That is, the pulsed light is incident on a wedge pair including the wedge 121 and the wedge 122. The pulsed light refracted by the wedge pair enters the birefringent element 123. By rotating the wedge 121 and the wedge 122 independently, the position and incident angle of the pulsed light incident on the birefringent element 123 can be adjusted.

The birefringent element 123 is a crystal element such as α-BBO, for example, and produces a birefringence effect on incident pulse light. This causes a time delay between the ordinary ray and the extraordinary ray in the pulsed light. That is, the propagation speed of the polarization component parallel to the slow axis is slower than that of the polarization component parallel to the fast axis. This gives a time delay between the orthogonal polarization components. The optical axis of the birefringent element 123 is arranged perpendicular to the optical axis of the pulsed light. If the P-polarized component is an ordinary ray and the S-polarized component is an extraordinary ray, the S-polarized component is delayed compared to the P-polarized component. For example, the birefringent element 123 causes a delay of 10 psec. As a result, as shown in FIG. 4A, two micro pulses shifted by 10 psec are generated. As a birefringent element, www. pantotek. com / Birefringent crystals / AlphaBBO. Crystals described in html and the like can be used. Alternatively, α-BBO crystal manufactured by CRYSTECH can also be used. Further, instead of the α-BBO crystal, a YVO ₄ crystal or a liquid crystal element may be used as the birefringent element 123. For example, by using a YVO ₄ crystal, it becomes possible to shape pulsed light having a wavelength of 400 nm to 5 μm.

  Similarly to the stacking λ / 2 plate 24, the stacking λ / 2 plate 126 rotates the polarization plane of linearly polarized light by 45 °. Therefore, as shown in FIG. 4B, + 45 ° micropulses and −45 ° micropulses are generated. The stacking λ / 2 plate 126 makes micropulses polarized by 45 degrees. Then, two micro pulses are incident on the second stack unit 132.

  The configuration of the stack units 132 and 133 is the same as that of the stack unit 131. However, the delay amount in the birefringent element 123 is different in each unit. The delay time in the birefringent element 123 of the stack unit 132 is 5 psec. Therefore, in the birefringent element 123 of the stack unit 132, four micro pulses as shown in FIG. 4D are generated. Then, the state shown in E of FIG. 4 is obtained by the stacking λ / 2 plate 126.

  Further, the delay time in the birefringent element 123 in the third stack unit 133 is 2.5 psec. Therefore, when passing through the birefringent element 123 of the stack unit 133, the state shown in FIG. Only the stack unit 133 at the third stage is not provided with the stacking λ / 2 plate 126. P-polarized micropulses and S-polarized micropulses are alternately arranged. A predetermined delay time can be obtained by changing the thickness of the α-BBO crystal. That is, the length of the birefringent element 123 is determined according to a given time delay. When increasing the delay time, the birefringent element 123 is made thick. For example, when a time delay of 10 psec is given using an α-BBO crystal rod manufactured by CRYTECH, the thickness of the crystal rod needs to be 2 cm. When a time delay of 5 psec is given, the thickness is 1 cm. When a time delay of 2.5 psec is given, the thickness is 0.5 cm. When increasing the delay time, it is preferable to use the birefringent element 123 having a high transmittance. Thereby, attenuation of the pulsed light in the birefringent element 123 can be reduced, and pulsed light with sufficient intensity can be obtained.

  In each stack unit, an extraction λ / 2 plate 124 and an extraction PBS 125 are inserted in the subsequent stage of the birefringent element 123 so as to be insertable. That is, in the stack units 131 and 132, the extraction λ / 2 plate 124 and the extraction PBS 125 are inserted between the birefringent element 123 and the stacking λ / 2 plate 126. Further, in the stack unit 133, the extraction λ / 2 plate 124 and the extraction PBS 125 can be arranged behind the birefringent element 123. For example, when the take-out λ / 2 plate 124 and the take-out PBS 125 are slid in the direction of the arrow, the take-out λ / 2 plate 124 and the take-out PBS 125 are on the optical path as shown by the dotted line in FIG. Be placed. At the time of adjusting the pulse shape, the extraction λ / 2 plate 124 and the extraction PBS 125 are arranged in the optical path. Then, the profile of each micro pulse light is measured. On the other hand, when the electron beam is used, the extraction λ / 2 plate 124 and the extraction PBS 125 are removed from the optical path. Thereby, it propagates through the optical system in a state where eight micro pulses are stacked. As described above, the extraction λ / 2 plate 124 and the extraction PBS 125 are detachably arranged in the optical path.

  When the extraction λ / 2 plate 124 and the extraction PBS 125 are arranged in the optical path, one micropulse does not enter the stacking λ / 2 plate 126. That is, the two micropulses from the birefringent element 123 propagate on the same optical axis and enter the extraction λ / 2 plate 124. However, an extraction PBS 125 is disposed behind the extraction λ / 2 plate 124. By adjusting the rotation angle of the extraction λ / 2 plate 124, only one micro pulse passes through the extraction PBS 125, and the other micro pulse is reflected. That is, only one of the P-polarized micropulse and the S-polarized micropulse passes through the extraction PBS 125 and enters the stacking λ / 2 plate 126. In other words, the extraction PBS 125 shields the other of the P-polarized micropulse and the S-polarized micropulse from entering the stacking λ / 2 plate 124. In this manner, the extraction PBS 125 blocks one micropulse from entering the stack unit 132 at the next stage according to the polarization state.

  Thus, in the state where the extraction λ / 2 plate 124 and the extraction PBS 125 are arranged in the optical path, one micropulse enters the stack unit 132. In a state where the extraction λ / 2 plate 124 and the extraction PBS 125 are removed from the optical path, both the P-polarized micropulse and the S-polarized micropulse enter the stack unit 132. Therefore, in the three stack units 131 to 133, when the extraction λ / 2 plate 124 and the extraction PBS 125 are inserted into the optical path, only one of the eight micro pulses is extracted as described above. Then, the extraction λ / 2 plate 124 is rotated in accordance with the micropulse to be extracted.

  Thus, in the stacking rod 120, the birefringent element 123 gives a time delay between the orthogonal polarization components of the pulsed light. A means for taking out one of the orthogonal polarization components of the pulsed light emitted from the birefringent element 123 from the other polarization component is detachably disposed in the optical path. Then, with the means for taking out one of the orthogonal polarization components of the pulsed light from the other polarization component inserted in the optical path, the pulse shape of the pulsed light from the birefringent element 123 is adjusted using the deformable mirror 40 or the like. To do. In order to adjust the pulse shape by the deformable mirror 40 or the like, measurement according to one polarization component is performed by the camera 47 or the beam monitor 55 or the like. By using the stacking rod 120 having such a configuration for the electron gun 100, the adjustment for each micro pulse can be performed as described above. Also, the adjustment can be easily automated.

  In the configuration of FIG. 4, the stacking λ / 2 plate 126 for 45-degree polarization is inserted between the birefringent elements 123, but it can be omitted. For example, the crystal axis of the birefringent element 123 may be rotated to tilt the crystal axis by 45 degrees. That is, by rotating the crystal rod instead of the stacking λ / 2 plate 126, the power of each polarization component can be adjusted. Thereby, the configuration can be greatly simplified.

  Furthermore, it is also possible to measure the polarization component branched by the extraction PBS 125. That is, the profile of the polarized component reflected without passing through the extraction PBS 125 may be measured, and the pulse shape may be adjusted based on the measurement result. For example, when a P-polarized micropulse is extracted by the extracting PBS and is incident on the stacking λ / 2 plate 126, the S-polarized micropulse reflected by the extracting PBS 125 is measured by a camera or the like. Then, the profile is adjusted based on the measurement result of the S-polarized micropulse. The means for extracting one of the polarization components may be a polarizing plate provided so as to be insertable / removable in the optical path. By rotating the polarizing plate, S-polarized micropulses or P-polarized micropulses can be taken out.

Embodiment 2 of the Invention
In the present embodiment, in the electron gun 100, a pulse stacker 70 shown in FIG. 6 is used instead of the pulse stacker 20 shown in FIG. The configuration other than the pulse stacker is the same as that shown in FIG. In the pulse stacker 70 shown in FIG. 6, four stack units 83a to 83d are used. Four stack units 83a to 83d are arranged in parallel. Thereby, eight micro pulses can be generated from one pulse. In the pulse stacker 70 as well, the description of the same contents as in the first embodiment is omitted. Furthermore, since the stack units 83a to 83d have the same configuration as the stack units 31 to 33 shown in the first embodiment, a part of the configuration is not shown. For example, the actuator 27, the fixed mirror pair 25, and the movable mirror pair 26 have the same configuration as in the first embodiment, and are not shown. In the present embodiment, the pulse stacker 70 has a configuration for making pulsed light into a rugby ball-like ellipsoid shape.

  A front stage side λ / 2 plate 84, a front stage side λ / 2 plate 85a, a front stage side λ / 2 plate 85b, a front stage side PBS 71, a front stage side PBS 71a, and a front stage side PBS 71b are arranged at the front stage of the stack units 83a to 83d. Yes. Further, in the subsequent stage of the stack units 83a to 83d, a mirror 81, a rear-stage branching beam splitter 74, a rear-stage branching beam splitter 76, a rear-stage branching beam splitter 78, a rear-stage synthesis beam splitter 75, and a rear-stage synthesis. A beam splitter 77, a rear-side synthesis beam splitter 79, an aperture 86b, an aperture 86c, and an aperture 86d are provided.

  In the subsequent stage of the stack units 83a to 83d, synthesis units 88a to 88c are provided. The synthesizing unit 88 a includes a rear stage side branching beam splitter 74, a rear stage side synthesizing beam splitter 75, and two mirrors 81. The synthesizing unit 88 b includes a rear stage side branching beam splitter 76, a rear stage side synthesizing beam splitter 77, and two mirrors 81. The synthesizing unit 88c includes a rear stage side branching beam splitter 78, a rear stage side synthesizing beam splitter 79, and two mirrors 81. Note that the synthesis unit 88a, the synthesis unit 88b, and the synthesis unit 88c have the same configuration.

  The pulsed light from the light source unit 10 enters the front-side λ / 2 plate 84. The front stage side λ / 2 plate 84 rotates the polarization axis of the pulsed light that is linearly polarized light. This rotates the polarization axis by 45 °. Then, the pulsed light emitted from the front stage side λ / 2 plate 84 enters the front stage side PBS 71. The first-stage front-side PBS 71 branches the incident pulsed light according to the polarization state. As a result, the light beam is branched into an S-polarized light beam and a P-polarized light beam. The P-polarized component light beam enters the front-stage PBS 71a via the second-stage front-side λ / 2 plate 85a, and the S-polarized component light beam enters the second-stage branching λ / 2 plate 85b. Then, it enters the pre-stage side PBS 71b. The λ / 2 plates for branching 85a and 85b have their polarization axes rotated by 45 °. In this way, the λ / 2 plate and the PBS are alternately passed twice. Thereby, the pulsed light is branched into four light beams. Specifically, two P-polarized beams and two S-polarized beams are generated.

  The four light beams are incident on the respective stack units 83a to 83d. Specifically, a P-polarized beam from the front-stage side PBS 71a enters the stack unit 83a, and an S-polarized beam from the front-stage side PBS 71a enters the stack unit 83b. Further, the P-polarized beam from the front-stage side PBS 71b enters the stack unit 83c, and the S-polarized beam from the front-stage side PBS 71b enters the stack unit 83d.

  The stack units 83 a to 83 d have substantially the same configuration as the stack unit 31. For example, the light beam incident on the stack unit 83a is branched into a P-polarized micropulse and an S-polarized micropulse. Here, the optical path length differences given in the stack units 83a to 83d are different. For example, in the stack unit 83a, an optical path length difference corresponding to a delay time of 2.5 psec is given. As a result, the S-polarized micropulse emitted from the stack unit 83a is delayed by 2.5 psec from the P-polarized micropulse. The stack unit 83b gives an optical path length difference corresponding to 7.5 psec delay time, the stack unit 83c gives an optical path length difference equivalent to 12.5 psec delay time, and the stack unit 83d gives 17 An optical path length difference corresponding to a delay time of .5 psec is given.

  Then, the micro pulses from the four stack units 83a to 83d are synthesized in order. Here, the eight micropulses are sequentially synthesized by the synthesis units 88a to 88c. Here, three synthesis units 88a, three synthesis units 88b, and a synthesis unit 88c are arranged in series. Then, the synthesis unit 88a synthesizes the micro pulses from the stack units 83a and 83b. As a result, a total of four micropulses are synthesized. The synthesis unit 88b synthesizes the micro pulse from the synthesis unit 88a and the micro pulse from the stack unit 83c. As a result, a total of six micro pulses are synthesized. The synthesis unit 88c synthesizes the micro pulse from the synthesis unit 88b and the micro pulse from the stack unit 83d. As a result, a total of eight micropulses are synthesized. Below, the synthesis | combination of the micro pulse in the synthesis | combination units 88a-88c is demonstrated in detail.

  The micro pulse of the stack unit 83a is incident on the rear-stage branching beam splitter 74 provided in the first-stage synthesis unit 88a via the mirror 81. The micro pulse of the stack unit 83b is incident on the post-stage branching beam splitter 74 provided in the first-stage synthesis unit 88a via the aperture 86b. The rear-stage branching beam splitter 74 is a non-polarizing beam splitter. Therefore, the incident light is divided regardless of the polarization state. As a result, the S-polarized micropulse and the P-polarized micropulse from the stack units 83a and 83b are branched. Then, the branched micro pulse is reflected by the mirror 81 and enters the post-stage synthesis beam splitter 75. The post-stage combining beam splitter 75 is a non-polarizing beam splitter, and combines the light beams branched by the post-stage combining beam splitter 75 regardless of the polarization state. As a result, four micro pulses are synthesized.

  Here, an aperture 86b is disposed between the stack unit 83b and the synthesis unit 88a. The aperture 86b has, for example, a circular opening. Accordingly, the light incident on the outside of the opening of the aperture 86b is shielded and cannot enter the post-stage branching beam splitter 74. Only the micropulses that have passed through the aperture 86b enter the post-stage branching beam splitter 74. Thereby, the beam diameter of the micro pulse from the stack unit 83b is limited and becomes smaller than the beam diameter of the micro pulse from the stack unit 83b. Thereby, the profile of the beam in the lateral direction changes.

  Further, the optical distance is changed between the stack unit 83a and the stack unit 83b. Thereby, an optical path length difference is given to the four micropulses. The timing of each micro pulse is shifted. Here, two micro pulses from the stack unit 83a are arranged between two micro pulses from the stack unit 83b. Then, four micro pulses are emitted from the synthesis unit 88a along the same optical axis.

  The micro pulse of the synthesizing unit 88a enters the post-stage branching beam splitter 76 provided in the second synthesizing unit 88b via the mirror 81. The micro pulse of the stack unit 83c is incident on the rear-stage branching beam splitter 76 provided in the second-stage synthesis unit 88b via the aperture 86c.

  The second synthesis unit 88b synthesizes the micro pulse from the synthesis unit 88a and the micro pulse from the stack unit 83c in the same manner. As a result, the four micropulses and the two micropulses are combined by the combining beam splitter 77 and emitted. Here, the optical distance is changed between the stack units 83a to 83c. Thereby, an optical path length difference is given to six micro pulses. The timing of each micro pulse is shifted. Here, four micro pulses from the synthesis unit 88a are arranged between two micro pulses from the stack unit 83c. Then, six micro pulses are emitted from the synthesis unit 88b along the same optical axis.

  Here, an aperture 86c is disposed between the stack unit 83c and the synthesis unit 88b. The aperture 86c has, for example, a circular opening. Therefore, the light that has entered the outside of the opening of the aperture 86c is shielded and cannot enter the post-stage branching beam splitter 76. Only the micro pulse that has passed through the aperture 86c is incident on the post-stage branching beam splitter 76. Thereby, the beam diameter of the micro pulse from the stack unit 83c is limited. Further, the opening of the aperture 86c is made smaller than the opening of the aperture 86b. Therefore, the beam diameter of the micro pulse from the stack unit 83c is smaller than the beam diameter of the micro pulse from the stack unit 83b. The aperture 86b changes the beam profile in the lateral direction.

  The micro pulse of the synthesis unit 88b is incident on the rear-stage branching beam splitter 78 provided in the third-stage synthesis unit 88c via the mirror 81. The micro pulse of the stack unit 83d is incident on the post-stage branching beam splitter 78 provided in the synthesis unit 88c through the aperture 86d.

  The third synthesis unit 88c synthesizes the micro pulse from the synthesis unit 88b and the micro pulse from the stack unit 83d in the same manner. As a result, the six micropulses and the two micropulses are combined by the combining beam splitter 79 and emitted. Here, the optical distance is changed between the stack units 83a to 83d. Thereby, an optical path length difference is given to eight micro pulses. The timing of each micro pulse is shifted. Here, six micro pulses from the synthesis unit 88b are arranged between two micro pulses from the stack unit 83d. Then, eight micro pulses are emitted from the synthesis unit 88d along the same optical axis.

  An aperture 86d is disposed between the stack unit 83d and the synthesis unit 88c. The aperture 86d has a circular opening that is smaller than the aperture 86c. Thereby, the beam diameter of the micro pulse from the stack unit 83d is smaller than the beam diameter of the micro pulse from the stack unit 83c. In this way, the apertures 86b to 86d limit the passage of the beam, so that the micropulse profile changes.

  Thus, the pulsed light emitted from the pulse stacker 70 has eight micro pulses arranged at predetermined time intervals. The beam diameter is changed for each micro pulse by the apertures 86b to 86c. That is, as shown in FIG. 6, the first and eighth micro pulses among the macro pulses pass through the aperture 86d, so that the beam diameter is minimized. In addition, since the second and seventh micro pulses pass through the aperture 86d, the beam diameter becomes the second smallest. Since the third and sixth micropulses pass through the aperture 86c, the beam diameter becomes the second largest. Since the fourth and fifth micro pulses do not pass through the aperture, the beam diameter is maximized.

  The beam diameter of the entire macro pulse increases with time and becomes maximum near the center of the macro pulse in the vertical direction. That is, the beam diameter becomes maximum at the timing corresponding to the fourth and fifth micro pulses. The beam diameter decreases with time from the vicinity of the center of the macro pulse in the vertical direction. The beam profile in the vertical direction changes with time. The profile in the vertical direction is symmetric. By using such a pulse stacker 70, an ellipsoidal macro pulse can be obtained. Of course, shaping other than the ellipsoid shape is also possible. As in the first embodiment, pulse shaping can be easily performed. Furthermore, various shapes of pulsed light can be shaped with a simple configuration. Further, by making the aperture diameters of the apertures 86b to 86d variable, they can be shaped into various shapes.

  In the above description, the profile is changed using the apertures 86b to 86d, but the profile changing means for changing the profile is not limited to this. For example, the profile can be changed using a pair of lenses. That is, a down collimating lens composed of a lens pair can be used. Then, the beam diameter is reduced to change the profile. When a lens pair that is a reduction optical system is used, the beam diameter can be changed without reducing the amount of light. Of course, the beam diameter may be enlarged by using the lens pair as an enlargement optical system. Furthermore, the profile may be changed using both the lens pair and the aperture. Furthermore, the profile may be changed using a spatial filter or the like having a different transmittance according to the position. It is preferable to change the beam diameter while maintaining the light beam as a parallel light beam. Further, an aperture or the like may be arranged between the synthesis unit 88a and the stack unit 83a to change the profile.

  Thus, in the present embodiment, the stack units 83a to 83d are arranged in parallel. For this reason, the profile changing means can be arranged in the optical path before synthesis. For this reason, it becomes possible to adjust micropulses individually. As a result, the pulsed light can be shaped into various shapes. In particular, adjustment in the vertical direction (direction parallel to the optical axis) is facilitated. In addition, since the P-polarized micropulses and the S-polarized micropulses are alternately arranged, optical interference can be reduced.

  As in the first embodiment, by adjusting the angle of the λ / 2 plate 24, the ratio of the light intensity of the P-polarized micropulse and the S-polarized micropulse can be adjusted. Further, by inserting the light shielding member 37, it is possible to measure individual micropulse shapes. Thereby, shaping of a more precise pulse shape becomes possible. Furthermore, one micro pulse can be removed by inserting the light shielding member 37 or rotating the λ / 2 plate 24. Thereby, the freedom degree which adjusts a beam diameter becomes high. That is, all micropulses can have different beam diameters. Therefore, pulse light having a desired shape can be obtained.

  Further, in the present embodiment, a measurement unit 90 that measures light from the latter-stage synthesis beam splitters 75, 77, and 79 is provided. The measuring unit 90 efficiently uses the light that leaks when combining the light. For example, the non-polarizing beam splitter is provided in the post-side synthesis beam splitter 75. For this reason, a loss occurs when combining the post-stage side combining beam splitter 75 light, and a part of the light does not enter the mirror 81. In other words, a light beam that does not enter the mirror 81 is incident on the measurement unit 90 from the post-stage synthesis beam splitter 75. One of the two light beams emitted from the post-stage synthesis beam splitter 75 enters the mirror 81, and the other enters the measurement unit 90. That is, the measurement unit 90 measures a light beam that is not the light beam that is combined by the post-stage-side combining beam splitter 75 and directed to the mirror 81.

  Here, the light beam incident on the measurement unit 90 among the light beams emitted from the post-stage-side synthesis beam splitter 75 is defined as measurement light. The light beam incident on the mirror 81 out of the light beams emitted from the post-stage synthesis beam splitter 75 is used light. The utilized light is incident on the photocathode 51 and used for generating electrons as described above. The measurement unit 90 measures the pulse shape using measurement light. Thereby, light can be used efficiently. Furthermore, pulse shaping while monitoring micro pulses is possible.

  The measuring unit 90 includes a camera for measuring the pulse shape. For example, a lateral profile can be measured by a two-dimensional camera or the like. Furthermore, the vertical profile may be measured by the measuring unit 90. Here, a preferred configuration example for measuring the profile in the vertical direction will be described with reference to FIG. FIG. 7 is a diagram showing the configuration of the measurement unit 90 that measures the profile in the vertical direction. The measurement unit 90 includes a measurement PBS 91, a mirror 92, a nonlinear optical element 93, a spectroscope 94, and a probe light modulator 95.

  A light beam from the post-synthesis beam splitter 75 is incident on the measurement PBS 91. Here, a light beam having two micro pulses is incident. For example, the stack unit 83b shields the micro pulse from the stack unit 83b. Specifically, two micropulses can be shielded by inserting the light shielding member 37 in the stack units 83a and 83b. In this case, a double pulse having two micro pulses is extracted from the post-stage synthesis beam splitter 75. The measurement PBS 91 reflects the double pulse in the direction of the mirror 92. Here, the double pulse which is the measurement object of the pulse shape is defined as measurement light.

  Further, the chirp pulse chirped by the probe optical modulator 95 is incident on the measurement PBS 91. This chirp pulse becomes probe light used for measuring the pulse shape. The chirp pulse has a pulse width longer than double pulse interval + pulse width. That is, the entire double pulse overlaps with the chirp pulse. This chirp pulse is used as probe light. The wavelength of the probe light changes with time. The chirp pulse is chirped into a rectangular shape by the probe light modulator 95. More preferably, it is a rectangle. You may make it use a center part near a flat by lengthening a chirp. However, there is a problem that when the chirp is lengthened, the laser intensity decreases. That is, in the chirp pulse, the wavelength changes with time. The chirped pulsed light can be extracted from the light source unit 10. For example, a part of the pulsed light emitted from the regenerative amplifier 14 or the multipath amplifier 15 can be extracted and used as a chirp pulse. Here, for example, the fundamental wave (790 nm) is extracted as the probe light. The measurement PBS 91 allows the chirp pulse to pass through while preventing the double pulse side pulse from spreading due to dispersion caused by transmission through the medium. Thereby, a double pulse and a chirp pulse are combined. Here, the chirp pulse and the double pulse are incident on the measurement PBS 91 in synchronization.

For example, the polarization direction of the double pulse is rotated by 45 °. Here, the polarization direction is rotated using a wavelength plate, a polarizer, or the like. In the measurement PBS 91, the S-polarized component or the P-polarized component of the double pulse is extracted. Further, a chirp pulse is taken out from the measurement PBS 91. Therefore, it is efficiently synthesized by the measurement PBS 91. The pulsed light synthesized by the measurement PBS 91 is reflected by the mirror 92 and enters the nonlinear optical element 93. The nonlinear optical element 93 is a nonlinear optical crystal such as KDP, for example. The nonlinear optical element 93 generates second harmonic generation (SHG), difference frequency generation (DFG), and sum frequency generation (SFG). Here, the non-linear optical element 93 generates a second harmonic or a sum frequency and a difference frequency of a double pulse and a chirp pulse. Note that the second harmonic can also be regarded as the sum frequency due to the nonlinear optical effect at the same wavelength. Since the chirp pulse and the double pulse are synchronized, the timings of incidence on the nonlinear optical element 93 coincide. In addition, the chirp pulse has a pulse width sufficiently longer than that of the double pulse. Thereby, a sum frequency or a difference frequency is efficiently generated.
For example, when the double pulse that is the measurement light is a triple wave, a difference frequency between the chirp pulse of the fundamental wave and the double pulse is generated. As a result, a double wave of the fundamental wave is generated. Even when the measurement light is a third harmonic wave in the ultraviolet region, it can be easily measured by using the difference frequency. Further, when the double pulse that is the measurement light is a fundamental wave, a double wave of the fundamental wave may be generated by the sum frequency of the chirp pulse and the double pulse.

  The sum frequency or difference frequency generated by the nonlinear optical element 93 enters the spectroscope 94. The spectroscope 94 measures the spectrum of the incident sum frequency. For example, the spectroscope 94 spatially disperses pulsed laser light incident through a slit or the like provided on the incident side according to the wavelength. In the case of the spectroscope 94 using a reflective diffraction grating, an optical system such as a concave mirror that guides light from the entrance slit to the spectroscopic element and a concave mirror that guides light split by the spectroscopic element to the two-dimensional array camera is provided. It has been. Of course, a spectroscope 94 having a configuration other than the above may be used. Incident light is dispersed by the spectroscope 94 in a direction perpendicular to the direction of the incident slit. That is, the spectroscope 94 wavelength-disperses the emitted light in a direction perpendicular to the line-shaped opening of the entrance slit. The outgoing light separated by the spectroscope 94 enters the two-dimensional array camera. In the two-dimensional array camera, light receiving elements are arranged in a matrix.

  By being split by the spectroscope 94, light of different wavelengths enters different light receiving pixels. That is, the incident light receiving pixel is shifted according to the wavelength. Thereby, the spectrum measurement of the sum frequency or the difference frequency becomes possible. For example, the light with the longest wavelength enters the light receiving pixel on one end side of the light receiving pixel row, and the light with the shortest wavelength enters the light receiving pixel on the other end side. The incident light receiving pixels are different depending on the wavelength. Further, since the chirped pulse laser beam is used, the light receiving pixel array corresponds to the time in the pulse laser beam. Therefore, light at different timings in the pulse enters different light receiving pixels. For example, the light at the front end of the pulse is incident on a light receiving pixel on one end side of the light receiving pixel row, and the light at the rear end of the pulse is incident on a light receiving pixel on the other end of the light receiving pixel row. Then, the spectrum of one pulse is detected by one frame of the camera of the spectroscope 94. That is, the spectrum of one pulse of the pulsed laser beam is acquired with one frame of the camera of the spectroscope 94. Thereby, the spectral distribution of sum frequency or difference frequency can be measured at high speed.

  Thus, the spectroscope 94 measures the spectrum of the sum frequency or the difference frequency of the double pulse and the chirp pulse. Furthermore, since a chirp pulse is used, the wavelength corresponds to time. The time interval of the micro pulses included in the double pulse corresponds to the peak wavelength interval of the spectrum. The spectrum includes information on the shape of a double pulse in the vertical direction. Thereby, the interval between micropulses can be measured accurately. Specifically, when the pulse interval is τ, the peak wavelength interval is a value based on τ. Thereby, the pulse interval τ can be calculated.

Thus, by using the nonlinear optical element 93 and the spectroscope 94, the pulse shape in the vertical direction can be accurately measured. Note that a double pulse of two colors may be used as the probe light instead of the chirp pulse. In this case, the fundamental pulse laser beam is modulated by the probe optical modulator 95. Then, double pulses having different wavelengths are generated. For example, the first pulse has a short wavelength and the second pulse has a long wavelength. Of course, this may be the opposite. In this way, pulsed light having different wavelengths depending on time is used as probe light.
Then, the two-color double pulse for probe and the double pulse for measurement are synchronized and made incident on the nonlinear optical element 93. That is, the double pulse as measurement light and the two-color double pulse as probe light are superimposed. As a result, a sum frequency and a difference frequency are generated in the nonlinear optical element 93. Here, the double pulse as the measurement light overlaps with the two-color double pulse. The spectroscope 94 measures the sum frequency or difference frequency with the two-color double pulse having different wavelengths.
For measurement of the pulse shape in the vertical direction, for example, PHAZLER (registered trademark) manufactured by FASTLITE can be used. Thereby, a pulse shape can be measured easily. For example, when the fundamental wave is used as measurement light, a double pulse that is measurement light is incident on PHAZLER (registered trademark). Thereby, the sum frequency of the two-color double pulse as the probe light and the measurement light can be generated. Then, the pulse shape in the vertical direction can be measured by measuring the spectrum with a PHAZLER (registered trademark) spectrometer. Thus, by using the probe light modulator 95 using AO modulation, the sum frequency or difference frequency of the probe light and the measurement light can be easily generated.

  Further, the number of micropulses included in the measurement light is not limited to two. For example, the measurement light may include four or eight micro pulses. In such a case, the PBS is detachably disposed at the subsequent stage of the pulse stacker 70. When taking out the eight micropulses, the light is converted to 45 ° polarization by the λ / 2 plate on the emission side of the pulse stacker 70. Then, the S polarization component or the P polarization component is extracted with PBS. Thereby, eight micro pulses can be taken out. Also, as the probe light, eight-color micropulses are prepared. That is, the same number of micropulses as the probe light are prepared as the probe light. Note that the micro pulse of the probe light may be a rectangular pulse. Alternatively, a chirp pulse having a pulse width that covers eight micro pulse trains of measurement light may be prepared as probe light. When extracting four micropulses, for example, one polarization component is masked by the stack unit of each stage. Thereby, for example, only the P-polarized light component is extracted. When measuring four micropulses, every other micropulse can be taken out. For example, it becomes possible to extract only P-polarized micropulses. By doing in this way, each micropulse interval and the shape of the vertical direction of each micropulse can be measured simultaneously. Of course, the above measurement method can also be used for the pulse stacker 20 and the stacking rod 120 shown in the first embodiment.

  Further, the pulse shape in the vertical direction can be measured for each micro pulse based on the measurement result of the spectroscope 94. Signal processing for this purpose will be described with reference to FIGS. The following signal processing can be performed by the control device 48, for example.

  First, the data shown in FIG. 8 is Fourier transformed. Note that the data shown in FIG. 8 is an example of a spectrum measured by the spectroscope 94. The time domain data shown in FIG. 9 is obtained by performing Fourier transform. Since the sum frequency is based on a chirp pulse and a double pulse, as shown in FIG. 9, for example, when a double pulse is taken as an example, two sidebands are generated. Two sidebands appear in the vicinity of ± τ. Here, paying attention to one sideband, the sideband is taken out with a window. Here, a side band appears in the vicinity of time τ. Cut out this sideband in the window. Thereby, the data shown in FIG. 10 is obtained. Here, inverse Fourier transform is performed on the data shown in FIG. As a result, data in the frequency domain is obtained as shown in FIG.

Here, the data shown in FIG. 11 is expressed by the following equation.
φ ₁ (ω) + ωτ

Note that φ ₁ (ω) is a function indicating the shape of the micropulse extracted as a sideband, and τ is the pulse interval as described above. Here, if the term of ωτ is deleted from the data shown in FIG. 11 and the data are joined together, the graph shown in FIG. 12 is obtained. Note that φ ₁ (t) shown in FIG. 12 includes information on frequency intensity distribution (amplitude spectrum) and frequency phase distribution (phase spectrum). When the data shown in FIG. 12 is Fourier-transformed, the time domain data shown in FIG. 13 is obtained. Here, the intensity distribution in the time domain has a pulse shape in the vertical direction.

  By performing such a measurement, the pulse shape in the vertical direction can be accurately measured for each micro pulse. Based on the measurement result, the pulse is shaped. For example, the control device 48 controls the optical modulator 13 based on the measurement result. Thereby, more accurate pulse shaping can be performed. The light leaking from the post-stage synthesis beam splitter 75 has the same shape as the micropulses that are actually stacked. Therefore, highly accurate measurement is possible. Similarly, the light emitted from the post-stage synthesis beam splitters 77 and 79 may be detected. Furthermore, two or more micro pulses may be measured simultaneously. Even in this case, the shape of the micro pulse can be measured by paying attention to each side band.

  Further, pulse light leaking from the combining PBSs 29a to 29d may be measured. The non-polarized component present in the pulsed light incident on the stack unit 83a leaks from the combining PBS 29a. Leaked light can be detected without entering the mirror 81 from the combining PBS 29a. Thereby, light can be used effectively. Furthermore, since pulses having substantially the same pulse shape can be measured, more accurate measurement is possible. Of course, the pulse shape in the lateral direction may be measured by the measuring unit 90 that detects leaking light. In this case, the measuring unit 90 is provided with a camera and a beam profiler. Here, the non-polarized light component becomes the measurement light incident on the measurement unit 90. Then, the polarization component becomes the utilization light incident on the mirror 81.

  The pulse stacker 70 may be used for applications other than the electron gun, as in the first embodiment. Of course, the first embodiment and the second embodiment may be appropriately combined. For example, even in the pulse stacker 70, the micropulse can be shielded by using the light shielding member 37 and the stacking λ / 2 plate 24. Each micropulse can be optimized. Further, in the pulse stacker 70 shown in the second embodiment, the incident side λ / 4 plate 21, the incident side λ / 2 plate 22, and the incident side PBS 23 may be used. Similarly, in the pulse stacker 70 shown in the second embodiment, the emission side λ / 2 plate 35 may be used. In the first and second embodiments, each micro pulse may not be completely deviated. That is, some of the micro pulses may overlap in time.

As described above, in the pulse stacker, one of the light beams after being branched by the branching PBS 28 is shielded by the light shielding member 37. For this reason, each micro pulse can be measured more accurately, and the pulse shape can be finely adjusted. Further, in the pulse stacker 70 according to the second embodiment, the beam profile is adjusted using the apertures 86b to 86d. Thereby, shaping for every micro pulse is attained. Further, the pulse shape is measured by the camera 47 and the measurement unit 90. Further, the beam monitor 55 performs measurement according to the light beam. Then, the pulse shape is shaped based on these measurement results. Thereby, it becomes possible to adjust more accurately. Therefore, the pulse shape can be shaped with high accuracy, and the desired bunch shape can be obtained.
It may overlap in time.

  In addition, the configurations shown in the first and second embodiments enable shaping of pulsed light in the deep ultraviolet and vacuum ultraviolet region to 157 nm. Thus, the present invention is suitable for shaping pulsed light having a wavelength of 157 nm to 400 nm. Furthermore, since it can be used in a wide wavelength band, it becomes possible to shape an ultrashort pulse.

Example The measurement result of the pulse shape shaped by the above-described pulse shaping device is shown in FIGS. 14 to 22 are diagrams showing the energy spectrum of the electron beam when the secondary dispersion in the optical modulator 13 is changed. Here, as described above, the electron beam in the bunch of the electron beam was spatially dispersed by the bending magnet. Then, a beam profiler was used as the beam monitor 55, and the lateral spatial distribution of the dispersed electron beam in the bunch was measured. As described above, the curvature of the bending magnet changes depending on the energy of electrons. Further, in the RF electron gun, the acceleration energy varies depending on time. Therefore, the pulse shape in the vertical direction was measured by measuring the energy distribution of the electron beam in the bunch.

14 to 22 are gray scale image data showing the spatial distribution of the electron beam in the dispersion section. In addition, a dispersion | distribution part is a location after bending an electron beam with a bending magnet. 14 to 22, electrons are dispersed in the horizontal direction according to energy. Here, in FIGS. 14 to 22, the left side is the high energy side, which is the tip side of the pulsed light. In FIGS. 14 to 22, the color image data acquired by actual measurement is displayed in gray scale. 14 to 22 show profile data at dotted lines.
Further, the second-order dispersion setting value of the optical modulator 13 is set to a negative value in order to perform second-order dispersion zero-pointing. Here, the set value of the optical modulator 13 is set to −13066 fs ² . And in this state, the compressor length in the pulse compressor 16 was changed, and the Fourier limit pulse was obtained. Further, in order to compensate for higher-order dispersion, the set value of the third-order dispersion of the optical modulator 13 is set to −7475 fs ³ and the set value of the fourth-order dispersion is set to −2680 fs ⁴ . In this state, if only the second-order dispersion of the optical modulator 13 is changed, it can be chirped linearly.

FIG. 14 is a diagram illustrating the spatial distribution of the electron beam in the dispersion portion when the second-order dispersion setting value of DAZZLER (registered trademark) which is the optical modulator 13 is 8500 fs ² . Similarly, FIG. 15 shows a secondary dispersion setting value of 10500 fs ² , FIG. 16 shows a secondary dispersion setting value of 12500 fs ² , FIG. 17 shows a secondary dispersion setting value of 14500 fs ² , and FIG. 18 shows a secondary dispersion setting. 19 is 15500 fs ² , FIG. 19 is the secondary dispersion setting value 16500 fs ² , FIG. 20 is the secondary dispersion setting value 18500 fs ² , FIG. 21 is the secondary dispersion setting value 20500 fs ² , and FIG. of the second-order dispersion set value when the 22500Fs ^2, shows the spatial distribution in the dispersing section.
Note that the second order dispersion of the optical modulator 13 itself is 13066 fs ² . Therefore, when the secondary dispersion at the Fourier limit is set to 0, the absolute value of the secondary dispersion given by the optical modulator 13 is obtained by adding 13066 fs ² to the set value of the secondary dispersion of the optical modulator 13. It becomes. Thus, second-order dispersion by the optical modulator 13, becomes 21566Fs ² in FIG. 14, becomes 23566Fs ² in FIG. 15, it becomes in Figure 16 25566Fs ^2, becomes 27566Fs ² in FIG. 17, the 28566Fs ² in FIG. 18 In FIG. 19, it becomes 29566 fs ² , in FIG. 20 it becomes 31666 fs ² , in FIG. 21 it becomes 33566 fs ² and in FIG. 33 it becomes 35566 fs ² .

By changing the set value of the secondary dispersion, the pulse shape in the vertical direction can be optimized. For example, when the optical modulator 13 is controlled to set the second-order dispersion setting value to 15500 fs ² or more, the micropulses are connected and integrated as a macropulse. Also, second-order dispersion setting up 12500Fs ² are not each micro pulses connected, at the 22500Fs ^2, micro pulses together is Job's tears like overlap. Thus, the stacking state can be improved by changing the set value of the second-order dispersion of the optical modulator 13.

Next, the measurement result of the Example which shape | molded the pulse shape of the horizontal direction is shown in FIGS. 23 to 29 are images obtained by measuring the spatial profile of pulsed light. Here, a spatial profile was measured using a laser profiler instead of the camera 47. 23 to 29 show image data acquired by the laser profiler. 23 to 29, profile data at the center of the circle is shown on the lower side and the left side. In FIGS. 23 to 29, the circle at the center shows a circle having a diameter of 0.8 mm. 23 to 29 display color image data acquired by actual measurement in gray scale. Below, the adjustment method is demonstrated according to the procedure.
(Step 0)
First, in the pulse stacker 20 shown in FIG. 3, the delay time at each stage was adjusted. Furthermore, as described above, the optimum secondary dispersion was set by the optical modulator 13, and it was confirmed that the energy of the electrons in the bunch were connected evenly. Here, it was determined from the energy spectrum of the electron beam that the micropulses in the macropulse were smoothly connected when the set value of the secondary dispersion of the optical modulator 13 was 14500 fs ² . And it set to this value and started laser space profile shaping. This step 0 is performed again after the following series of steps is completed (after optimization of the laser space profile).

(Step 1)
As an initial condition for shaping, eight micropulses from the pulse stacker 20 were incident on the center of the deformable mirror 40, and were roughly adjusted so that they were at approximately the same position on the profiler equidistant from the photocathode 51. These can be adjusted by controlling the angle of the top / bottom / left / right of the wedge 56 in front of the pulse stacker 20. Actually, the vertical and horizontal angles of the two parallel windows are controlled and adjusted instead of the wedge 56. The spatial profile of the pulsed light at this stage is as shown in FIG.

(Step 2)
Rotation control of the angle of the stacking λ / 2 plate 24 of the masked stack unit is made so that the pulse intensity of the eight micropulses becomes the same by the power meter immediately after the pulse stacker 20 (before the emission side λ / 2 plate 35). did. At this time, the masking was finely adjusted so that the same micropulse intensity was obtained in a pair of (NNS, NNP), (NSN, NPN), and (SNN, PNN). Here, N, S, and P indicate mask states at each stage. That is, N indicates a state in which the mask is not masked, P indicates a state in which the S polarization component is masked and the P polarization component is transmitted, and S indicates a state in which the P polarization component is masked and the P polarization component is transmitted. ing. As described above, the mask state of the three-stage pulse stacker 20 can be specified by writing N, S, and P side by side.

  For example, NNS is a state in which the first and second stack units 31 and 32 have no mask, and the third stack unit 33 masks P-polarized micropulses (passing S-polarized micropulses). Is permitted). NNP is a state in which the first and second stack units 31 and 32 do not have a mask, and the third stack unit 33 masks an S-polarized micropulse (passing through a P-polarized micropulse). Is permitted). The same applies to other mask states.

  Accordingly, when the micropulse intensity of the dual pair of (NNS, NNP) is made equal, the first and second stack units 31 and 32 are not masked, and one of them is masked by the third stack unit 33. . A power meter is disposed in front of the emission side λ / 2 plate 35 to measure the power of each micro pulse. That is, the power of the S polarization component is measured in the NNS mask state, and the power of the P polarization component is measured in the NNP mask state. Then, the respective powers are compared, and the stacking λ / 2 plate 24 of the stack unit 33 is rotated so that the S-polarized component and the P-polarized component have the same power. Thereby, the dual micropulse intensity of (NNS, NNP) can be made equal.

  Similarly, the stack units 31 and 32 in the first stage and the second stage are masked in order so that the S-polarization component and the P-polarization component in each stage have the same intensity. The two plates 24 are rotated. When only the second stack unit 32 is masked, the micropulse intensity becomes equal for the dual of (NSN, NPN). Further, if only the first stack unit 31 is masked, the micropulse intensity becomes equal for the dual of (SNN, PNN). In this way, the optimum rotation angle of the stacking λ / 2 plate 25 was determined for each stage.

  In order to compensate the polarization dependency of the reflectance and transmittance in the laser transport optical system, the emission side λ / 2 plate 35 is set so that the pulse intensities of the eight micropulses are the same on the camera 47 and the like. The rotation was controlled.

  The measurement data in step 2 are shown in FIGS. FIGS. 24 to 27 are those at the time of fine adjustment performed again after the following series of steps is completed. FIG. 24 shows measurement data when the mask state is SNN, and FIG. 25 shows measurement data when the mask state is PNN. 24 and 25 show an example in which the alignment is shifted between the dual micropulses. In FIG. 25, the incident position of the micro pulse is shifted from the center of the circle.

  FIG. 26 shows measurement data when the mask state is NSN, and shows measurement data when the mask state is NPN. In FIG. 26 and FIG. 27, the intensity between dual micropulses is different. The intensity of the micropulse with the mask state of NPN is lower than that of the micropulse with NSN. In such a case, the rotation angle of the stacking λ / 2 plate is adjusted to make the intensity of the dual micropulses the same. After that, fine adjustment is performed so that the profiles are overlapped on the camera 47 between the dual masking.

(Step 3)
All 59 channels of the deformable mirror 40 were applied with the same voltage set. Then, the voltage applied was all 0 to 255 V, and the applied voltage at which the eight micropulses were about the target beam diameter was determined. For example, when aiming for a spot diameter of 0.8 mm on the cathode surface, an applied voltage of 115V was most suitable. The measurement data at the stage of step 3 is shown in FIG. Then, from this state, the cylindrical pulsed light intensity is searched by a genetic algorithm so as to be within a circle having a center diameter of 0.8 mm. From here, it becomes the starting point of the search.

(Step 4)
The initial gene of the genetic algorithm, which is the control algorithm of the deformable mirror 40, was prepared with the same voltage set. Here, as described above, since the gene in the vicinity of all 115V is expected to be excellent, the number of individuals was increased in order to increase the survival rate. The mutation probability was 1%. Attempting with a genetic algorithm in actual 30,000 steps resulted in 281 mutations.
Early genes (100 total)
All 85-99 per set 1 × 15 = 15
All 100-104 2 sets each 2 × 5 = 10
All 105-125 3 sets each 3 × 21 = 63
All 126-135 per set 1 × 10 = 10
All 85-99 2 sets 2 × 1 = 2
By the way, the last two sets of all 115V are replaced with genes (DNA) that are considered to be excellent sets after several attempts of the optimization algorithm. This increases the survival rate of excellent genes.
The genetic algorithm controls to maximize a certain evaluation function. The element of the evaluation function is, for example, H.264. Tomizawa, T .; Asaka, H .; Dewa, H.M. Hanaki, T .; Kobayashi, A .; Mizuno, S.M. Suzuki, T .; Taniuchi, and K. Yanagida, <Status of SPring-8 photocathode RF Gun for Future light Sourcesc, Processeds of the 27th International Free laser, 200 Nine parameters in Table 1 of 138-141 were multiplied by each coefficient to make a linear combination. And each coefficient was set to 50, 200, 170, 200, 300, 0, 200, 0, 10 from the top of the table.

(Step 5)
The deformable mirror 40 was shaped by a genetic algorithm while alternately masking so that the total number of steps was 30,000. Here, masking was performed in the following order.
Overall (NNN) -SSS-Overall-SSP-Overall-SPS-Overall-PSS-Overall-SPP-Overall-PSP-Overall-PPS-Overall-PPP-Overall Adjustment of the entire macro pulse in each step is in 2000 steps went. Therefore, since the whole adjustment was performed nine times in one cycle, the whole adjustment was performed 18000 steps. Each micropulse was adjusted 8 times, 1500 steps each. Therefore, 12000 steps were performed to shape each micropulse in one cycle. In total, 30000 steps were performed in one cycle. The spatial profile in step 5 is as shown in FIG.

  Thus, the measurement of the whole macro pulse and the measurement for each micro pulse were alternately performed, and the pulse shape was adjusted as needed according to the measurement result. That is, after the whole measurement of a plurality of steps was completed, micropulse measurement of a plurality of steps was performed. And the whole measurement and the micro pulse measurement were performed alternately. In the micro pulse measurement, the micro pulses to be masked are changed in order, and all the micro pulses are measured in order. Then, using a genetic algorithm, the deformable mirror 40 was adjusted to optimize the pulse shape.

Thereafter, the precision adjustment in Step 2 was performed. And the misalignment between each micro pulse and the nonuniformity of pulse energy were eliminated. This state is shown in FIGS. The positional deviation of each micropulse on the profiler was within 10 μm, and the difference from the average value of the pulse energy of 8 micropulses was within ± 10%.
Further, Step 0 was performed, and the set value of the secondary dispersion of the optical modulator 13 was set so that the energy spectrum and emittance of the electron beam were minimized. Here, the setting value of the secondary dispersion of the optical modulator 13 is changed to 18500 fs ² , judging from the value of the secondary dispersion of the optical modulator 13. After setting this set value, the two sets of all 115V in Step 4 were replaced with the two optimal genes (DNA) selected in the optimization algorithm trial in the previous 30000 steps. Then, optimization of 30000 steps was performed.

  As for the genetic algorithm, when the initial gene population is prepared, it may be preferable that the genetic algorithm is created with a set in which all the same applied voltage is applied at a square interval in the first step. In this way, the search space can be maximized. This is because when the deformable mirror 30 is an electrostatic actuator system, the deformation displacement amount of the deformable mirror 40 is proportional to the square of the voltage applied to each electrode. This method is therefore used in the very first steps. Note that this method need not be used when the deformable mirror 30 is of the piezo actuator type.

It is a figure which shows typically the structure of the electron gun concerning this invention. It is a figure which shows typically the structure of the light source part used for the electron gun concerning this invention. It is a figure which shows typically the structure of the pulse stacker concerning Embodiment 1 of this invention. It is a timing chart which shows the pulse waveform in a pulse stacker. It is a figure which shows typically the structure of the stacking rod concerning the modification of Embodiment 1 of this invention. It is a figure which shows typically the structure of the pulse stacker concerning Embodiment 2 of this invention. It is a figure which shows the structure of the measurement part which measures the pulse shape in a time direction. It is a figure for demonstrating the signal processing which measures the pulse shape in a time direction. It is a figure for demonstrating the signal processing which measures the pulse shape in a time direction. It is a figure for demonstrating the signal processing which measures the pulse shape in a time direction. It is a figure for demonstrating the signal processing which measures the pulse shape in a time direction. It is a figure for demonstrating the signal processing which measures the pulse shape in a time direction. It is a figure for demonstrating the signal processing which measures the pulse shape in a time direction. In the Example of this invention, it is a figure which shows the energy distribution of the electron beam in a dispersion | distribution part when the secondary dispersion of an optical modulator is changed. In the Example of this invention, it is a figure which shows the energy distribution of the electron beam in a dispersion | distribution part when the secondary dispersion of an optical modulator is changed. In the Example of this invention, it is a figure which shows the energy distribution of the electron beam in a dispersion | distribution part when the secondary dispersion of an optical modulator is changed. In the Example of this invention, it is a figure which shows the energy distribution of the electron beam in a dispersion | distribution part when the secondary dispersion of an optical modulator is changed. In the Example of this invention, it is a figure which shows the energy distribution of the electron beam in a dispersion | distribution part when the secondary dispersion of an optical modulator is changed. In the Example of this invention, it is a figure which shows the energy distribution of the electron beam in a dispersion | distribution part when the secondary dispersion of an optical modulator is changed. In the Example of this invention, it is a figure which shows the energy distribution of the electron beam in a dispersion | distribution part when the secondary dispersion of an optical modulator is changed. In the Example of this invention, it is a figure which shows the energy distribution of the electron beam in a dispersion | distribution part when the secondary dispersion of an optical modulator is changed. In the Example of this invention, it is a figure which shows the energy distribution of the electron beam in a dispersion | distribution part when the secondary dispersion of an optical modulator is changed. In the Example of this invention, it is a figure which shows the spatial profile of pulsed light. In the Example of this invention, it is a figure which shows the spatial profile of pulsed light. In the Example of this invention, it is a figure which shows the spatial profile of pulsed light. In the Example of this invention, it is a figure which shows the spatial profile of pulsed light. In the Example of this invention, it is a figure which shows the spatial profile of pulsed light. In the Example of this invention, it is a figure which shows the spatial profile of pulsed light. In the Example of this invention, it is a figure which shows the spatial profile of pulsed light.

Explanation of symbols

10 light source unit, 11 laser oscillator, 12 pulse stretcher, 13 light modulator,
14 regenerative amplifiers, 15 multipath amplifiers, 16 pulse compressors,
17 wavelength converter,
20 pulse stacker, 21 incident side λ / 4 plate, 22 incident side λ / 2 plate,
23 PBS on the incident side, 24 λ / 2 plate for stack, 25 fixed mirror pair,
26 movable mirror pair, 27 actuator, 28 PBS for branching,
28a-28d PBS for branching, 29 PBS for synthesis, 29a-29d PBS for synthesis,
30 mirrors, 31 stack units, 32 stack units,
33 stack unit, 34 mirror, 35 emission side λ / 2 plate, 37 light shielding member,
40 deformable mirror, 41 mirror, 42 lens, 43 lens, 44 lens,
45 mirror, 46 beam splitter, 47 camera, 48 control device,
51 Photocathode, 52 Resonator, 53 Microwave source, 55 Beam monitor 70 Pulse stacker 71 Pre-stage side PBS, 71a Pre-stage side PBS, 71b Pre-stage side PBS,
74 Rear-side branching beam splitter, 75 Rear-stage combining beam splitter,
76 latter stage side beam splitter, 77 latter stage beam splitter,
78 Post-stage side beam splitter, 79 Post-stage side beam splitter,
81 mirror, 84 front side λ / 2 plate, 85a front side λ / 2 plate,
85b Front stage side λ / 2 plate, 86b to 86d Aperture 90 measurement unit, 91 measurement PBS, 92 mirror, 93 nonlinear optical element,
94 spectrometer, 95 light modulator for probe,
120 stacking rods, 121 wedges, 122 wedges,
123 Birefringent element, 124 λ / 2 wave plate for extraction, 125 PBS for extraction,
126 λ / 2 plate for stack, 131 stack unit, 132 stack unit,
133 stack unit,

Claims (15)

Optical branching means for splitting the pulsed light into two light beams;
Light combining means for combining two light beams branched by the light branching means;
A delay unit that shifts timing by giving an optical path length difference to the pulse light of two light beams incident on the light combining unit from the light branching unit;
Adjusting means for adjusting the pulse shape of the pulsed light from the photosynthesis means;
A light blocking member capable of blocking one light beam incident on the light combining unit between the light branching unit and the light combining unit;
A pulse shaping device comprising: a measuring device that performs measurement according to the other light beam in a state where the one light beam is shielded by the light shielding member in order to adjust the pulse shape by the adjusting means. A λ / 2 plate rotatably provided is disposed in front of the light branching means,
The pulse shaping device according to claim 1, wherein the light combining unit and the light branching unit are polarization beam splitters that branch light according to a polarization state. The measuring device measures the pulse shape of the other light beam;
The pulse shaping device according to claim 1 or 2, wherein the adjustment unit adjusts the pulse shape based on the pulse shape of the other light beam measured by the measuring device.   4. The pulse shaping device according to claim 1, wherein the optical path length difference given by the delay means is variable. The light combining unit is a polarization beam splitter that combines the one light beam and the other light beam, and the polarized light component emitted from the light combining unit is used light, and the non-polarized component leaking from the light combining unit is measured light. age,
The pulse shaping device according to any one of claims 1 to 4, wherein the pulse shape of the utilization light is adjusted based on a measurement result of the pulse shape of the measurement light. A first stack unit that synthesizes two pulsed lights having different timings and emits a first light beam; and
A second stack unit that synthesizes two pieces of pulsed light with different timings and emits a second light beam; and
A rear-stage side light combining means for combining the first and second light beams at a subsequent stage of the first and second stack units;
A profile changing means for changing the profile of the first light beam, which is disposed between the latter-stage photosynthesis means and the stack unit;
Each of the first and second stack units is constituted by the pulse shaping device according to any one of claims 1 to 5 ,
Pulses emitted from the first and second stack units as the first and second light beams by using the two light beams synthesized by the light synthesizing unit in a state where the light shielding member does not shield the light beam. Shaping device. The latter-stage light combining means is a polarization beam splitter that combines the one light beam and the other light beam. The polarized light component emitted from the latter-stage light combining means is used as light and leaks from the latter-stage light combining means. The non-polarized component is used as measurement light.
The pulse shaping device according to claim 6 , wherein the pulse shape of the utilization light is adjusted based on a measurement result of the pulse shape of the measurement light. Means for generating probe light whose wavelength changes with time;
A nonlinear optical element on which the measurement light and the probe light are incident synchronously;
A spectroscope for measuring a spectrum of light emitted from the nonlinear optical element, and
The pulse shaping device according to claim 5 or 7, wherein a pulse shape in a traveling direction of the pulsed light is adjusted based on a measurement result of the spectroscope.   The pulse shaping device according to claim 6, wherein the post-stage side light combining means is a non-polarizing beam splitter. The pulse shaping device according to any one of claims 1 to 9,
An electron gun comprising: a photocathode on which pulsed light shaped by the pulse shaping device is incident. Measure the bunch shape of the electron beam generated at the photocathode,
The electron gun according to claim 10, wherein the pulse shape is adjusted based on a measurement result of the bunch shape. After the electrons in the bunches of the electron beam are spatially dispersed according to energy, the spatial distribution of electrons in the bunches is measured,
The electron gun according to claim 11, wherein a pulse shape in a traveling direction of the pulsed light is adjusted based on a measurement result of the spatial distribution. Optical branching means for splitting the pulsed light into two light beams;
Light combining means for combining two light beams branched by the light branching means;
A pulse shaping method using a pulse shaping device comprising: delay means that shifts the timing by giving an optical path length difference to pulse light of two light beams incident on the light combining means from the light branching means,
Shielding one of the two light beams branched by the light branching means;
Performing measurement according to the other light beam in a state where the one light beam is shielded;
Adjusting a pulse shape based on a measurement result in the step of performing the measurement. Of the two light beams branched by the light branching means, the light beam to be shielded is switched from the one light beam to the other light beam, the other light beam is shielded, and the one light beam A step of not shielding the light,
A step of performing measurement according to one light beam in a state where the other light beam is shielded,
The pulse shaping method according to claim 13, wherein in the step of adjusting the pulse shape, the pulse shape is adjusted based on a measurement result in the step of performing measurement according to the one light beam. Synthesizing two pieces of pulsed light having different timings and emitting a first light beam; and
Synthesizing two pieces of pulsed light that are out of timing and emitting a second light beam; and
Changing the profile of the first light beam;
Synthesizing and emitting the second light beam and the first light beam having the changed profile, and
15. A pulse shaping method, wherein each of the first and second light beams is a light beam including two pulse lights shaped by the pulse shaping method according to claim 13 or 14.