JP6202316B2 - Chirp pulse amplifier - Google Patents

Description

  The present invention relates to a pulse amplifying apparatus that amplifies an optical pulse while controlling chirp, and more particularly to a chirped pulse amplifying apparatus that can amplify even a short pulse light in the femtosecond range with high output.

  In the “chirp pulse amplification method”, a low-power optical pulse from an oscillator is dispersed by a stretcher and the time width is once stretched, and then amplified by an amplifier including a laser gain medium and compressed. This is a method in which the optical pulse is given a dispersion opposite in sign to compress the time width of the optical pulse and returned to obtain a high output optical pulse.

Here, the time width tau [s], the energy E [J], the beam cross-sectional area A peak intensity of the optical pulse of ^{[cm 2] I [W /} cm 2] is, approximately, I = E / (τA) It can be expressed as. In other words, in the chirped pulse amplification method, if the time width τ is stretched with a stretcher, the peak intensity I of the light pulse in the laser medium can be reduced, so that laser medium destruction and pulse waveform distortion due to nonlinear effects are suppressed. It can be done.

  As described in Non-Patent Documents 1 and 2, in the chirp pulse amplification method, in general, a combination of a stretcher and a compressor in which diffraction gratings are arranged in parallel to control dispersion is widely used. Here, in the diffraction grating of the compressor disposed in the subsequent stage of the amplifier, a high breakdown threshold having sufficient resistance against a high-power optical pulse amplified by the amplifier is required. In other words, if the amplification factor of the amplifier is to be increased, a very special and expensive diffraction grating must be used. Further, the output energy extraction efficiency is limited by the diffraction efficiency.

  On the other hand, Patent Document 1 discloses a method called “down chirp pulse amplification method” that does not use a diffraction grating in the compressor. In such an amplification method, an element including a bulk optical block made of a transparent material that imparts substance dispersion is used, and a time pulse is compressed by passing an optical pulse through the element.

  Here, it is also considered that a linear body such as a fiber is used as an element for giving material dispersion. However, in a small-diameter element such as a linear body, for example, as can be seen from the above-described equation of peak intensity I, the beam cross-sectional area A of the input beam must be reduced, and the peak intensity I increases. Therefore, it is strongly influenced by the nonlinear effect. Therefore, it is preferable to suppress the nonlinear effect by using an element including a bulk optical block that can input and pass a large-diameter beam.

US Pat. No. 7,072,101

H.Takada, M.Kakehata, K.Torizuka, “High-repetition-rate 12 fs pulse amplification by a Ti: sapphire regenerative amplifier system,” Optics Letters Vol.31, No.8, pp.1145-1147 (2006) . F.Roser, T.Eidam, J.Rothardt, O.Schmidt, DNSchimpf, J.Limpert, A.Tunnermann, “Millijoule pulse energy high repetition rate femtosecond fiber chirped-pulse amplification system,” Optics Letters Vol.32, No .24, pp.3495-3497 (2007).

  Short pulse light, particularly short pulse light in the femtosecond range, is more strongly affected by nonlinear effects. Therefore, as described above, a chirped pulse amplifying apparatus using an element including a bulk optical block as a compressor is desired. On the other hand, an element including an optical block is larger than an element using a linear body such as a fiber, and in particular, the linear body can be wound with a certain degree of curvature. End up.

  The present invention has been made in view of the above situation, and the object of the present invention is to provide a compact chirped pulse amplifying apparatus that can amplify even a short pulsed light in the femtosecond range with high output. It is to provide.

  The chirped pulse amplifying device according to the present invention is a chirped pulse amplifying device for amplifying the pulsed light from the oscillator through the stretcher, the amplifier and the compressor, extending the time width of the pulse, and compressing and outputting the time width. The compressor includes a bulky optical block made of a material exhibiting material dispersion, and the optical block is provided with at least a plurality of straight-passing paths of the pulsed light.

  According to this invention, it is possible to amplify even a short pulse light in the femtosecond range with a high output while being compact. Moreover, highly efficient compression is possible by using a transparent material with small transmission loss for an optical block.

  In the above-described invention, the straight path may be provided by repeatedly reflecting on the surface of the optical block. According to this invention, it is possible to obtain a more compact chirped pulse amplifying device that can amplify even a short pulse light in the femtosecond range with a high output.

  In the above-described invention, the optical block may be a polygonal column. The optical block may be provided with a control port that changes an incident angle of the pulsed light. According to this invention, the amount of dispersion to be compensated can be changed, and accurate dispersion compensation can be provided even for short-pulse light in the femtosecond range.

  In the above-described invention, the straight path may be provided by repeatedly emitting and injecting the pulsed light from the optical block. The straight path may be provided via an antireflection film provided on the surface of the optical block and a reflecting mirror provided outside the optical block. According to this invention, it is possible to amplify even a short pulse light in the femtosecond range with a high output even though it is simple and compact.

  In the above-described invention, the stretcher includes a bulky optical block made of a material exhibiting a substance dispersion having a sign different from that of the substance dispersion of the compressor, and is provided with at least a plurality of straight passage paths of the pulsed light. It may be characterized by being. According to this invention, a chirped pulse amplifying device that can amplify even a short pulse light in the femtosecond range with high output but has excellent operational stability can be obtained in a compact manner.

It is a block diagram of a chirp pulse amplifier. It is a top view which shows the optical path | route of the chirp pulse amplifier of this invention. It is a graph which shows compensation of the secondary dispersion of the chirp pulse amplifier of the present invention. It is a top view which shows the optical path | route of the compressor of the chirp pulse amplifier of this invention. It is a top view which shows the optical path | route of the compressor of the chirp pulse amplifier of this invention. It is a perspective view which shows the optical path | route of the compressor of the chirp pulse amplifier of this invention. It is a top view which shows the optical path | route of the compressor of FIG. It is a table | surface regarding the N value of a compressor, an incident angle, and the physical-property value of an optical material. It is a top view which shows the optical path | route of the compressor of the chirp pulse amplifier of this invention. It is a figure which shows the incident angle and optical path | route to a compressor in the chirp pulse amplifier of this invention. It is a graph which shows the relationship between the incident angle to a compressor, and optical path length.

  In general, the material dispersion of a bulk optical block (transparent material) is very small compared to the dispersion that a diffraction grating can provide. That is, in the chirped pulse amplifying device, since the dispersion that can be compensated when the optical block is used in the compressor is small, the dispersion of the reverse sign given by the stretcher must be reduced in advance, and the time of the optical pulse that can be stretched The width is small. Therefore, the time width of the optical pulse in the laser medium of the amplifier is also shortened, and the peak intensity I increases when the above formula for obtaining the peak intensity I is referred. In other words, it is strongly influenced by the non-linear effect, easily destroys the amplification medium and deteriorates the waveform of the optical pulse, and becomes a factor that limits the output that can be amplified. Therefore, the inventor of the present application pays attention to the fact that the amount of material dispersion given by the optical block is proportional to the internal optical path length, and considers providing a long optical path inside the optical block to obtain large dispersion. .

  The chirped pulse amplifying apparatus according to one embodiment of the present invention will be described in detail with reference to FIGS.

  As shown in FIGS. 1 and 2, the chirped pulse amplification device 1 includes an oscillator 10, a stretcher 12, an amplifier 14, and a compressor 16 that sequentially process and control an optical pulse.

  The oscillator 10 generates low-power pulse light, and transmits short pulse light, particularly short pulse light in the femtosecond range, to the isolator 21 and transmits the light to the subsequent stretcher 12. For example, as disclosed in non-patent literature, the oscillator 10 is a laser device capable of pulse oscillation, and may use a mode-locked Yb fiber laser that performs pulse oscillation at a center wavelength of 1040 nm. Typically, it is composed of a ring resonator using a Yb-doped fiber having a gain at a center wavelength of 1040 nm, and passive mode locking is performed using nonlinear polarization rotation by three wave plates to generate pulse oscillation. Do. Here, a typical optical pulse repetition frequency of 118 MHz and an output of 80 mW were employed.

  The stretcher 12 provides dispersion to the low-power pulse light guided from the oscillator 10 to stretch the time width of the light pulse. Typically, pulse light is reciprocated between two parallel-arranged diffraction gratings 12a to provide dispersion. Here, a folding mirror 13a is provided on the optical path between the two diffraction gratings 12a.

The number of grooves of the diffraction grating 12a of the stretcher 12 is typically 600 lines / mm, and is arranged in a Littrow arrangement so as to be close to the Littrow angle (incident angle and diffraction angle are equal) that maximizes the diffraction efficiency. ing. One of the diffraction gratings 12a is placed on the movable stage 13b so that the distance between the two diffraction gratings 12a can be adjusted between 220 and 240 mm. By such adjustment, the range of the secondary dispersion value that can be given by the stretcher 12 is variable from −3.5 × 10 ⁵ to −3.8 × 10 ⁵ [fs ² ].

  Further, the stretcher 12 is not limited to the above. For example, a chirped fiber Bragg grating may be used to chirp the grating spacing to disperse light reflected by the Bragg grating. By appropriately setting the direction of the chirp, negative dispersion can be given to the reflected light.

  The amplifier 14 amplifies the optical pulse whose time width is extended in the extender 12 by a laser gain medium to increase the output, and after passing through the telescope optical system 23 composed of a lens pair, adjusts the divergence angle and aperture of the light beam. This is transmitted to the compressor 16.

The amplifier 14 may typically be a Yb fiber laser amplifier. As the gain fiber 14a, a polarization-maintaining double clad fiber can be used. It is preferable to use a photonic crystal fiber having a large core diameter because the nonlinear effect can be further suppressed. For example, the excitation light is supplied by the semiconductor laser 14b, is incident on the excitation cladding of the fiber 14a via the dichroic mirror 14c and the lens 14d, and when entering the core from the excitation cladding, the Yb ions in the core are excited and amplified. It is. The dichroic mirror 14c is designed to transmit the amplification wavelength band 1020 to 1100 nm and reflect the excitation wavelength 976 nm.

  The compressor 16 passes the high-power optical pulse amplified by the amplifier 14 through the optical block (transparent material) 16a to give substance dispersion, and compensates for the dispersion given by the stretcher 12, that is, the time width of the optical pulse. Compress. If dispersion is given also in the oscillator 10 and the amplifier 14, the total dispersion of these dispersions and the dispersion of the stretcher 12 is compensated by the compressor 16, and the entire chirped pulse amplification device 1 is compensated. It is preferable that the residual dispersion is zero.

  The compressor 16 is provided with the total reflection mirrors 17 on both sides of the optical block 16a so as to be spaced apart, and repeatedly reflects the light pulse between them, thereby providing at least a plurality of straight passage paths for the pulsed light in the optical block 16a. The length is lengthened.

By the way, the approximate value of the secondary dispersion per 1 mm length with respect to the wavelength of the pulse laser, that is, the secondary dispersion per unit length is defined as d [fs ² / mm]. When the physical length of the optical path passing through the bulk optical block 16a is L [mm], the secondary dispersion D [fs ² ] given by the optical block 16a is the secondary dispersion d [fs] per unit length. ² / mm] and the physical length L [mm] of the optical path. That is, the amount of dispersion given by the optical block 16a increases in proportion to the optical path length. Therefore, in the compressor 16, at least a plurality of straight traveling paths of pulsed light inside the bulk-shaped optical block 16a are provided, and the optical path length passing through the interior is lengthened.

  Further, by applying the antireflection coating 16b to both end faces of the optical block 16a, it is possible to remarkably suppress the Fresnel reflection loss at the end face of the optical block 16a. The antireflection coating 16b is not essential, but in general, an optical material having a large dispersion has a large refractive index, so that the Fresnel reflection loss tends to increase, and when high-efficiency compression in the compressor 16 is desired. preferable.

As a typical example, the optical block 16a is made of SF57 glass manufactured by Schott Corporation having a length of 25 cm. Since a Yb fiber laser having a wavelength of 1040 nm is used for the oscillator 10 and the amplifier 14, the antireflection coating 16b is an antireflection film for wavelengths of 1000 to 1100nm, and the total reflection mirror 17 is a total reflection dielectric multilayer mirror for the same wavelength range. . In FIG. 2, using the reflection by the total reflection mirror 17, the pulsed light passes through the optical block 16a for seven passes through a straight path. Thereby, it is possible to give a substance dispersion corresponding to glass having a length of 25 × 7 = 175 cm. Since the secondary dispersion value of SF57 glass with respect to light having a wavelength of 1040 nm is approximately d = 150 [fs ² / mm], the secondary dispersion value given by the compressor 16 is D = 2.6 × 10 6. ⁵ [fs ² ]. The secondary dispersion value generated in the oscillator 10 and the amplifier 14 and other optical systems is typically estimated to be 1.1 × 10 ⁵ [fs ² ], and can be given by the above-described stretcher. With the above, it is possible to adjust the total secondary dispersion of the entire amplifying apparatus 1 to zero.

  FIG. 3 shows the measurement result of the time width of the optical pulse output from the compressor 16 with the amplified output of 3W. The distance between the two diffraction gratings 12a of the stretcher 12 is optimized so that the optical pulse width is minimized by adjusting the movable stage 13b. The solid line is a measured autocorrelation waveform, and assuming a correlation factor value of 1.5 for a general waveform, the measured pulse width (full width at half maximum) is 170 [fs]. On the other hand, the dotted line is a Fourier limit pulse (pulse width 150 [fs]) obtained from the Fourier transform of the measurement spectrum. Since the solid line and the dotted line substantially match, it indicates that the compressor 16 can correctly compensate the second-order dispersion and obtain a pulse width close to the Fourier limit.

  In the above, an Er fiber laser mode-locked oscillator having a center wavelength of 1550 nm can also be used as the oscillator 10. In such a wavelength range, the secondary dispersion values of quartz and BK7 are negative. As the stretcher 12, a normal dispersion fiber having positive secondary dispersion in the wavelength 1550 nm band is used, and positive secondary dispersion is applied to the output pulse from the oscillator 10. As the amplifier 14, an Er-Yb fiber laser amplifier using a fiber in which Er ions and Yb ions are co-doped as a gain fiber is used. The pulse output from the stretcher 12 and temporally stretched is passed through the isolator 21 and then amplified by the amplifier 14. The compressor 16 uses quartz or BK7 as the bulk optical block 16a. When the amplified high-power pulsed light is put into the compressor 16, negative secondary dispersion is given by quartz or BK7, and the positive secondary dispersion previously given by the stretcher 12 is compensated, and pulse compression is performed. Can do it.

  4 and 5 show other compressors 36 and 46 for the compressor 16 described above.

In FIG. 4, total reflection coating 37 made of a dielectric multilayer film or a metal film is applied to both end faces of the optical block 36a. Further, an input port 38 and an output port 39 were provided on a part of both end faces of the optical block 36a, and an antireflection coating was provided on the part. That is, the pulsed light that has once entered the optical block 36a from the input port 38 is emitted from the output port 39 after being subjected to multiple reflection at the portion of the optical block 36a without being emitted outside. Thereby, at least a plurality of straight traveling paths of the pulsed light inside the bulk-shaped optical block 36a can be provided. In FIG. 4, pulsed light passes through the optical block 36 a through a straight path through reflection by the total reflection coating 37, but as described later, by changing the incident angle to the optical block 36 a, The number of round trips of the passage route can be adjusted.

In FIG. 5, total reflection mirrors 47 are arranged on both sides of the optical block 46a so as to be spaced apart from each other. However, the antireflection coating is not provided on both end surfaces of the optical block 46a, and pulse light is incident at a Brewster angle. . That is, for p-polarized light incident at an incident angle θ _B ,
tan θ _B = n ₂ / n ₁
It is known that the reflectance becomes zero when the light is incident at the Brewster angle. Here, n ₁ and n ₂ are the refractive indexes of the medium on the incident side and the emission side, respectively.

For example, if the light from the in outside air into the inside of the optical block 46a is incident, and, by setting the angle in the case where light from the optical block 46a into the air is emitted so as to satisfy the equation of the incident angle theta _B, If the light beam is p-polarized light, reflection loss can be remarkably suppressed on the incident surface and the exit surface. Specifically, when SF57 glass is used, since the refractive index with respect to a wavelength of 1040 nm is 1.81, if the optical block 46a is arranged so that the incident angle on the air side is 61.1 degrees, the reflection loss is reduced. it can. In addition, if both end faces of the optical block 46a are made parallel, the exit angle when exiting from the optical block 46a is also the Brewster angle. That is, reflection loss can be sufficiently suppressed without requiring an antireflection coating.

  As described above, according to the present embodiment, as a high-power short pulse laser device, it is effective for precision processing that requires less thermal influence and medical surgery that requires less damage to the affected part. is there. In addition, since the configuration is relatively simple, a high-power short-pulse laser device can be provided at a low price.

  Subsequently, an embodiment using another compressor 26 for the above-described compressor 16 will be described below. The oscillator 10, the stretcher 12, and the amplifier 14 constituting the chirped pulse amplification device 1 are the same as those described above unless otherwise specified, and are not limited.

  As shown in FIG. 6, the compressor 26 confines the pulsed light in the optical block 26a using total internal reflection, lengthens the optical path length by providing at least a plurality of straight-passing paths, Compensates for second order dispersion. Since there is no pulsed light that is refracted from the optical block 26a and emitted to the outside, loss of pulse energy can be suppressed. In the figure, the compressor 26 including the regular hexagonal prism optical block 26a is shown. However, it may be a columnar body having a bottom surface of a regular N-gon (N is an integer of 3 or more). Although a glass material is used here, various optical materials can be used as appropriate.

  The light incident location P1 near the upper portion of the optical block 26a is provided with a convex input port 27 to the optical block 26a, or the optical block 26a is cut and a concave portion is cut as shown in the lower right circle of FIG. An input port 27 'is provided to provide an input surface 27a where the input light is directly incident. In addition, it is preferable to appropriately apply an antireflection coating on the input surface 27a because Fresnel reflection loss can be prevented. Alternatively, the light may be incident at the above Brewster angle to prevent reflection of p-polarized light. An output port 28 is provided at the light exit point P2 near the lower portion of the optical block 26a, which is formed in the same manner as the input port 27.

  Referring also to FIG. 7 (a), when the optical block 26a is viewed from above in the vertical direction, the pulsed light input from the incident point P1 at the midpoint P of the side AB is reflected at the point Q and further adjacent. In the same manner, reflection is repeated at the midpoint of each side, and the inside of the optical block 26a of a regular N prism is circulated. If the input light is input from the light incident point P1 by being slightly tilted upward or downward in the vertical direction from the horizontal direction, the light will be slightly above or below the input point P in the vertical direction after making one round inside the regular N-gon. Come back to the point located at. FIG. 7 (a) shows a case in which it is tilted downward. That is, when the output port 28 is provided at a point where the pulsed light spirally circulates inside the optical block 26a and reaches the inside of the regular N-gonal circle several times, the pulsed light is emitted from the optical block 26a to the outside. Can be output.

  Here, the incident angle θ at the time of reflection at the midpoint of the side of the N-prism optical block 26a is represented by θ = 90 degrees−180 degrees / N. FIG. 8A shows incident angle values up to N = 9. That is, referring to the value of the critical angle in FIG. 8B, for example, when light having a wavelength of 1040 nm is used, in order to satisfy the total reflection condition that the incident angle is equal to or greater than the critical angle, SF57, SF6 In the four types of glass bodies, quartz and BK7, N ≧ 4, that is, a regular N prism more than a regular quadrangle. FIG. 7B shows an example in which the optical path is calculated for light with a wavelength of 1040 nm in SF57.

  FIG. 9 (a) shows the result of calculation similar to FIG. 7 (b) using an unequal side hexagonal column in which the length of the side is changed by about 5% from the regular hexagonal column optical block 26a. It was. By making the polygonal side of the cross section of the optical block 26a an unequal side, the position of the optical path from the light incident location P1 slightly changes with each turn. Here, after 6 rounds, the light reaches a surface that does not satisfy the total reflection condition, and a light exit point P2 is given. According to this, light can be emitted from the optical block 26a without requiring processing of a special shape such as the output port 28 (see FIGS. 6 and 7) or the like to the light output portion P2.

  In FIG. 9B, the incidence of the pulsed light on the unequal hexagonal column is performed by the right angle incident surface. In such a case, it is not necessary to apply a specially shaped process like the input port 27 (see FIGS. 6 and 7) to the light incident place P1. By making the polygonal side of the cross section of the optical block 26a an unequal side, the straight path (light path position) slightly changes for each turn, and reaches a surface that does not satisfy the total reflection condition after two and a half laps and emits light. Location P2 is given. According to this, not only the input port 27 but also the output port 28 need not be processed with a special shape.

  As described above, the optical block 26a is not necessarily a regular polygonal prism, and need not be input from the midpoint of each side as long as the total reflection condition is satisfied inside the regular polygonal prism.

  FIG. 10 shows the optical path when the incident angle is changed at the light incident point P1 at the midpoint of the side of the regular hexagonal prism optical block 26a. FIG. 10A and FIG. 10B are optical path calculation results when the incident angle is changed to 59 degrees and 61 degrees, respectively. The optical path length changes continuously corresponding to the incident angle.

FIG. 11 shows the incident angle dependence of the physical distance of the optical path. The value of the physical distance of the optical path is expressed as a numerical value (ratio) with respect to the length of the side of the regular hexagonal optical block 26a as 100. As can be seen, the amount of dispersion given by the compressor 26 can be continuously changed by providing a control port at the light incident location P1 and slightly changing the incident angle. With this dispersion amount adjusting mechanism, the total residual dispersion amount of the chirped pulse amplification device 1 can be accurately zeroed.

  By the way, in the above-described compressor 16 (26, 36, 46), at least a plurality of straight passage paths are provided by using an optical block made of an optical material that gives dispersion opposite to that of the optical block 16a (26a, 36a, 46a). Thus, it is possible to provide the stretcher 12 having a longer optical path length. Such a stretcher has advantages similar to the compressor 16 (26, 36, 46) described above, and can provide, for example, a dispersion amount adjusting mechanism. In other words, the chirped pulse amplification device using this can accurately reduce the total residual dispersion amount to zero, and can be amplified with high output even with a short pulsed light in the femtosecond range, but has excellent operational stability. And can.

  So far, representative examples and modified examples based on the examples have been described, but the present invention is not necessarily limited thereto. Those skilled in the art will recognize a variety of alternative embodiments without departing from the scope of the appended claims.

DESCRIPTION OF SYMBOLS 1 Chirp pulse amplifier 10 Oscillator 12 Stretcher 12a Diffraction grating 14 Amplifier 14a Gain fiber 14b Semiconductor laser 14c Dichroic mirror 14d Lens 16, 26, 36, 46 Compressor 16a, 26a, 36a, 46a Optical block 23 Telescope optical system

Claims (5)

A chirped pulse amplifying device that sequentially outputs pulse light from an oscillator through a stretcher, an amplifier, and a compressor, extends the time width of the pulse, amplifies the compressed pulse, and outputs the compressed time width;
The compressor comprises a bulk optical block of Do Ri regular polygonal pillar of a material exhibiting material dispersion, the the optical block was repeated reflections at least a plurality give Rutotomoni surface straight passing path of the pulsed light A pair of input ports and output ports are provided that circulate and provide a circular optical path, and the input port changes the physical distance of the circular optical path by changing the incident angle of the pulsed light to continuously change the amount of dispersion. by output from the output port by said stretcher and chirped pulse amplification system, characterized that you compensate for residual dispersion given in the amplifier. 2. The chirped pulse amplification device according to claim 1, wherein the optical block is a regular hexagonal prism. The chirped pulse amplification device according to claim 1 or 2, wherein reflection on the surface is repeated to make a spiral. 4. The chirped pulse amplification device according to claim 3, wherein the amplifier is made of a photonic crystal fiber. The stretcher includes a bulky optical block made of a material exhibiting a material dispersion having a sign different from that of the material dispersion of the compressor, and is provided with at least a plurality of straightly passing paths of the pulsed light. A chirped pulse amplification device according to any one of claims 1 to 4 .

Images