JP2016509698A - Method and apparatus for fiber delivery of a high power laser beam - Google Patents

Abstract

In various embodiments, the optical fiber (11) includes a core having a relatively large area selected to increase the threshold for stimulated Raman scattering or stimulated Brillouin scattering, or both, the core having a high aspect ratio. And has a first refractive index. The core is narrow in the fast axis direction and wide in the slow axis direction. As a result, the fiber is mechanically flexible in the fast axis direction and mechanically rigid in the slow axis direction.

Description

  The present disclosure relates to a fiber supply apparatus and method for a high-power laser beam.

  The optical fiber may be used to transmit a laser beam from a laser light source to a desired position. The use of optical fibers to transmit a laser beam from a laser source to a desired location is an important factor for success in many laser applications. This is because the optical fiber does not include free-space optical communication and provides a flexible transmission path that allows the optical path to be reset in real time. For example, industrial lasers operating at wavelengths that can be transmitted through silica fibers are almost universally using such optical fibers to transmit a laser beam from a laser source to a desired location. Usually, the output end of the optical fiber is attached to a robot arm that guides the output beam to the workpiece. Another example is medical applications. In medical applications, optical fibers are used to transmit a laser beam through a blood vessel to a specific location. At the specific location, a focused beam is used to achieve the desired medical effect. Another example is found in the design of solid state lasers. In solid state laser designs, radiation from the pump diode is typically delivered to the active region of the solid state laser via a fiber or bundle of fibers. Yet another example is a defense system. In defense systems, it is desirable to guide the laser beam to the inner gimbal of the beam director without having to propagate through the optical elements that make up the complex sequence of moving and guiding the laser beam.

  Although the first three examples listed above have already been implemented in their respective application areas, the use of optical fibers to deliver laser beams in defense systems has so far been limited by two challenges. First, the optical fiber in the conventional system has the required average or peak without suffering from laser beam degradation due to optical damage or nonlinear optical processes such as stimulated Brillouin scattering (SBS) or stimulated Raman scattering (SRS). It is required to have a core area large enough to be adaptable to the output. Second, if a sufficiently large core is developed, the fiber must be designed so that the input beam quality—which must be nearly diffraction limited for many military applications—is not degraded by propagation through the fiber. .

  Optical fibers are widely used for supplying laser beams in the industrial laser market. These optical fibers have the function of supplying a very high power laser beam of interest. However, these conventional optical fibers are designed to transport the radiation of high multimode industrial lasers. These conventional optical fibers inevitably have a core diameter of 0.5-1 mm and a large numerical aperture (NA) of 0.1-0.2 for high multimode. At present, there is no practical way to maintain high beam quality when using conventional optical fibers for beam delivery. Due to the multimode aspects of these conventional optical fibers, strong mode coupling to higher order transverse electromagnetic modes guided along with the fundamental transverse electromagnetic mode and other lowest order transverse electromagnetic modes by bending the optical fiber Happens. If the lowest order transverse electromagnetic mode is emitted in these conventional optical fibers, the lowest order mode loses most of its output when providing a higher order transverse electromagnetic mode. As a result, the quality of the output beam is generally better than 50-100 times the diffraction limit (XDL), even if the input beam is nearly diffraction limited and the output beam suffers only a small output loss. growing. Thus, these conventional optical fibers meet the requirements for power supply, but do not meet the requirements for beam quality supply.

  Another problem with typical beam delivery fibers for high power lasers is that the fiber becomes inflexible as it grows to accommodate higher laser power with increasing core diameter. . In some respects, the basic purpose of the fiber—mechanical flexibility—becomes severely limited as the fiber diameter increases.

  To overcome the above disadvantages of high quality beam transport, large mode area (LMA) optical fiber designs may be implemented. LMA optical fibers can guide a few higher order transverse electromagnetic modes while maintaining beam quality above about 1.3 times the diffraction limit (XDL). An LMA optical fiber is a standard large core feed fiber in that it has a relatively small core diameter of about 20 μm to about 30 μm for a signal wavelength of about 1 μm and a reduced NA of about 0.06. Different. LMA fiber must be properly coiled in order to maintain good beam quality. For a specified core diameter, configuring the coil induces bending losses to all transverse electromagnetic modes, but all higher order modes constitute a much larger coil than the desired lowest order transverse mode. Suffers from the loss induced by this. Therefore, by configuring the coil, the radiation loss of the higher order transverse electromagnetic mode is increased, and the higher order transverse electromagnetic mode is removed out of the core. As a result, a laser beam is “cleaned up” and a beam having an almost fundamental transverse electromagnetic mode is obtained. With optimized coil radius, the fundamental transverse electromagnetic mode loss remains at a low and acceptable level.

  A 10 kW fiber laser that generates a laser beam supplied to a beam director used in industrial applications is usually based on LMA fiber. Conventional LMA beam delivery fibers may be considered to meet beam delivery requirements. The LMA fiber can obviously be adapted to the laser power, especially since the laser beam is initially generated by the LMA fiber. However, there are other items that need to be considered. In particular, stimulated Raman scattering (SRS) or stimulated Brillouin scattering (SBS) may begin when the beam delivery fiber length reaches several meters, as required in typical applications.

US Pat. No. 5,208,699 US Pat. No. 7,042,631 US Pat. No. 7,860,360 U.S. Pat. No. 7,983,312 US patent application Ser. No. 12 / 358,983

  What is needed is to provide an optical fiber that solves the above-mentioned deficiencies of conventional optical fibers used to provide high power laser beams and takes into account the above considerations-including SRS and SBS- It is.

  One embodiment of the present disclosure provides an optical fiber that includes a core having a relatively large area selected to increase the threshold for stimulated Raman scattering or stimulated Brillouin scattering, or both. The core has a cross section extending at a high aspect ratio and a first refractive index. The core is narrow in the fast axis direction and wide in the slow axis direction. As a result, the fiber is mechanically flexible in the fast axis direction and mechanically rigid in the slow axis direction.

  Another embodiment of the present disclosure includes a laser, an optical fiber having an input end and an output end, and an optical coupler configured to input radiation from the laser through the input end of the optical fiber. An optical device is provided. The optical fiber includes a core having a relatively large area selected to increase the threshold for stimulated Raman scattering or stimulated Brillouin scattering, or both. The core has a cross section extending at a high aspect ratio and a first refractive index. The core is narrow in the fast axis direction and wide in the slow axis direction. As a result, the fiber is mechanically flexible in the fast axis direction and mechanically rigid in the slow axis direction.

  Another embodiment of the present disclosure provides an optical device having an optical fiber having an input end and an output end, and an end cap connected to the output end of the optical fiber. The optical fiber includes a core having a relatively large area selected to increase the threshold for stimulated Raman scattering or stimulated Brillouin scattering, or both. The core has a cross section extending at a high aspect ratio and a first refractive index. The core is narrow in the fast axis direction and wide in the slow axis direction. As a result, the fiber is mechanically flexible in the fast axis direction and mechanically rigid in the slow axis direction.

  Another embodiment of the present disclosure provides a fiber delivery method for a high power laser beam. The method includes providing an optical fiber that includes a core having a relatively large area selected to increase a threshold for stimulated Raman scattering or stimulated Brillouin scattering, or both, wherein the core has a high aspect ratio. The fiber has a cross section that extends and has a first refractive index, and the core is narrow in the fast axis direction and wide in the slow axis direction, so that the fiber is mechanically mechanical in the fast axis direction. A stage that is flexible and mechanically rigid in the slow axis direction. The method further comprises emitting a laser beam into the fiber, the laser beam having a width that is narrower than the width of the core along the slow axis direction and the fast axis direction; and Propagating the laser beam through the fiber while avoiding the generation of stimulated Raman scattering or stimulated Brillouin scattering or both.

  Yet another embodiment provides an optical fiber comprising a core having a relatively large area selected to increase the threshold of stimulated Raman scattering or stimulated Brillouin scattering or both, wherein the core is high A method for providing a fiber of a high power laser beam having a step having a cross section extending at an aspect ratio and having a first refractive index is provided. The fiber core is narrow in the fast axis direction and wide in the slow axis direction. As a result, the fiber is mechanically flexible in the fast axis direction and mechanically rigid in the slow axis direction. The method further includes emitting a laser beam into the fiber, the laser beam having a width that is narrower than the width of the core along the slow axis direction and the fast axis direction. The method further propagates the laser beam through the fiber to mitigate the occurrence of stimulated Raman scattering or stimulated Brillouin scattering, or both, while taking any higher order transverse greater than the loss of the low order transverse electromagnetic mode. Substantially eliminating the higher order transverse modes along the slow axis in the fiber core by providing electromagnetic mode losses.

  The above and other features and characteristics of the present disclosure, as well as the method of operation, the function of the relevant elements of the structure, the combination of parts, and the efficiency of manufacture are described in the following detailed description and claims with reference to the accompanying drawings It becomes clearer by examining. All of the drawings constitute part of the present specification. In the drawings, like reference numerals indicate corresponding parts in the various views. In one embodiment, the structure depicted herein is not drawn to scale. However, it should be noted that the figures are for illustration and explanation purposes only and are not intended to limit the basic concepts of the present invention.

Why the core and NA of the annular core feed fiber must be specially sized for the power and beam quality of the high energy laser (HEL) beam, otherwise the system suffers from loss of efficiency or beam quality It is a schematic diagram showing what will happen. Schematic showing how a high aspect ratio core (HARC) fiber provides additional design freedom, resulting in a group of optimization curves depending on the specified core aspect ratio. is there. FIG. 2 is a schematic diagram of a HARC fiber following a curved path that may be required in a beam delivery application according to an embodiment of the present disclosure. 1 is a schematic cross-sectional view of a HARC fiber according to an embodiment of the present disclosure. 1 is a schematic diagram of a cross-section of a semi-guiding high aspect ratio core (SHARC) fiber according to an embodiment of the present disclosure. FIG. 1 is a schematic diagram of an optical coupler that couples the output of a laser to a HARC or SHARC fiber according to one embodiment. FIG. FIG. 4 illustrates in gray scale the results of a 3D propagation simulation of a laser beam through an optical coupler and the HARC fiber illustrated in FIG. 3 according to one embodiment. FIG. 2 is a schematic view of an output end cap provided at the end of a semi-waveguide fiber illustrated in FIG. 1 according to one embodiment. 2 is a theoretical model representing the focal length induced by torsion as a function of torsion angle and a plot of focal length induced by measured torsion as a function of torsion angle. 1 is a schematic diagram of an optical fiber module according to an embodiment of the present disclosure. FIG. It is a one part perspective view of the optical fiber module by another Example. FIG. 11a is a schematic diagram of an embodiment of the Laurin group method using a fiber structure. FIG. 11b is a schematic diagram of an example of the Laurin group method using a fiber structure. FIG. 11c is a schematic diagram of an embodiment of the Laurin group method using a fiber structure. FIG. 11a is a side view of a portion of the fiber structure used in the Laurin group method of FIGS. 11a-11c. 12b is a cross-sectional view of a portion of the fiber structure shown in FIG. 12a. 1 is a schematic diagram of a hollow core waveguide with 90 ° twist that can be used as a twist module to interconnect two untwisted glass HARC / SHARC fibers according to an embodiment of the present disclosure. FIG.

  Defense-related applications typically can tolerate the output of a single high power (eg, 10 kW or higher) solid or fiber laser with beam quality better than about 1.3 times the diffraction limit (XDL), and Benefits are obtained from the availability of beam supply means that can supply the laser beam to a remote location. In such defense applications, the beam delivery fiber must be able to adapt to such high power levels while maintaining the same beam quality as the original laser. In addition to the above-described aspects, such defense applications include the ability of the beam delivery fiber to carry a high power laser beam to the inner gimbal of the beam director and to the expected motion of the gimbal assembly in the anticipated area scenario. It is stated that it is flexible and has a maximum length of about 10 m or more so that it has an appropriate length to fit.

  At a laser wavelength of 1 μm and a spectral width wider than about 0.01 nm, one of the fundamental limitations of conventional fibers in this type of application is to operate without generating SRS. For example, in a typical application using a 10 kW fiber laser, the output is sufficiently large so that the length of a conventional beam delivery fiber is limited to less than about 2 m. However, in certain defense applications, this limited length is inappropriate and constrained.

  In general, SRS degrades efficiency and possibly the beam quality of the transmitted laser beam. In particular, SRS generates a counter-propagating beam, which can cause the majority of the initial signal strength to return to the laser rather than the intended target. At the same time, SRS widens the frequency band of signal strength that continues to propagate in the forward direction. Such spread is unacceptable for many applications. SRS may also generate a forward propagating beam that generally includes not only higher order modes but also fundamental modes. This process degrades the beam quality of the transmitted light.

For continuous wave (CW) radiation or pulses longer than about 100 ps, the conditions under which SRS occurs in the optical fiber are expressed as:

g (2P / A) L <25 (1)

Here, g is the Raman gain depending on the medium, P is the signal intensity, A is the area of the core, and L is the length of the fiber. Factor 2 on the left side of Equation (1) is an approximation that the effective area of the propagating fundamental optical transverse electromagnetic mode is approximately half the actual core area.

For example, in fused silica where g = 0.01 cm / GW, at P = 10 kW, the relationship between length L and core area A can be determined as follows.

L <(25A / 2gP) = 1.25A (2)

Here, the length at which SRS occurs is L (unit: cm), and A is the core area (unit: μm ² ) of the fiber.

Some defense applications further specify low power beam fiber delivery as part of a coherent laser detection and ranging (LADAR) system. Even if the nominal power of this application is as low as 1 kW, the spectral width is very narrow below 10 kHz. In this case, the limiting nonlinear process is stimulated Brillouin scattering (SBS). Equation (1) also applies to SBS. However, with SBS, the gain g of fused silica increases at 5 cm / GW. For example, for a signal strength of 1 kW, the length limit is determined using equation (3).

L _B <0.025A (3)

Here, _{L B} is the length of occurrence of SBS (unit cm), and, A is the core area of the fiber (in [mu] m ^2).

For a conventional LMA fiber having a relatively large core diameter of about 30 μm, the core area is about 700 μm ² . If the core area of the LMA fiber of about 700 μm ² is used, the limit length L can be calculated to be about 8.4 m using Equation (2). However, note that this length corresponds to the limit length before the SRS threshold is reached. The safe length value may be half of this length to provide a double margin for the occurrence of SRS. Moreover, if the HEL is a fiber laser, the calculated length limit must include the fiber laser itself, assuming that both fibers are made of silica. For example, at a power of 10 kW, the fiber laser itself uses up most of the acceptable length (even when the fiber laser core is longer than 30 μm). As a result, the tolerance for further length of the beam delivery fiber is minimal. Therefore, the conventional LMA fiber cannot meet the requirement of a beam supply length of about 10 m without causing SRS.

  The fiber has been changed to a larger area than is available with conventional LMA fibers. In principle, such large area fibers can be changed to longer lengths without causing SRS. However, changing the diameter of a single mode or number mode fiber core must be realized by further reducing the numerical aperture (NA) of the core to less than about 0.03. This reduction in NA is driven by the need to limit the number of guided transverse modes to a relatively small number. Otherwise, the mode identification mechanism used to maintain single mode operation at high power will not work. Specifically, as the size of the core increases for a given NA, the effective mode exponents in all core modes are very closely spaced. As a result, the higher order mode is almost degenerated from the lowest order mode. As a result, the mode mixing induced by bending between the lowest order mode and many higher order modes becomes very strong, and the output is switched from the lowest order mode to the higher order mode. To avoid this unintended result, it is important to avoid even slight bends and deformations in large core fibers. Another problem with such fibers is that the reduced NA causes higher order modes to leak out and it is no longer possible to flex the fiber without inducing unacceptable transmission losses. Increasing the core makes the fiber more susceptible to refractive index non-uniformity. As a result, the difficulty of maintaining the superior mode quality desired from state-of-the-art lasers is increased.

  By designing a fiber that does not bend, the bending sensitivity of a low NA fiber with a large core can be avoided. In fact, “rod-like” fiber lasers with core diameters up to 100 μm and very low NA achieved impressive power scaling. However, in these fibers, the core is embedded in a quartz rod with a diameter of about 2 mm, which makes such fibers inflexible. The conclusion is that the size of the core area of conventional fibers cannot be realistically changed while maintaining the flexibility required for beam delivery applications.

  Until now, only the problem related to increasing the core area of the fiber has been considered in that high power operation can be realized without causing SRS or SBS. However, even if a solution to suppress such stimulated scattering is found, the large core annular beam delivery fiber also suffers from different fundamental limitations that are best understood by referring to FIG. FIG. 1 shows a graph of core area (for a given NA) of a conventional annular core fiber versus beam quality (XDL). The graph in this figure schematically represents the core area (for a given NA) required to fit a laser beam as a function of the beam quality of the laser beam. Since the beam quality is generally expressed in one dimension rather than in two dimensions (ie, how much the beam diameter exceeds the diffraction limit), the required core area is roughly relative to the square of the beam quality. Change. The parabola in FIG. 1 shows the change in core area as a function of beam quality.

  In the upper left of the graph of FIG. 1, the situation where the actual core area of the beam delivery fiber exceeds the core area predetermined by the parabola is illustrated. Thus, the input beam does not fill the core, as shown schematically in FIG. For example, assuming that the fiber dynamically deflects as the beam director moves to track the target until the beam propagates through the designated 10 m fiber, the output beam is shown schematically in FIG. As shown, the entire core area and the total NA are satisfied. Correspondingly, the overall core area and total NA both exceed the original beam area and divergence. The result is beam quality degradation. The emitted beam can fill the core area but not the NA. But keep in mind that this does not solve the problem. The NA is inevitably filled even if only a portion of the NA can be filled by the emitted beam. Therefore, mode scrambling is induced by fiber deflection that inevitably occurs due to the motion of the beam director. The result of a high NA in this way is a degradation of the transmitted beam quality.

  In the lower right of the graph of FIG. 1, a situation is shown where the core area is smaller than the size of the beam cross-sectional area (eg, the beam diameter is greater than the fiber diameter). Specifically, the input beam overfills the supply fiber core. The output beam “out” fills the core and the power emitted to the cladding is lost. Therefore, significant efficiency loss occurs when using small size fibers. Similar losses occur when the input beam size is reduced to fit within the fiber. As a result, the divergence exceeds NA at the fiber input.

  The only satisfactory solution in this example is to operate strictly on the parabolic curve of FIG. As the quality of the input beam improves, the area of the annular fiber core must increase proportionally. The reverse is also true. This raises the obvious problem that there is no annular fiber that can accommodate the radiation of the high power laser system currently being developed by Raytheon. These lasers have good beam quality of less than 2XDL. Thus, beam quality can be maintained outside the annular beam delivery fiber only if the fiber is selected from the small core area available at the bottom of the parabola. Instead, however, at high power levels above 100 kW, a very large core area is required to suppress loss of efficiency with negative consequences including optical damage and non-linear reduction—eg, SRS—.

  As a general rule, specify a high multimode feed fiber with a core area large enough to accommodate high laser power without encountering stimulated scattering or other intensity dependent processes, and purify the output beam to achieve the desired good beam A method of restoring quality can be used. For example, this can be achieved using the phase conjugate architecture described in US Pat. In this case, the laser architecture is a main oscillation output amplifier (MOPA) in which a beam delivery fiber and fiber amplifier are arranged to pass twice by reflection from a phase conjugate mirror (PCM) between paths from a simple fiber laser. Must be corrected. Although phase conjugate compensation for distortion of passive multimode fibers having a core diameter of about 400 μm has been shown in the past, application to large core fibers is believed to be problematic due to limitations of the PCM compensation function. In addition, the complexity of the phase conjugate architecture adds cost and risk.

  Without an appropriate fiber-based method, one can choose from two main fiber-independent methods that can be used to deliver a signal beam from a high power laser to the beam director turret. The first method is to combine the laser beam so that it enters a free space away from the beam director and direct the beam to the turret (provided inside the turret device, which is commonly used in large optical microscopes). In this method, an optical path is set so as to pass through the turret by using an elbow-shaped curved path (including a series of optical systems). Another method is to shorten the distance from the output fiber of the laser head to the beam director turret by providing the laser head itself on the beam director device. In this case, a turret that needs to be able to move quickly to direct the beam at the target must carry the additional weight of the laser device. This is considered serious. In the first method, turret design is complicated by elbow-shaped paths. The optics in the elbow-shaped path is sufficient to fit the high energy laser beam without being damaged or heated to such an extent that the high energy laser beam is optically distorted by various mechanisms. It needs to be a proper size. The optical system must also maintain alignment as the optical path changes over time and environment. This requirement adds to the cost and complexity of the system.

  A dramatic new innovation from all of the methods described above is to continue to use the waveguide, but to change the shape of the core. This method begins by recognizing the basic conclusion of FIG. 1 that using an annular core fiber is not possible with a high power, near diffraction limited laser beam. However, this new method also recognizes that the situation is quite different for high aspect ratio (HARC) fibers, as shown schematically in FIG. FIG. 2 shows a graph representing the area of the fiber core as a function of the beam quality (XDL) of the laser. HARC fiber has a rectangular cross section. A rectangular core has a narrow “fast axis” direction, which is one of the core dimensions. In one embodiment, the fiber core may be narrow enough to guide only a single fast axis mode or possibly a few low order fast axis modes in the fast axis direction. The length of the fast axis cladding may also be narrow enough to provide the mechanical flexibility of the fiber in this direction. In this way, the fiber can be bent in the fast axial direction. Since the fiber can be single mode at this length, the fiber can be bent without distorting the beam quality. On the other hand, the cross-sectional area of the core can be very large by extending the length of the core along the length of the orthogonal “slow axis”. In particular, the core area may be large enough that very high laser power can be emitted without exceeding the SRS threshold.

  Some of the features of HARC can be evaluated by comparing the graph of FIG. 1 with the graph of FIG. First, note that there is a significant change in the scale of the vertical axis in the plot of FIG. 2 compared to the plot of FIG. It can be seen that the HARC fiber provides a total of two degrees of freedom to specify the core. The two degrees of freedom are area and aspect ratio. This makes it possible to generate a group of curves each representing the aspect ratio of a given core (the length of the core in the slow axis divided by the length of the core in the fast axis). . A 1: 1 aspect ratio suffers from the same limitations as an annular core fiber. However, with higher aspect ratios, the fast axis length is adapted to the beam quality of the input beam (whether or not diffraction limited), and the core can also cause damage, exceeding the SRS threshold. It is possible to simultaneously determine the length of the slow axis so that it can be adapted to the overall laser power without any problems. Regardless of what beam quality and power level is required, the HARC fiber may be identified with an appropriate core aspect ratio.

  The HARC method has been proposed in US Pat. No. 6,057,836 to increase the resistance to laser damage resulting from high energy short pulses in a hollow core beam delivery waveguide. The self-imaging property of the rectangular core is used to image the output surface of the waveguide to the input surface by transporting the beam through the rectangular core waveguide. However, this method is extremely sensitive not only to the deformation of the core shape of the waveguide but also to the deflection of the supply path. Self-imaging is based on the tracking of the relative phase of all modes with a very strict tolerance level of less than 1 radian. Minor imperfections in the core shape during waveguide fabrication, propagation distance deviation from self-imaging distance, and simple waveguide deflection can all destroy the relative phase between modes. There is. As a result, the self-imaging phenomenon and the related advantages cannot be obtained. There is no practical way to transport a coherent beam using this method without destroying the desired self-imaging.

  Yet another HARC method according to one embodiment may be implemented without the limitations imposed by relying on self-imaging. This alternative utilizes a fiber that can be made from glass or other transparent material with a similarly high aspect ratio core embedded in a high aspect ratio cladding. FIG. 3 is a schematic diagram of a HARC fiber 11 following a path with a curvature that may be required in beam delivery applications. Note that because the HARC fiber 11 has a geometric shape like a ribbon, the HARC fiber 11 can only bend with a thin length. This is shown schematically in FIG. FIG. 4a is a schematic illustration of a transverse section 4 of a fiber 11 according to one embodiment. The fiber 11 has a core 42 and a clad 44. The coating 46 covers the cladding 44. In one example, the coating 46 may be a polymer coating or any other flexible coating. The HARC fiber 11 has a rectangular cross section as shown in FIG. 4a. In one embodiment, the rectangular core 42 of the HARC fiber 11 generally has an aspect ratio that can range from about 30: 1 to 100: 1 or more, depending on the laser power transmitted through the fiber 11 and the application. Have The slow axis and the fast axis are illustrated in FIG. 4a. The slow axis is the direction of the long dimension of the rectangular cross section of the fiber 11. The fast axis is the direction of the small dimension of the rectangular cross section of the fiber 11.

  As shown in FIG. 4 a, the core 42 is centered within the fiber 11. The vertical axis AA then divides the fiber 11 into two substantially identical and symmetric halves, and the horizontal axis BB divides the fiber 11 into two substantially identical and symmetric halves. However, as will be readily appreciated, the core 42 need not be centered within the fiber 11. For example, in one embodiment, the core 42 may be offset from the axis of symmetry BB of the fiber 11 (eg, the core 42 is provided “upper” or “lower” of the fiber 11 so as to approach the upper or lower portion of the coating 46. Good). Alternatively, or in addition, the core 42 may be offset from the axis of symmetry AA of the fiber 11 (eg, the core 42 is provided on the left or right side of the fiber 11 to approach the left or right side of the coating 46). Good).

The HARC fiber 11 is distinguished from conventional LMA fiber by identifying a high aspect ratio rectangular core 42 having a very large area (eg, up to 30000 μm ² or more). The area of the core 42 is increased by stretching the core 42 with only one dimension (slow axial direction) while maintaining other dimensions (fast axial direction) approximately the same as the original radius of the LMA core. The core thickness and numerical aperture of the fast axis may be chosen so that the fast axis direction of the core guides either a single mode or a few lower order modes, similar to an LMA fiber. In one embodiment, the fast axis thickness may be chosen to match the input beam quality to the fast axis. As a result, the coupled light substantially meets the fast axis dimension and NA of the core without significant power loss when incident on the cladding. According to this method, the cladding surrounding the core is also kept relatively thin in the fast axial direction so that the fiber is mechanically flexible in the fast axial direction. In one embodiment, since the laser beam is compatible with the core fast axis dimensions and NA, and the cladding is thin, the fiber will compromise the fast axis beam quality while maintaining the same input divergence. Without, laser radiation can be transported over any trajectory including fast axis bends.

Accordingly, the fast axis core parameters may be specified to maintain the quality of the input beam in the presence of fast axial bends. The parameters of the slow axis core are specified complementarily so that the fiber can be adapted to high laser power without the limitations of stimulated Raman scattering or Brillouin scattering or optical damage. For example, assuming that the core 42 dimension of the HARC fiber 11 is about 10 μm (which is single mode in the fast axis direction) and 0.5 mm (in the slow axis direction), the area of the core 42 is about 5000 μm ^2. , SRS threshold length L can be determined to be approximately 62.5 m for a 10 kW signal by using equation (2). Therefore, by allowing twice the margin, the HARC fiber 11 can carry a high-power laser beam by a length of about 30 m (that is, about half of 62.5 m) without causing SRS. For narrowband applications is limited by the SBS, equation (3) suggests that the size of the core is equal area of the core is a 40000Myuemu ² at 20 [mu] m × 2 mm, allowing the length of 10m . Thus, by allowing twice the margin, this fiber can carry a 1 kW narrowband laser beam for a length of about 5 m without causing SBS. As a result, by selecting appropriate dimensions for the core 42 of the fiber 11, the threshold of occurrence of SRS, SBS, or both can be increased.

  Given a properly designed HARC fiber carrying a high power laser beam without encountering SRS or SBS, determine the effect of core size and shape on the output beam quality of the laser beam propagating along the fiber It is worth doing. If the core end of the HARC fiber becomes reflective due to total internal reflection from the cladding material, and the input radiation meets the core dimensions and NA, the resulting beam quality is inadequate Conceivable. In addition, the output beam quality from such a guide may be inadequate if the emitted beam has a good beam quality but is improperly emitted. As described above, when the following procedure according to one embodiment is utilized, the quality of the output slow axis beam may actually be comparable to the quality of the input beam.

  First, in one embodiment, the beam is emitted directly along the fiber axis. Such emission preserves the divergence of the input light through all paths. The reason for this is that any reflection from the end of the slow axis core (parallel to the fiber axis and the beam propagation direction) reverses the angular tilt of the beam with respect to the slow axis, but the reflection is This is because the size is not increased. Because the high aspect ratio rectangular shape of the fiber withstands bending in the slow axis, HARC fibers that are bent in the fast axis remain effectively straight along the slow axis. This fact and specified emission conditions allow the input angular spectrum of laser radiation to be symmetric with respect to the fiber axis. Thus, reflections at the end of the slow axis cannot increase beam divergence.

  Second, in one embodiment, the beam is emitted so that it enters the HARC fiber to fully meet the slow axis dimensions of the core. For good input beam quality, the emitted beam excites only low order slow axis modes. Under these conditions, diffraction does not broaden the spot size of the input beam, and the effective beam size is substantially constant throughout the entire fiber path and is limited only by the width of the core. As mentioned above, HARC fibers that are bent in a fast axial direction do not suffer from slow axis bending, even in complex flexible paths. Thus, the divergence of the input beam to the output end of the fiber is preserved. This means that the slow axis beam quality (expressed by the product of beam size and beam divergence) is the same at the input and output ends of the fiber.

  Even if the shape of the HARC fiber causes bending only in the fast axis direction, a fiber that follows a complex path when delivering a high-power laser beam requires not only fast axis bending but also fiber twisting. there is a possibility. Twisting a portion of the HARC fiber can produce an effective negative flat lens that can defocus the light and increase beam divergence in the slow axis plane. Even if the fiber axis is effectively straight in the slow axis direction, twisting can compromise the beam quality of the slow axis.

  In the above situation, further steps may be implemented to maintain good beam quality. First, the entire propagation path may be designed as a series of two types of fiber ribbon deformations, with a purely fast axis bending or purely twisting without bending. The parts having each of the two types of deformation may be joined to each other. However, any given part of the fiber must have only one type of deformation. Under these conditions, the slow axis beam quality is automatically preserved while the fiber propagates through a purely bent region where the fiber is effectively straight and there is no torsional lens effect.

  Second, compensation for torsion-induced lens effects may be performed in each of the torsional regions to eliminate beam quality distortion with a negative torsion lens. Various methods of lens compensation may be implemented. For example, any of the two lens compensation methods described hereinafter in this application may be embedded in the fiber core as disclosed in Patent Document 3. The contents of Patent Document 3 are incorporated herein by reference. All compensation methods are based on the fact that the twisted lens behaves optically as an effective negative 1D gradient index (GRIN) lens with a focal length that depends on both the twist region and the twist angle. Compensation may be performed by building a positive GRIN lens in the core of the twisted region that is equal to the power of the negative lens originating from the twisted portion of the fiber and that has the opposite lens power. The positive lens effect may be generated by generating a slow axis refractive index change in the core, as is usually done to produce commercially available 2D GRIN lenses. A similar focusing effect also creates a core thickness profile along the slow axis, specifically by making the core thicker in the middle along the fiber axis and narrower near the ends. May be generated. Yet another compensation method is to provide one or more cylindrical lenses at each end of the torsional region. Here the lens is designed to produce a net positive lens effect equivalent to the defocus power of the twist-inducing lens.

  The method proposed above can be useful for transporting multi-mode input radiation over complex, dynamically flexible trajectories without sacrificing input beam quality. In this case, extreme tolerance for compensation is not necessary. As a result, slight optical coupling between the transverse modes is allowed. This is because slight optical coupling does not significantly increase the divergence of light and simply redistributes the phase between many excited fiber modes.

  Some fiber delivery applications may be required to deliver the highest single transverse mode high power radiation. The operation described in the above paragraph appears to be insufficient for single transverse mode input radiation. Bending the fiber can still involve mode mixing. This is because the tolerance to eliminate this effect in a full scale system is considered too impractical to be practical. Thus, even if only a single transverse mode is coupled at the HARC fiber core input, deflecting the path induces a redistribution of higher order mode output. Since the HARC fiber guides all modes without loss, higher order modes are present at the output with the fundamental mode and interference between the modes results, which compromises the quality of the output beam. Thus, a fundamentally different approach may be required to deliver truly single mode radiation over a deflected multimode fiber.

  This fundamentally different approach to fiber beam delivery of high power laser beams is based on other novel types of fibers, specifically semi-guided high aspect ratio core (SHARC) fibers. SHARC fibers have been proposed as amplifiers that produce high power lasers. Examples of SHARC fiber lasers are described in US Pat. The entire contents of Patent Documents 4 and 5 are incorporated herein by reference.

  FIG. 4b is a schematic cross-sectional view of a SHARC fiber 10 according to one embodiment. The fiber 10 has a core 12, a signal or fast axis cladding 14, a mode index matching (NIM) cladding 16, and a coating 18. The core 12 is provided between the two signal claddings 14, i.e., sandwiched. The core 12 and the signal cladding 14 are surrounded by the NIM cladding 16. The coating 18 covers the pump cladding 16. In one example, the coating 18 may be a polymer coating or any other flexible coating. The SHARC fiber 10 has a rectangular cross section as shown in FIG. 4b. In one embodiment, the rectangular core 12 of the SHARC fiber 10 typically has an aspect ratio that can range from about 30: 1 to 100: 1 or more, depending on the laser power transmitted through the fiber 10 and the application. Have The slow axis and the fast axis are illustrated in FIG. 4b. The slow axis is the direction of the long dimension of the rectangular cross section of the fiber 10. The fast axis is the direction of the small dimension of the rectangular cross section of the fiber 10.

  As shown in FIG. 4 b, the core 12 is centered within the fiber 10. The vertical axis AA then divides the fiber 10 into two substantially identical and symmetric halves, and the horizontal axis BB divides the fiber 10 into two substantially identical and symmetric halves. However, as will be readily appreciated, the core 12 and / or cladding 14 need not be centered within the fiber 10. For example, in one embodiment, the core 12 and the cladding 14 may be offset from the axis of symmetry BB of the fiber 10 (eg, the core 12 and the cladding 14 may be “top” of the fiber 10 to approach the top or bottom of the coating 18 or It may be provided at the “bottom”). Alternatively or in addition, the core 12 may be offset from the axis of symmetry AA of the fiber 10 (eg, the core 12 and the cladding 14 are on the left or right side of the fiber 10 to approach the left or right side of the coating 18). May be provided).

As with the HARC fiber 11, the SHARC fiber 10 is distinguished from conventional LMA fibers by identifying a high aspect ratio rectangular core 12 having a very large area (eg, up to 30000 μm ² or more). The area of the core 12 is increased by stretching the core 12 with only one dimension (slow axial direction) while maintaining other dimensions (fast axial direction) approximately the same as the original radius of the LMA core. The core 12 is designed to be semi-waveguide. Thus, conventional refractive index guiding by total internal reflection (TIR) occurs along the two large surfaces of the rectangular core at the interface between the cladding 14 and the core 12. On the other hand, the slow axial narrow core end 13 is designed not to support TIR. In one embodiment, the refractive index step at the core-edge boundary 13 is small enough—eg, less than 500 ppm (eg, 10 ⁻⁴ ) —so that higher order core modes are (in the slow axis direction) Along) freely diffracts outside the core. Thus, all wide axis transverse electromagnetic modes (ie modes along the slow axis) suffer from radiation loss through the core end 13. Although the SHARC fiber 10 has been described with reference to a particular fiber structure to create a semi-waveguide core, it is possible that other structures can be used to produce the same semi-waveguide characteristics. Easy to understand.

One characteristic that the SHARC fiber 10 shares with the HARC fiber 11 is that the external dimension of the fiber along the fast axis is relatively small (eg, approximately about 30000 μm ² ), even if the core 12 area can be very large (> 30000 μm ² ). (Not exceeding 500 μm). In addition, in one embodiment, the numerical aperture NA of the core 12 along the fast axis is relatively small. For example, a core thickness of about 15 to 25 μm is about 0.06. This value is sufficient to prevent the signal from leaking out through the fast shaft cladding under a reasonable bend radius. Under these conditions, the SHARC fiber 10 is mechanically flexible in the fast axial direction and the SHARC fiber 10 inherently bends in this direction without causing fundamental mode loss, or A coil may be constructed. In contrast, as explained in the previous paragraph, a “rod-like” fiber with a very large diameter and low NA annular core is extremely sensitive to bending and does not sacrifice beam quality, In addition, a low loss coil can be configured without suffering from a large diffraction loss. Thus, by providing bendability and mechanical flexibility while providing a properly aligned fast axis core thickness and NA, the SHARC fiber 10 is a candidate for high power laser fiber delivery. Satisfy at least some of the criteria.
In addition to these criteria, as previously described for the HARC fiber, the SHARC fiber 10 requires a laser power supply requirement without substantially causing SRS or SBS by raising the threshold for stimulated Raman scattering or stimulated Brillouin scattering. Provide a sufficiently large core area to satisfy

  A further criterion is to determine whether the SHARC fiber 10 can provide a relatively high power laser beam without degrading the output and beam quality of the input laser beam. The SHARC fiber 10 can essentially operate reliably due to its unique core design. Instead of relying on conventional wavelength-based mode rejection—which is not effective at multimode wavelengths, it will result in a severely degraded output beam—the SHARC fiber 10 is affected by the core end 13. It relies on having a sufficiently wide core in the slow axis direction that naturally allows unacceptable beam propagation.

The collimated input beam is large enough so that the Fresnel length L _{Fr to} W _b ² n / λ exceeds the fiber length L (where n is the refractive index of the core 12 and λ is the wavelength of the laser) If the core 12 has a slow axis beam width w _b and the material of the core 12 is optically uniform, the beam will not be reflected from the end of the core. As a result, the collimated output beam retains the quality of the input beam. In addition, bending the SHARC fiber 10 in the fast axial direction does not affect the quality of the beam in the slow axial direction, as the model shows and experiments warrant. One reason for this fact is that the very high aspect ratio of the core facilitates the separation of diffraction processes along the slow and fast axis directions. “Loss filtering” if any possible imperfections induce wavefront distortion, or if other processes scatter the radiation into a more divergent ray from the original laser beam profile. The purification by prevents the beam quality from being deteriorated. Loss filtering utilizes the fact that higher order transverse electromagnetic modes representing divergent light diffract more strongly than the desired lowest order transverse electromagnetic mode and leak out of the semi-waveguide core region. This leakage is shown schematically in FIG. 4 as three diverging arrows on each side of the core end 13. The application of loss filtering to mode control in SHARC fiber amplifiers and lasers is described in detail in US Pat. The contents of Patent Document 5 are incorporated herein by reference.

  The SHARC fiber 10 is single mode along the thin fast axis, or is effectively single in this fast axis by using the same coil construction method described above at the LMA fiber location. Configured to be a mode. In the latter case, the thickness of the flat core 12 is relatively small, and at the same time, by reducing the numerical aperture NA of the core 12 along the fast axis direction, a number of guided fast axis modes can be reduced to 1 to several. Reduce to 1 mode. The mode is cleared by the fiber bend inherent in forming the coil in the fast axial direction. That is, the higher order fast axis modes are eliminated, and essentially only the fundamental transverse electromagnetic mode propagates without radiation loss. However, configuring the coil has sufficient ability to maintain a single mode if the fiber configures a static coil, but for beam delivery applications, the appropriate coil diameter is always maintained during dynamic operation. It is not easy to do. It is therefore preferable to design the fiber so that it is single mode in the fast axial direction.

  In the active version of SHARC fiber 10, slow axis mode control is achieved by utilizing a combination of two complementary mechanisms: (i) loss filtering and (ii) gain filtering. The loss filtering mechanism is based on the fact that all slow axis modes experience a loss of radiation incident on the cladding 16 via the “open” core end 13. Refractive index step so as to substantially eliminate higher order transverse electromagnetic modes along the slow axis in SHARC fiber 10 by providing higher order transverse electromagnetic mode losses greater than the loss of lower order transverse electromagnetic mode propagation. (Refractive index difference between the refractive index of the clad 16 and the refractive index of the core 12) is designed to be substantially zero. Radiation loss depends strongly on the mode, and prefers the fundamental mode with less loss than others. The gain filtering mechanism is realized by designing the gain stripe width to be narrower than the core width in the slow axis direction. The mode overlap due to such gain stripes is much larger in the fundamental mode than in any higher order slow axis mode. As a result, the gain is significantly improved in the basic slow axis mode. Detailed studies on the propagation of the SHARC fiber 10 have quantified the mode-dependent loss and gain discrimination or filtering described above for multiple lowest order transverse electromagnetic modes. These studies ensure that loss identification is adequate to achieve the desired mode control of the active fiber.

  When the SHARC fiber 10 is configured to be passive—in the case of laser propagation and delivery—since there is no gain medium in the SHARC fiber 10, the gain filtering mode control mechanism is not effective overall. However, loss filtering may be used to remove any higher order slow axis modes that may be generated while propagating through the core 12.

However, the loss filtering mechanism provides limited functionality. Loss filtering can be used to remove higher order transverse electromagnetic modes propagating in the SHARC fiber 10, but can only be used within certain limits. For example, for a slow axis core width of about 2 mm, the lowest order mode has a propagation loss rate α ₀ of about 0.004 m ⁻¹ . This reaches about 4% for a supply length of 10 m. The loss rate α _m of the mode m changes with respect to the square of the slow axis mode number, that is, α _m is proportional to α ₀ (m + 1) ² (m = 0, 1, 2, 3,...). since mode 1 and alpha ₁ and alpha ₂ respectively representing the loss factor of 2 each 0.016 M ^-1 and 0.036 m ^-1. These values correspond to losses of about 15% and about 30% for a fiber length of 10 m, respectively. Such losses are not negligible, but significant losses in excess of 90% or the most effective transverse electromagnetic mode discrimination occurs only in higher order modes with m> 5. Thus, when the fundamental transverse electromagnetic mode is distorted as it propagates through the core 12, higher order slow axis aberrations are effectively filtered, while lower order aberrations remain in the output beam and are transmitted. The quality of the beam is reduced.

  A number of methods have been identified for mitigating or substantially eliminating low order aberrations in the transmitted beam. These methods described in the following paragraphs may be used separately or in combination to achieve the desired effect.

First, in one embodiment, the width of the core 12 in the slow axis direction can be reduced to an optimum value that is sufficiently high in a mode with a loss factor of m = 1 (ie, the first higher order mode). The model shows that the radiation loss rate α _m through the slow axis end 13 of the core varies with the cube of the width w, ie α _m is proportional to 1 / w ³ . For this reason, the loss rate α _m increases 4.6 times in all modes when a core width w of 1.2 mm is used instead of 2 mm. As a result, the transmission loss in the higher order transverse electromagnetic mode after 10 m propagation increases by about 53% at m = 1 and by about 82% at m = 2. The loss becomes very strong, exceeding 95% at m = 2. Such loss identification may be sufficient for mode “purification” (ie, higher order mode identification). As a result, only the fundamental transverse electromagnetic mode is emitted to the SHARC fiber 10, while when the fundamental transverse electromagnetic mode propagates through the optical imperfection in the core 12, the fundamental transverse electromagnetic mode is triggered. The fact that higher order transverse electromagnetic modes occur is taken into account. However, one problem with this method is that the cost to improve beam quality results in increased radiation loss in the low order fundamental transverse electromagnetic mode. For example, in the core 12 having a width of about 1.2 mm and the fiber 10 having a length of about 10 m, the loss of the fundamental transverse electromagnetic mode to the cladding 16 reaches about 18%.

Second, since propagation loss occurs when the signal beam is incident on the core end where loss is likely to occur, the loss is eliminated by widening the core 12 sufficiently that the beam does not reach the core end 13. Can be done. This situation can be realized if the following conditions are met.
(A) In the input beam width w _b inside the core 12, the fiber length L is specified to be less than the Fresnel range L _{Fr to} w _b ² n / λ.
(B) The core material is substantially optically uniform.
(C) The core width w exceeds the width of the diffraction signal at the output end of the fiber.
(D) The defocus effect of the slow axis lens effect induced by arbitrary twist is weak compared to the diffraction effect corresponding to a particular mode size. This condition can occur because the torsion angle is small or because negative torsion lenses are accurately compensated.

For example, in the SHARC fiber 10 having a length of about 10 m, the input beam width w _b is selected to exceed 2.2 mm and the width of the core 12 is selected to exceed 3 mm so as to satisfy the above-described conditions. It is. Since optical uniformity is a common condition for realizing a high performance fiber with minimal mode distortion, condition (b) is not unique to the SHARC fiber 10.

  For both the HARC fiber and the SHARC fiber, collectively referred to as (S) HARC fiber, in the following paragraphs, two separate mode distortion sources need to be controlled. The first strain source is the optical imperfection of the core. Such imperfections can be caused by either the non-uniformity of the glass (e.g. silica) constituting the core or the inaccurate shape of the core-in particular the fluctuations in the thickness of the core. Accurate selection and refining of the core glass material is usually used to achieve the best core-glass uniformity, with precise control of temperature uniformity during the drawing process. However, if any core imperfections induce wavefront distortion, or if other processes scatter the radiation to make it more divergent from the original beam profile, then cleanup by loss filtering in the SHARC fiber This option helps to prevent beam quality degradation in the output beam.

  The second optical strain source is fiber deformation. The deformation of the fiber can be suppressed as follows. First, the flat fiber cores 12, 42 of the fibers 10, 11 can each be purely single mode in the fast axial direction. This may cause the fast axis mode profile to be insensitive to any bend along its fast axis direction. Second, as the model shows and experiments have warranted, bending (S) HARC fiber in the fast axis direction has no detrimental effect on the beam quality of the slow axis. As mentioned above, one reason for this fact is that the very large aspect ratio facilitates complete separation of the diffraction process along the slow axis and the diffraction process along the fast axis. is there. However, bending the cores 12 and 42 in the slow axis direction may impair the slow axis beam quality. For this reason, another allowance may be taken to prevent the fibers 10, 11 from bending on a slow axis. In one embodiment, this is accomplished, for example, by specifying that the fibers 10, 11 have a rectangular core 12, 42 (eg, as illustrated in FIGS. 4a and 4b) and a rectangular profile. obtain. In one embodiment, an aspect ratio of the outer shape of the fibers 10 and 11 (ie, the ratio of the width and height of the fibers 10 and 11) of 3 or more may be sufficient for this purpose. Thereby, the whole fibers 10 and 11 behave as ribbons. As a result, the fibers 10 and 11 as a whole are easily adapted to bending and twisting in the fast axial direction, but strongly resist deformation in the slow axial direction.

The model and experiment are consistently that (S) HARC fibers 10,11 do not affect the beam quality of the substantially slow axis, while twisting induces further divergence in the laser beam, thereby causing the fibers 10, 11 shows that it functions as an effective negative lens that affects only the slow axis direction. In one embodiment, the effective negative lens focal length F varies with F = −ΔL / φ ² . Here, ΔL is the twist length, and φ is the twist angle integrated over the length ΔL. A twist-induced lens generated in separate regions of the fibers 10, 11 can provide a coupled lens that is the sum of the individual lenses, regardless of whether the twist is in the same direction. Thus, in one embodiment, if the ribbon fibers 10, 11 undergo a uniform twist along the entire length L of the fiber, the integrated twist angle is much longer than the fiber length (ie F >> L). ) And to suppress defocusing of the signal beam based on the lens effect, it must be less than about 0.3 radians. The same expression is to localize the twist over a short portion of the fiber to fit a large overall twist, ΔL << L is undesirable because it produces a strong negative flat lens. Show.

  Thus, another way to ensure the transmission of a high quality laser beam is to optimize the trajectory and twist of the fibers 10, 11 when the beam delivery is made through a non-straight path with defects. . A universally optimal fiber trajectory cannot be defined for all applications. This is because the trajectory depends on the shape of a specific path and the positions of the input mount and output mount. The goal of optimization is to distribute the deformation along the fibers 10 and 11 so as to alleviate local twists by adjusting the fibers 10 and 11 so that the trajectory includes a series of bends. By distributing the deformation along the fibers 10, 11, the overall twist can be suppressed and the twist rate can be distributed along the fibers 10, 11 to be as smooth as possible.

Still other methods may be used specifically to deal with twist-inducing lenses. If twisting of the supply fiber is unavoidable, the design may be improved so that the twisted portion of the fiber occurs over the fiber length δL near the output end of the fiber. This length is then specified to be shorter than the intrinsic distortion length L _dist associated with the twist-induced focal length F. Specifically, it is _{δL <L dist ~w core / δθ} F. Where δθ _F = (w _b / 2F) is the twist-induced divergence angle, w _core and w _b are the core width and beam width, and F = δL / φ ² is related to the twist angle φ. The focal length of the negative lens. The latter condition for δL is satisfied if w _core > w _b φ ² if the _{core is} sufficiently wide compared to the beam size. This is easily met if the twist angle φ is less than 0.5 radians, but if the twist angle exceeds 45 °, it requires an expanded core width, w _core > w _b To do. Such geometrically shaped output beams are not distorted but carry a wavefront with curvature. Therefore, an additional optical system that maintains beam collimation is required. Slow axis lens with fixed cylindrical position (for static feed fiber path) or adjustable position (for dynamic feed fiber path) eliminates residual effects of twisted lens Can be added at the fiber output to do this.

  In the previous paragraph, various methods have been described in which (S) HARC beam delivery fibers 10, 11 are configured to provide an output beam having a beam quality comparable to the input signal. However, for SHARC fiber options, some of these methods can cause some loss of signal or radiation output. These radiation losses simply represent the power remaining inside the cladding rather than inside the core. These radiation losses do not represent the heat inside the fiber that must be dissipated. Thus, optical power “lost” inside the cladding may be extracted from the fiber.

  There are several ways to remove the signal output from the cladding. In one embodiment, for example, the numerical aperture NA of the cladding 16 at the outer surface is designed low. As a result, most of the power lost from the core can be completely ejected from the fiber 10. Alternatively, in other embodiments, a silica splice can be periodically attached around or coupled to the fiber 10 along the length of the fiber 10. In one embodiment, the splice ring may be index-matched to the clad 16 of the SHARC fiber 10 to extract the output from the clad 16 and to guide the output to a suitable absorber. In other words, the refractive index of the joint ring is substantially equal to the refractive index of the clad 16.

  For high power applications, one way to extract the optical output in the cladding 16 is as follows. First, the SHARC fiber 10 has a coating 18 (eg, a polymer coating) that protects the glass fiber surface. In one embodiment, the polymer coating may be chosen for high transmission at the signal wavelength. As a result, the output in the cladding 16 can jump out of the fiber 10 without being absorbed by the coating 18.

The second method is to include the fiber 10 inside a water-cooled jacket that can be configured to remove power by convection. An analysis was performed to ensure that the method described above was appropriate even for high loss rates, such as 0.016 m ⁻¹ . The optical power lost from the fiber core 12 was set to be equal to the output q _colant transported from the fiber by convection through the cooled cylinder. The optical power lost from the fiber core 12 is given by equation (4). The power transported by convection from the fiber 10 through the cooling sleeve is given by equation (5).

ここで、Ｐ_ｉｎは入力放射出力に対応し、Ｐ_ｏｕｔは出力放射出力に対応し、αはファイバ１０の放散率で、Ｌはファイバ１０の長さで、Ｃ_ｐはファイバ１０の周辺の流体の熱容量で、ｄｍ／ｄｔは流体の質量流で、ΔＴは入力での流体と出力での流体との温度差に対応し、ρは冷却流体の密度で、Ａは流体流の断面積で、かつ、Ｕは流体の流速である。

Here, P _in corresponds to the input radiation output, P _out corresponds to the output radiation output, α is the dissipation factor of the fiber 10, L is the length of the fiber 10, and C _p is the fluid around the fiber 10. Dm / dt is the mass flow of the fluid, ΔT corresponds to the temperature difference between the fluid at the input and the fluid at the output, ρ is the density of the cooling fluid, A is the cross-sectional area of the fluid flow, U is the flow rate of the fluid.

Thus, in one embodiment, for a fiber length of about 10 m with a core loss rate of about 0.016 m ⁻¹ , the optical power q _optical lost from the core 12 is calculated to be about 1.5 kW. By using a cooled cylinder around the fiber 10 with a water cross section of 1 cm ² and using an allowable temperature rise ΔT of 10 ° C. between the input and output manifolds, the required water The mass flow is calculated to be about 37.4 g / s. However, it is known that the heat capacity C _p of water is 4.186 J / (g · ° C.). Since the density ρ of water is 1 g / cm ³ , this mass flow is equal to the flow rate U = 9.36 cm / s and the volume flow 2.25 l / min. This can be easily maintained by a moderate water pump. Although water is used in this example, any other suitable coolant may be used.

  The integration of the (S) HARC fibers 10 and 11 for supplying the high-power laser beam from the laser source to the point of use may be performed as follows. (S) The input ends of the HARC fibers 10 and 11 may be connected to a laser source. The laser source is, for example, but not limited to, an LMA fiber laser (eg, a 10 kW LMA fiber laser) via an optical coupler 20 described in the following paragraphs referring to FIG.

  FIG. 5 schematically shows an optical coupler 20 according to one embodiment. In one embodiment, the optical coupler 20 is configured to couple the output of a fiber laser (not shown) to the (S) HARC fibers 10, 11 to provide laser energy to the desired point of use. . In one embodiment, neither the fiber laser (signal fiber) nor the (S) HARC fiber (supply fiber) 10, 11 can transmit the laser signal 24 to the (S) HARC fiber 10 without requiring free space optical communication. , 11 are fused and connected to opposite ends 21 and 22 of the optical coupler 20 so as to constitute a monolithic device. (S) The output ends of the HARC fibers 10, 11 may be fused and connected to the end caps shown in FIG. The end cap allows the high power beam output by the (S) HARC fibers 10, 11 to be expanded and reshaped without using free space optical communication. Once the output beam jumps out of the end cap, the signal may be processed in the same way as any other free space beam at the inner gimbal. A detailed description of the end cap will be given after describing the coupler.

  The optical coupler 20 reshapes the diffraction-limited annular input laser beam 24 (eg, having a diameter of about 20-30 μm) into a collimated high aspect ratio elliptical beam for injection into the (S) HARC fibers 10, 11. To do. The narrow dimension of the reshaped beam remains comparable to the size of the initial beam 24. However, the beam is expanded and collimated as necessary with orthogonal width dimensions up to 2 mm or more. In one embodiment, the optical coupler 20 may be implemented as an integrated member. In one embodiment, the optical coupler 20 is configured to have a small size, and two very different optical fibers (input fiber laser and (S) HARC fiber 10, 11) are connected to the ends 21 and 22. It is configured to allow connection to the optical coupler 20 above. In one embodiment, input end 21 is configured to receive commercial annular core fiber and output end 22 is configured to receive (S) HARC fibers 10, 11. However, as can be readily understood, the input end 21 and the output end 22 are each (S) HARC fiber--for example (S) HARC fiber from (S) HARC fiber laser at input end 21; and The input end 22 may be configured to receive the (S) HARC beam supply fiber 10, 11- from the (S) HARC fiber laser. In this latter case, where the beam delivery fiber is applied in a situation where it has a different core dimension than the dimension of the (S) HARC fiber laser core, the coupler 20 connects with the respective core dimensions of the two (S) HARC fibers. Designed as such.

  In the fast axis direction (perpendicular to the plane of FIG. 5), the optical coupler 20 utilizes a normal refractive index based waveguide to maintain a constant fast axis dimension. Here it is assumed that the annular core diameter is the same as the thin dimension of the rectangular core. If the application requires that the fast axis dimensions be different, waveguides based on the fast axis refractive index will cause the length of the optical coupler 20 to be as long as necessary to match the annular beam to the rectangular thin dimensions of the cores 12,42. It may taper or expand along the length.

  In the slow axis direction, the beam jumping out of the annular opening is allowed to diffract freely. However, in one embodiment, the medium of the optical coupler 21 is selected to have a gradient index profile that functions as a one-dimensional (1D) gradient index (GRIN) lens. In one embodiment, the length of the optical coupler 20 is specified so that the slow axis GRIN lens is ¼ pitch long that essentially collimates the diverging beam that emerges from the input fiber of the annular core. Both the 1/4 pitch length of the signal beam and the final slow axis dimension can be controlled by appropriately identifying the refractive index profile in the GRIN portion.

  To confirm the performance of the optical coupler 20, a 3D propagation simulation showing coupling from a relatively small core diameter of the LMA input fiber (eg 20 μm) to a (S) HARC fiber 10, 11 having a core of 20 μm × 1 mm, for example. Was executed. FIG. 6 shows the result of the 3D propagation simulation according to one embodiment in grayscale. The vertical axis (y) represents the distance that the laser propagates inside the LMA fiber and the optical coupler 20. The horizontal axis (x) represents the dimension of the LMA fiber and the optical coupler 20 along the slow axis direction. As shown in FIG. 6, when the input fiber (for example, LMA fiber) jumps out, the laser beam propagates through the optical coupler 20 having a flat GRIN lens in the direction of the white arrow. As the laser beam propagates along the y-axis, the laser beam remains guided in a fast axial direction perpendicular to the planes of FIGS. 4a and 4b. However, the laser beam begins to diffract as it propagates in the slow axis direction (x-axis direction as shown in FIG. 6) under the influence of the flat GRIN lens in the optical coupler 20. As shown in FIG. 6, the effect of the planar GRIN lens systematically causes beam divergence on the slow axis until the beam is collimated after propagating within a quarter pitch length (eg, 20 mm) within the optical coupler 20. To change it. At the end of the quarter pitch length, the laser beam is expanded to fill the width (for example, 1 mm) of the (S) HARC fibers 10 and 11 at the end of the optical coupler 20. While the optical coupler 20 is described with reference to a specific size laser beam and fiber, it will be appreciated that the optical coupler 20 may be designed and configured to fit other sized input and output fibers. Let's go.

  FIG. 7 is a schematic view of an output end cap 50 according to one embodiment. The function of the end cap 50 is that the signal beam in the (S) HARC fibers 10 and 11 diverges greatly in the fast axis direction and diverges small in the slow axis direction and enters the output end cap 50. Based on the facts. In one embodiment, end cap 50 is made from bulk fused silica. Due to this divergence angle difference, there is a propagation distance in the end cap 50 at which the beam size in the fast axis direction is substantially equal to the beam size in the slow axis direction. End cap 50 terminates at this position. This position corresponds to the exit surface 52 of the end cap 50. In one embodiment, the exit surface 52 is shaped to mitigate the divergence in the fast axial direction and match the slow axial value without changing the slow axial divergence. Thus, the exit beam has an aspect ratio of approximately 1 and has essentially the same divergence in two orthogonal lateral directions. For example, if the cores 12 and 42 of the (S) HARC fibers 10 and 11 have a size of 10 μm × 2 mm, the analysis results show that the end cap 50 length is about 30 mm and the radius of curvature of the exit surface 52 is about 25 mm. As can be readily appreciated, if one simply wants to reduce the beam aspect ratio but does not want the final aspect ratio to be 1, the end cap length is reduced to such a desired final aspect ratio. It may be specified to a length that gives.

  In one embodiment, a second end cap may be provided for the (S) HARC fibers 10 and 11. The second end cap may be connected to the input end of the (S) HARC fibers 10 and 11. The second end cap may be configured to receive a free space laser beam from a solid state laser or other type of laser so that the free space laser beam is input to the (S) HARC fibers 10,11. In one embodiment, the second end cap may be similar to the end cap 50 described above.

As can be seen from the previous paragraph, the method of fiber feeding a high power laser beam comprises providing a fiber (fiber 11). The fiber (fiber 11) has a core (core 42) having an elongated cross section so as to have a high aspect ratio along the slow axis direction of the fiber (fiber 11). The core (core 42) has a first refractive index. The fiber (fiber 11) includes a cladding having a second refractive index. The method further includes emitting a laser beam (eg, a laser beam having an output of about 10 kW or greater) into the fiber (fiber 11). The laser beam has a width w _b smaller than the width w of the core (core 42) along the slow axis direction. The method further comprises propagating the laser beam through the fiber (fiber 11) while mitigating the generation of SRS and / or SBS.

As can be seen from the previous paragraph, the method of fiber feeding a high power laser beam includes providing a fiber (fiber 10). The fiber (fiber 10) includes a core (core 12) having a first refractive index having a cross section elongated to have a high aspect ratio along a slow axis direction of the fiber (fiber 10); A first signal cladding and a second signal cladding (cladding 14) having a second refractive index provided in contact with the core (core 12) and sandwiching the core (core 12); and the core (core 12) A third cladding (cladding 16) having a third refractive index substantially surrounding at least the side end (end 13) is provided. The difference between the first refractive index of the core (core 12) and the third refractive index of the third cladding is small (for example, less than 500 ppm (for example, 10 ⁻⁴ )). The method further includes emitting a laser beam (eg, a laser beam having an output of about 10 kW or greater) into the fiber (fiber 10). The laser beam has a width w _b smaller than the width w of the core (core 12) along the slow axis direction. The method substantially eliminates the higher order lateral slow axis modes propagating along the fiber (fiber 10) while mitigating the generation of SRS and / or SBS, thereby reducing the lower order lateral direction. The method further includes propagating the laser beam through the fiber (fiber 10) to provide a higher order transverse slow axis electromagnetic mode loss that is greater than the slow axis electromagnetic mode loss.

  There are various applications for the embodiments of the present disclosure. For example, in a battleship laser system, the (S) HARC fibers 10 and 11 make it possible to install laser light source (s) behind the deck. The beam from the laser source (s) can be re-routed to go directly to the beam director device via the (S) HARC fibers 10, 11 without requiring an elbow-shaped path. . For example, in an aircraft-based laser weapon system, the SHARC fibers 10, 11 allow packaging of laser light source (s) at a location on the vehicle that allows the aircraft's center of gravity and flight characteristics to be optimized. To do. From these positions, the laser can be delivered to the desired point by using (S) HARC fibers 10,11.

  Thus, the (S) HARC fibers 10, 11 described herein have a wide range of applications including delivery of an arbitrary laser beam to a desired location while enabling efficient packaging. (S) One of the advantages provided by HARC fibers 10, 11 is that of packaging on military platforms that occurs when the output fiber from the laser source is constructed to be only a few meters long. It is to solve the difficulty. (S) HARC fibers 10, 11 enable transmission of the laser beam (s) from the fiber laser light source to the beam director turret device. In the beam director turret device, the beam (s) from the fiber (s) may be emitted into free space and combined.

  During beam delivery, the fibers 10, 11 can be used, for example, to propagate the HEL beam from the laser generator to the inner gimbal of the beam control system. This replaces the more typical elbow-shaped path. The fibers 10, 11 must be able to bend irregularly by irregular deformations. This is because the beam director can move (sometimes quickly) with two degrees of freedom. That is, the fibers 10 and 11 must be able to adapt to any angular freedom of the assumed biaxial gimbal assembly. In one embodiment, the fiber is in each of two correspondences so that the fibers 10, 11 are consistently bent only in one direction and change its direction when needed by twisting the fiber 90 ° C. It may be aligned along each of the displacement directions. This method allows the fibers 10, 11 to accommodate angular displacement along two orthogonal directions. However, as described above, twisting the rectangular core of the HARC fiber can produce a negative gradient index (GRIN) lens in the slow axis plane of the rectangular core. Therefore, there is a risk of distortion caused by twisting the fibers 10 and 11 by 90 °.

FIG. 8 shows the calculation results of a theoretical model representing the focal length induced by torsion as a function of torsion angle, and the focal length induced by torsion measured over a range of torsion angles ranging from 0 to 0.35 radians. ing. As shown in FIG. 8, with a constant twist rate along a twist length of about 30 cm, the twist lens monotonically increased to reach 0.4 diopters at 0.35 radians. In the range of torsion angles of at least 0-0.35 radians, the model is consistent with the data. The model is based on the following formula:

F _T = L _T / φ ² (6)

Here, the focal length F _T twist lens is a function of the fiber length L _T of twist angle φ and twist is imposed.

  Since the length of the supply fiber can be about 10 m or more, the twisted lens is so strong that it cannot be adequately compensated by simply using the correction optics at either end of the beam supply fiber 10,11. There is a fear. Thus, in some embodiments, twisting in the beam delivery fiber may only occur at a predetermined one along the entire fiber length. For example, this can be accomplished using a fiber optic twist module 100 that can be connected to the beam delivery fibers 10, 11 at appropriate locations along the beam delivery fibers 10, 11.

  In the embodiment illustrated in FIG. 9, the optical twist module 100 has a first end 102, a second end 104, and a twisted portion 106 between the first end 102 and the second end 104. An optical fiber 101 is included. The twisted portion 106 creates a distortion effect. The optical twist module 100 includes a compensator 108 that compensates for distortion effects, and further includes a housing 110 configured to store at least a portion of the optical fiber 101 including the twisted portion 106. In one embodiment, the housing 110 is a hermetically sealed housing. The first end 102 of the fiber 101 is configured to connect to the input fiber 112 to operate. Second end 104 is configured to connect to output fiber 114 for operation. In some embodiments, input fiber 112 and output fiber 114 may be integrated with fiber 101. In other embodiments, input fiber 112, output fiber 114, and fiber 101 may be separate components that are operatively connected to each other. It is anticipated that the fibers 101, 102 and / or 114 may be either HARC fibers or SHARC fibers.

  In the embodiment of FIG. 9, the distortion compensator 108 includes a lens arrangement configured to compensate for distortion effects arising from the twisted portion 106. According to FIG. 8 and equation (6), at a twist angle of about 90 °, the focal length of the twisted lens should be about half the length of the twisted region of the fiber. Accordingly, torsional lens compensation may be achieved by using one or more free space cylindrical lenses. Cylindrical lenses may be used for free-space coupling of an annular beam incident on a planar waveguide located at the focal point of the lens image plane.

  In one embodiment, the torsion module 100 includes a distortion compensator 108 that includes a plurality of cylindrical lenses 116a-116f coupled in series, and optionally at least one provided at the input and / or output ends. It also has a polarizing member (not shown). Lenses 116a-116f may be conventional refractive lenses. However, other types of lenses, such as flat GRIN lenses, may be used. Lenses 116a-116f and fiber 101 may be included in a housing 110 having an input fiber 112 and an output fiber 114 at each end. Module 100 may be implemented in a manner similar to various telecommunications components, such as isolators or dielectric filters. For example, isolators and dielectric filters all embody free space components contained within a module having a fiber pigtail that allows the module to be simply coupled to a fiber-based system. Specifically, the module 100 has an input fiber 112 and an output fiber 114 at each end that allow these components to be coupled to a fiber-based system. Module 100 may be designed to maintain optical alignment between optical elements over a range of environmental conditions.

  The input end of the input fiber 112 may be coupled to an incoming HARC fiber or SHARC fiber (not shown in FIG. 9). The output end of the output fiber 114 may be coupled to an outgoing HARC fiber or SHARC fiber (not shown in FIG. 9). The cylindrical lenses 116 a, 116 b may be configured to collimate the beam and guide the beam to the twisted portion 106 of the fiber 101. The wide dimensions of the fiber may be sufficiently wide so that essentially no diffraction spread occurs at this dimension. However, it is also anticipated that additional lenses can be provided to collimate the beam in the slow axis direction. The lens 116c is configured to focus the slow axial beam as a “pre-distortion” that at least partially compensates for the flat negative torsion lens arising from the twisted portion 106 of the fiber 101. good. As shown in FIG. 9, the slow axis direction of the fiber 101 after 90 ° twisting in the twisted portion 106 is perpendicular to the slow axis direction of the fiber 101 in front of the twisted portion 106. Lens 116d may be configured to complete torsional lens compensation by imaging a slow axial beam onto output fiber 114. The lenses 116e, 116f may be configured to collimate the fast axial beam and guide the beam toward the output fiber 114. The output fiber 114 may then be coupled to another HARC supply fiber (not shown in FIG. 9). As described above, the lenses 116 a-116 f, the fiber 101 having the twisted portion 106, and the input fiber 112 and a part of the output fiber 114 may be included in the housing 110.

  In other embodiments, the fiber optic twist module 100 (see FIG. 10) is an all fiber module or all glass module (not shown in FIG. 10) that does not jump out of the fiber medium until the light beam reaches the output end of the fiber 114. Not used). Here, the output end of the fiber 114 transitions to free space or is coupled to another fiber. The other fiber may be a HARC fiber or a SHARC fiber. Note that fiber 101 and fiber input end 112 and output end 114 are not shown in FIG. 10, but module 100 has such features included within housing 110. In such an embodiment, all of the fiber 101, input end 112, and output end 114 may be the combined length of a single fiber. In such an embodiment, instead of using the above-described free space lens device, the module 100 may include another form of distortion compensation device. The compensation device or method is based on the fact that the twisted lens operates optically as an effective negative 1D gradient index (GRIN) lens, the focal length of which depends on both the length of the twisted region and the twist angle. . Thus, in one embodiment, the distortion compensator has a positive lens effect that has a lens power that is opposite and substantially equal to the power of the negative GRIN lens effect arising from the twisted portion 106. That is, the compensator may comprise building a positive GRIN lens in the core of the twisted region that has a lens power equal to and opposite to the power of the negative lens originating from the twisted portion of the fiber. In one embodiment, the positive lens effect is generated by creating a slow axis refractive index change in the core, as is usually done to produce commercially available 2DGRIN lenses. Alternatively, or in addition, a positive lens effect can be generated by forming a core 101 thickness profile along the slow axis.

  Any one of the built-in GRIN lens compensators described above may be used. In one embodiment, the module 100 may also include an end mount device that includes end mounts 120, 122 that may be used to adjust the length of the torsional portion 106. The end mounts 120, 122 are used to reduce or remove any residual lens power from either the torsion 106 or the compensator used in the module 100 by fine-tuning torsional lens compensation. It ’s good. As shown in equation (6), for a given twist angle, the twist lens may be adjusted by changing the twist length. Thus, in embodiments having a module 100 that includes a 90 ° twisted portion over length L, the module 100 may be adjusted during assembly to remove any net optical distortion.

  In the embodiment illustrated in FIG. 10, each end mount 120, 122 is provided with a slot 124 constructed and arranged to have a thickness and width comparable to the thin dimensions of the fiber 101. At least a portion of the fiber 101 is contained within the slot 124. The housing 110, which in this embodiment takes the form of a tube, has longitudinal slots 126 and 127 provided therein. In one embodiment, slot 126 extends the entire length of the housing. One function of slot 126 is to allow the fiber to be inserted into the housing from the side. The slot 127 (some of which are not shown in this perspective view) is several times longer than the length of the end mounts 120,122. The slots 126 and 127 are constructed and arranged to receive the flanges 128 of the end mounts 120, 122 when the end mounts 120, 122 are connected to the housing 110. In one embodiment, the end mounts 120 and 122 are rotated 90 degrees relative to each other and are provided to enter the slots 126 on the first end 130 and the second end 132 of the housing 110, respectively. It is done. Slots 126 and 127 may be constructed and arranged to be longer than the length of end mounts 120 and 122. As a result, the end mounts 120 and 122 slide in the longitudinal direction along the housing 110, and as a result, the distance between the end mounts 120 and 122 can be changed. By changing the spacing between the end mounts 120 and 122, the twist length, or strength of the twist lens, may be adjusted while maintaining a constant twist angle. A reference light beam may be emitted to the module 100. The quality of the transmitted beam may be monitored to determine if evidence of torsional lenses remains. If distortion is detected, the twist length may be adjusted by adjusting the spacing between the end mounts 120 and 122. While the end mounts 120, 122 are adjusted to adjust the twist length so as to remove or mitigate any net optical distortion in the reference beam, the quality of the transmitted beam is continuous. Can be monitored.

  In other embodiments, module 100 may be constructed and arranged to allow in-situ length adjustment. In such an embodiment, after the final system integration of the beam delivery fiber system, the part to be passed or slid through is the housing 110 and the end mount 120, so that the final optimum adjustment is possible during the end-to-end test. 122 may be added. One embodiment takes advantage of the inherent fit within the housing 110 due to the slot 126 and squeezes the housing 110 into end mounts 120 and 122 similar to standard collet clamps found in machine shops. To use a small band clamp. This allows for fine alignment after installation and provides a relatively simple method of fixing the position for permanent or repetitive adjustment.

In yet another embodiment, the fiber optic module may be divided into three parts (first part, second part, and third part) having lengths L ₁ , L ₂ , and L ₃ , respectively. The first and third portions may each have a 45 ° twisted portion of the fiber. Together, the 45 ° twisted portions are equivalent to the 90 ° twisted portions. In such embodiments, the distortion compensation device may include one of the two all-glass strain compensation methods and devices described above. However, due to manufacturing tolerances, compensation is not sufficient and distortion may remain. By adjusting the twist lengths L ₁ and L ₃ , the twist angle at the two ends may vary while the twist angle remains constant. Thus, the reference laser beam may be emitted to the module. The quality of the transmitted beam may be monitored to determine if there is a residual twist lens effect. If the distortion is detected, for determining the value of the twist length and L ₃ which may be alleviated or removed optical distortion net in the reference beam, while the beam quality is continuously monitored, twist length L ₁ and L ₃ may be adjusted.

In yet another embodiment, the fiber optic twist module may be divided into three parts (first part, second part, and third part) having lengths L ₁ , L ₂ , and L ₃ , respectively. Each of the first and third portions may initially have a 45 ° twisted portion of the fiber. Together, the 45 ° twisted portions are equivalent to the 90 ° twisted portions. However, in this embodiment, the distortion compensator may be configured to compensate for distortion resulting from the twisted portion. This overcompensation can be eliminated by giving further twist to the second part and increasing the twist above 45 °. While a twist with a selected angle is being provided to the second part, the reference beam can be emitted to the module and the quality of the transmitted beam can be monitored. Thus, the twist in the second part may be adjusted while monitoring the quality of the transmitted beam to determine a sufficient twist in the second part to remove or mitigate the net distortion. As the second part of the fiber is twisted, the opposite ends of the fiber optic module have a twist angle at the third part of less than 45 ° and are complementary to the twist angle at the first part. Note that it must be securely held or fixed to ensure that it is.

  In the embodiment described above, once the torsion module is adjusted so that the net strain is relieved or eliminated, the mechanical structure is twisted to all lengths and to prevent compensation from changing during module use. Can be used to permanently hold the value of. In one embodiment, end mounts 120 and 122 may be coupled to housing 100, for example, to permanently fix the compensation for optimal alignment. Thus, the module can be used immediately when needed in a fiber-based system.

  The data and model plotted in FIG. 8 do not account for distortions other than the linear extension of the core end under the influence of the torsion. Thus, at the twist angle applied in the experiment that produced the data plotted in FIG. 8, there was no distortion other than a simple GRIN lens. However, in at least some embodiments, and in at least some situations, a twisted lens resulting from a large twist angle can include not only a simple flat GRIN lens, but also polarization distortion. When such polarization distortion is generated, such polarization distortion is substantially the same as the polarization state of the input beam with the polarization state of the beam exiting the twist module, but with the twist angle using at least one polarization element. May be compensated by adjusting to rotate. For example, it is well known in the art that the orientation of a linearly polarized beam may be rotated by an angle 2θ by passing the beam through a half-wave plate whose optical axis makes an angle θ with respect to the initial polarization. . In addition, a fractional wave plate may change linearly polarized light to elliptically polarized light or vice versa.

  In some embodiments and in some situations, a torsional lens resulting from a large torsion angle may also include additional higher order distortions. When such a strain is generated, the twist length may be increased to produce a twist rate in which higher order strains are relaxed or eliminated. However, if the twist length exceeds the allowable length of the module, the total length of twist is divided into a series of small twists added to the total required twist, and the fiber has a smaller module size. It can be folded between twists to provide. For example, if a 20 ° twist can be achieved without further distortion, a 90 ° twist may consist of four 20 ° twists and one 10 ° twist. Alternatives may include 5 18 ° twists or 6 15 ° twists. Thus, the module may be divided into a number of parts, and each part may include a twisted part added to the total required twist.

  Yet another embodiment of the torsional mode includes a twisted hollow core waveguide, as shown in FIG. This method avoids higher order distortions that can occur in rectangular cores composed of glass or other solid materials. The hollow waveguide may be made of two thin parallel flexible plates 151, 153 that reflect light propagating between them. The thin plates 151, 153 may be made of metal, bare glass, uncoated glass, glass coated with metal, or other materials that transmit light with little optical loss. This embodiment offers several advantages. First, the hollow waveguide is highly resistant to laser loss. Air or vacuum that fills the space between the plates has a very high damage threshold. When light radiation is incident on the reflecting surface at a large incident angle, the damage threshold of those surfaces also increases. Second, the plate may be specified to be much wider than the propagating beam itself. This avoids reflection of the beam from the slow axis end of the waveguide. Third, the thin plates 151 and 153 can be formed with a slow axial direction. As a result, the cross-sectional shape of the hollow core is corrected to either a concave shape or a convex shape. This design freedom can be used to fine tune the positive lens effect that compensates for torsional lenses. The corrector 108 can be used to stretch and collimate the beam size at the torsion module input, and again shrink the beam to match the slow axis width of the optical fiber at the module output. In one embodiment, the dimensions of the hollow core may substantially match the respective core dimensions of the HARC input and output fibers. With this configuration, it is possible to use a relatively simple interface optical system having a magnification of ˜1. However, if the hollow core propagation loss is greater than acceptable, the beam is stretched before entering the hollow core to take advantage of the waveguide propagation loss being inversely proportional to the third power of the waveguide dimension. good. For example, by doubling the fast shaft dimensions, the propagation loss is reduced by a factor of eight.

  Some beam delivery systems have a portion that is rotatable relative to a stationary bottom. That is, the fibers 10 and 11 may be connected to both a stationary object and a rotating object. In such a situation, the beam delivery fibers 10, 11 must be able to transition from a stationary object to a rotating object without twisting the beam delivery fibers 10, 11. In order to prevent twisting of the fibers 10, 11, the fibers 10, 11 may be grouped on a flat tape-like structure, for example to form an optical fiber device or structure 134 as illustrated in FIG. 12a. In the embodiment of FIG. 12a, six fibers are grouped or placed between two layers of polyimide (see FIG. 12b). Additional layers may be added to strengthen the fiber structure 134. Note also that fewer layers may be provided in the fiber structure 134. For example, in some embodiments, the fiber optic device structure 134 may include a single layer of fiber 10. The central fiber may be configured to suppress bending stresses. The fiber structure 134 may be sufficiently flexible to provide a “rolling loop”. The rotation loop will be described in detail later. This fiber structure 134 also makes it possible to prevent unintentional twists in the fiber by bending only about the fast axis of the rectangular fiber.

  FIGS. 11 a-c show an example of a “rotating loop” method using fiber structure 134 or (S) HARC fibers 10, 11. For example, in the case of a fiber structure 134 as illustrated in FIG. 11a, the inner portion with rotating curvature is provided to be concentric with the outer portion with fixed curvature of the beam director hardware. In the embodiment of FIG. 11a, the radius of the inner portion is 6 inches and the radius of the outer portion is 12 inches. These radii may vary to support a particular embodiment. These radii are limited only by the amount of bending that the resulting bending stress induces loss and / or mode coupling in the fiber. The tolerance to loss and mode coupling induced by such bending sets the minimum allowable bending between the inner and outer housings and in the “rotating loop” portion of the inner hub itself. The fiber structure 134 is mechanically constrained between two concentric portions 138 and 140. The fiber structure 134 may be connected to the inner portion 138 at point A and connected to the outer loop at point B. The fiber structure 134 may have a loop 142 formed between points A and B. The position of the loop 142 may change to maintain the length of the fiber structure 134 between points A and B during rotation of the inner portion 138. In one embodiment, the fiber structure 134 has a length of 1 m between points A and B. In one embodiment, the system has an azimuthal range of about ± 150 ° relative to the center of line of sight (line C). FIG. 11 a shows a “rotating loop” constituted by a fiber structure 134 that rotates the orientation of the inner portion 138 in the counterclockwise direction by + 150 °. As illustrated in FIG. 11a, the position of the loop 142 has moved approximately 50 ° from the position illustrated in FIG. 11b to provide rotation of the beam director. FIG. 11c has been moved −50 ° from the position shown in FIG. 11b to rotate the inner portion 138 in the −150 ° azimuth direction clockwise from the position shown in FIG. 11b. The fiber structure 134 and the “rotating loop” method using the fiber structure 134 allow the fiber 10 to be provided with a degree of freedom of fiber rotation without twisting the fiber 10.

  The fiber structure 134 allows the transition of the fiber with the first rotational degree of freedom described above. After a transition using the “rotating loop” method, the fiber optic twist module 100 may be coupled to a beam delivery fiber. Thereby, the beam supply fiber can provide an arbitrary bending required in the second degree of freedom. The configuration of a rotating loop using additional fiber optic modules 100 and / or fiber structures 134 may be provided to other portions of the fiber to provide further rotation and / or bending. Although the above embodiments have been described with reference to the fiber structure 134, similar embodiments may be provided for (S) HARC fibers 10,11.

  In this disclosure, the term “optical fiber” is used broadly to include any fiber capable of transmitting or carrying radiation in any wavelength range, including visible, ultraviolet, and infrared. For example, the fiber may be made of fused silica, but may be made of other fiber materials that provide the desired flexibility and desired transmission characteristics. For example, the fiber may be made of a chalcogenide material that has the desired transmission characteristics in the mid-infrared range.

  The above description has been given for purposes of illustration based on what is currently considered to be the most practical embodiment. However, such details are merely for that purpose, and the basic concept of the invention is not limited to the disclosed embodiments, but, in contrast, belongs to the spirit and scope of the claimed invention. Note that it is intended to cover modified and equivalent configurations. For example, it is noted that this disclosure allows one or more features of any embodiment to be combined with one or more features of other embodiments, where possible.

  In addition, since many modifications and variations will readily occur to those skilled in the art, it is not desirable to limit the basic concept of the invention to the exact configuration and operation described herein. Accordingly, all suitable modifications and equivalents must be considered as belonging to the spirit and scope of the basic concept of the invention.

Claims (52)

An optical fiber having a core with a relatively large area selected to increase the threshold for stimulated Raman scattering or stimulated Brillouin scattering, or both,
The core has a cross section extending at a high aspect ratio and has a first refractive index;
The core is narrow in the fast axis direction and wide in the slow axis direction, so that the fiber is mechanically flexible in the fast axis direction and mechanically in the slow axis direction. Having rigidity,
Optical fiber. A first signal clad and a second signal clad provided in contact with the core and having a second refractive index across the core; and
The optical fiber of claim 1, further comprising a third cladding having a third refractive index substantially surrounding at least a slow axis end of the core.
To provide a higher order transverse electromagnetic mode loss greater than a lower order transverse electromagnetic mode loss to substantially eliminate higher order transverse modes propagating along the slow axis within the fiber, The difference between the first refractive index of the core and the third refractive index of the third cladding is less than about 500 ppm;
Optical fiber.   The optical fiber of claim 1, wherein the width of the core along the slow axis and the fast axis is chosen to be greater than the width of the laser beam input to the fiber.   The optical fiber of claim 3, wherein the width of the core exceeds the width of the diffracted laser beam at the output end of the fiber. The width w _{b of the} laser beam is such that the Fresnel length L _Fr proportional to w _b ² n / λ exceeds the fiber length L,
n is the first refractive index of the core, and
λ is the wavelength of the laser beam input to the fiber,
The optical fiber according to claim 3.   The refractive index of the third cladding is such that the numerical aperture of the third cladding at the outer interface is 0.06 so as to allow laser radiation leaking into the third cladding to leak into the third cladding. The optical fiber according to claim 2, wherein the optical fiber is selected to be less.   The optical fiber of claim 2, further comprising a splice attached to the third cladding, wherein the refractive index of the splice is essentially equal to the refractive index of the third cladding.   The optical fiber of claim 2, further comprising a coating covering the third cladding, wherein the coating is configured to be transparent to laser radiation that leaks into the third cladding.   The optical fiber according to claim 2, wherein the third cladding substantially surrounds the core and the first signal cladding and the second signal cladding.   The optical fiber of claim 1, wherein the fiber is configured to transmit a laser beam having an output of about 10 kW or greater.   The optical fiber of claim 1, wherein the core has a substantially rectangular shape extending in the slow axial direction.   The fiber has a substantially rectangular outer shape extending in the slow axis so as to allow mechanical deflection in the fast axis and resist mechanical deflection in the slow axis. The optical fiber according to claim 10. The optical fiber of claim 1, further comprising a twisted portion that occurs over a fiber length near the output end of the fiber,
The fiber length near the output end of the fiber is specified to be shorter than the inherent strain length associated with the twist-induced focal length.
Optical fiber. The inherent strain length L _dist is proportional to w _core / δθ _F ,
δθ _F = (w _b / 2F) is a torsion-induced divergence angle, w _core is the width of the core, w _b is an input laser beam, and F = δL / φ ² is a torsion-induction associated with the torsion angle φ The focal length,
The optical fiber according to claim 13.   The optical fiber of claim 1, wherein the divergence of the beam from the input end to the output end of the fiber is preserved.   The optical fiber according to claim 1, further comprising a bent portion or a twisted portion bent in the fast axial direction, or both the bent portion and the twisted portion.   17. The optical fiber of claim 16, wherein the twisted portion is different from the bent portion so that the twisted portion maintains substantially the same beam quality throughout the fiber. The twisted portion of the fiber induces optical distortion inside the fiber;
The optical distortion is compensated by constructing means for compensating the optical distortion in the core of the twisted portion;
As a result, a transmission beam is obtained in which no net optical distortion remains after passing through the twisted part.
The optical fiber according to claim 16. The optical distortion is a negative lens inside the fiber, and
The negative lens is compensated by constructing a positive lens in the core of the twisted portion opposite to the negative lens power and having a lens power that is substantially equal;
The optical fiber according to claim 18. The optical distortion is a polarization distortion inside the fiber, and
The polarization distortion is compensated by means of at least one polarizing member;
The optical fiber according to claim 18.   The positive lens creates a slow axis refractive index change in the core, or thickens the core in the middle along the fiber axis and narrows near the end of the core 20. The optical fiber of claim 19, wherein the optical fiber is generated by forming a thickness profile of the core along the slow axis. The size of the negative lens induced by the twist is adjusted to counteract the positive lens;
As a result, a transmission beam is obtained in which no net beam quality distortion remains after passing through the twisted portion
The optical fiber according to claim 19.   23. The optical fiber of claim 22, wherein the twist-induced negative lens is adjusted by changing the length of the twisted portion.   The generation of stimulated Raman scattering and stimulated Brillouin scattering is dependent on the length of the fiber, the area of the core of the fiber, the output of laser radiation propagating in the fiber, or the medium dependent Raman gain, or a combination thereof. The optical fiber of claim 1, controlled by selection.   For a given output of the laser radiation and a given Raman gain, the core is large enough to prevent the stimulated Raman scattering or stimulated Brillouin scattering or both from occurring at fiber lengths greater than about 2 m. 25. The optical fiber of claim 24, configured to have a surface area. laser;
An optical fiber having an input end and an output end; and
An optical coupler configured to input radiation from the laser through an input end of the optical fiber;
An optical device comprising:
The optical fiber has a core with a relatively large area selected to increase the threshold of stimulated Raman scattering or stimulated Brillouin scattering or both,
The core has a cross section extending at a high aspect ratio and has a first refractive index;
The core is narrow in the fast axis direction and wide in the slow axis direction, so that the fiber is mechanically flexible in the fast axis direction and mechanically in the slow axis direction. Having rigidity,
Optical device. A first signal clad and a second signal clad provided in contact with the core and having a second refractive index across the core; and
27. The optical device of claim 26, further comprising a third cladding that substantially surrounds at least a slow axis end of the core and has a third refractive index.
To provide a higher order transverse electromagnetic mode loss greater than a lower order transverse electromagnetic mode loss to substantially eliminate higher order transverse modes propagating along the slow axis within the fiber, The difference between the first refractive index of the core and the third refractive index of the third cladding is less than about 500 ppm;
Optical device. The laser comprises a fiber laser, and
The optical coupler has a flat refractive index profile (GRIN) lens;
The plane of the refractive index profile is substantially parallel to the slow axis of the fiber, and
The optical coupler is configured to couple the fiber laser to an input end of the optical fiber;
27. The optical device according to claim 26.   29. The optical device according to claim 28, wherein the length of the optical coupler is selected to be substantially equal to a quarter pitch length of the flat GRIN lens.   27. The optical apparatus of claim 26, wherein the laser and the optical fiber are attached to opposing ends of the optical coupler so as to form a monolithic device.   27. The optical device of claim 26, wherein the laser and the optical fiber are attached to opposite ends of the optical coupler by using melt bonding.   27. The optical apparatus of claim 26, wherein the optical coupler is configured to reshape the laser beam into a collimated high aspect ratio elliptical beam for input to the optical fiber.   27. The optical device of claim 26, wherein the optical coupler comprises at least one free space lens configured to guide radiation from the laser to the optical fiber. An optical fiber having an input end and an output end; and
A first end cap connected to the output end of the optical fiber;
An optical device comprising:
The optical fiber has a core with a relatively large area selected to increase the threshold of stimulated Raman scattering or stimulated Brillouin scattering or both,
The core has a cross section extending at a high aspect ratio and has a first refractive index;
The core is narrow in the fast axis direction and wide in the slow axis direction, so that the fiber is mechanically flexible in the fast axis direction and mechanically in the slow axis direction. Having rigidity,
Optical device. The first end cap has an exit surface;
The laser beam input through the input end protrudes from the first end cap through the emission surface,
The exit surface is provided at a position related to the output end of the optical fiber,
In the certain position, the size of the laser beam in the fast axis direction is substantially equal to the size of the laser beam in the slow axis direction,
35. The optical device according to claim 34.   The exit surface of the first end cap changes the divergence of the laser beam along the fast axis direction without changing the divergence of the laser beam along the slow axis direction. 36. The optical device of claim 35, shaped to reduce to laser beam divergence.   37. The optical apparatus of claim 36, wherein the divergence of the laser beam along the fast axis is substantially equal to the divergence of the laser beam along the slow axis.   35. The optical apparatus of claim 34, further comprising a second end cap connected to the input end of the optical fiber, wherein the second end cap is configured to receive a laser beam. . Providing a fiber comprising a core having a relatively large area selected to increase the threshold of stimulated Raman scattering or stimulated Brillouin scattering, or both, wherein the core has a cross section extending at a high aspect ratio. The core is narrow in the fast axis direction and wide in the slow axis direction, so that the fiber is mechanically flexible in the fast axis direction. And mechanically rigid in the slow axis direction;
Emitting a laser beam into the fiber, the laser beam having a width that is narrower than a width of the core along the slow axis direction and the fast axis direction; and
Propagating the laser beam through the fiber while avoiding the generation of stimulated Raman scattering or stimulated Brillouin scattering or both;
A method of supplying a fiber with a high-power laser beam.   40. The method of claim 39, further comprising providing the laser beam to a desired location by bending or twisting or both bending and twisting the fiber along at least a portion of the fiber.   Bending and twisting the fiber bends the first portion of the fiber and maintains a second portion of the fiber different from the first portion so as to maintain substantially the same beam quality throughout the fiber. 41. The method of claim 40, comprising twisting.   Compensating optical distortion in the second portion of the fiber induced by twisting the second portion of the fiber by providing means for compensating the optical distortion to the core of the second portion. 42. The method of claim 41, further comprising obtaining a transmission beam that results in no net optical distortion remaining after passing through the twisted portion.   Compensating the optical distortion constructs a negative lens in the fiber with a positive lens having a lens power opposite and substantially equal to the lens power of the negative lens in the core of the twisted portion. 43. The method of claim 42, further comprising: compensating.   Adjusting the size of the negative twist-inducing lens to counteract the positive lens, so that essentially no net beam quality distortion remains after passing the twisted second portion. 44. The method of claim 43, further comprising obtaining a transmission beam.   45. The method of claim 44, further comprising adjusting the negative twist-inducing lens by changing a length of the twisted second portion. Providing a fiber comprising a core having a relatively large area selected to increase the threshold of stimulated Raman scattering or stimulated Brillouin scattering, or both, wherein the core has a cross section extending at a high aspect ratio. The core is narrow in the fast axis direction and wide in the slow axis direction, so that the fiber is mechanically flexible in the fast axis direction. And mechanically rigid in the slow axis direction;
Emitting a laser beam into the fiber, the laser beam having a width that is narrower than a width of the core along the slow axis direction and the fast axis direction; and
Propagating the laser beam through the fiber reduces any higher order transverse electromagnetic mode losses that are greater than the lower order transverse electromagnetic mode losses while mitigating the occurrence of stimulated Raman scattering or stimulated Brillouin scattering or both. Providing, substantially removing the higher order transverse electromagnetic mode along the slow axial direction in the fiber;
A method of supplying a fiber with a high-power laser beam.   47. The method of claim 46, further comprising providing a laser beam to a desired location by bending or twisting the fiber along at least a portion of the fiber, or both bending and twisting.   Bending and twisting the fiber bends the first portion of the fiber and maintains a second portion of the fiber different from the first portion so as to maintain substantially the same beam quality throughout the fiber. 48. The method of claim 47, comprising twisting.   Compensating optical distortion in the second portion of the fiber induced by twisting the second portion of the fiber by providing means for compensating the optical distortion to the core of the second portion. 49. The method of claim 48, further comprising obtaining a transmission beam that results in no net optical distortion remaining after passing through the twisted portion.   Compensating the optical distortion constructs a negative lens in the fiber with a positive lens having a lens power opposite and substantially equal to the lens power of the negative lens in the core of the twisted portion. 50. The method of claim 49, further comprising: compensating.   Adjusting the size of the negative twist-inducing lens to counteract the positive lens, so that essentially no net beam quality distortion remains after passing the twisted second portion. 51. The method of claim 50, further comprising obtaining a transmission beam.   48. The method of claim 47, further comprising adjusting the negative twist-inducing lens by changing a length of the twisted second portion.

Images