DE112005003885B3 - Large core holey fibers, fiber amplifiers or fiber lasers - Google Patents

Abstract

A multimode optical fiber for propagating the fundamental mode and a plurality of non-fundamental low-order modes of the multimode optical fiber, comprising:a cladding region having cladding features arranged in a matrix, the cladding features having an im have a substantially constant spacing Λ and an average size d, and the cladding features are arranged substantially in a plurality of rows N and in a uniform hexagonal array;a core region surrounded by the cladding region, the core region is formed in the matrix in place of a central cladding feature and in place of one or more inner rows of the cladding features of the hexagonal array such that the multimode optical fiber allows propagation of the plurality of non-fundamental low-order modes; andwherein the cladding region and the core region provide an effective refractive index for the fundamental mode and the plurality of low-order non-fundamental modes;wherein an effective refractive index difference between the fundamental mode and the low-order non-fundamental modes is established such that the mode spacing between the fundamental mode and the first non-fundamental mode is large enough to reduce mode coupling in the multimode optical fiber such that a single mode input signal in the fundamental mode is substantially single mode after propagation down the multimode optical fiber -signal is output in the fundamental mode.

Description

BACKGROUND

area

The present teachings relate to optical fibers in general and, more particularly, to singlemode and multimode holey fibers having large core dimensions and to devices and systems using such fibers, such as lasers and amplifiers.

Description of the prior art

A holey fiber is an optical fiber in which optical guidance is provided by an array of "holes" arranged around a core. Such holes can have a refractive index lower than that of the core and together form a cladding region (in the following description the term cladding region is referred to by the German translation "Mantelbereich"), which has an effective refractive index , which is also lower than that of the core.

In general, the term holey fiber is used here to denote a fiber that has a large number of cladding features (in the following description the term cladding features is referred to by the German translation "Mantel properties"), e.g. holes arranged in a matrix, wherein the cladding features may have a refractive index different from that of the matrix. Accordingly, light incident on the cladding features may be reflected or scattered off.

A photonic crystal fiber (PCF) is a holey fiber with the holes arranged in a regular fashion. In one arrangement, the cladding region may be formed by holes arranged in a hexagonal close-packed fashion and the core may be formed in the central region of such a hexagonally patterned cladding region. The core may be in the central region where a central hole or one or more inner "rings" of holes would otherwise be in the extended array of holes. PCFs formed in the manner described thus have a core dimension and a cladding region thickness.

As with traditional fibers, maintaining single mode propagation is desirable in many PCF applications. Conventional PCFs operating in the single-mode propagation regime have core dimensions and cladding region thicknesses that are similar to those of traditional non-holey single-mode fibers. A limitation of such single-mode fibers and conventional PCFs is that the core dimension is limited to a relatively small value. In such small core fibers, high optical power levels result at peak intensities that induce nonlinearities in the core and fiber properties. High performance applications of such fibers are therefore either impractical or not possible.

the end WO 0 239 159 A1 Microstructured single-mode optical fibers are known in which inner cladding features have normalized dimensions in the range 0.35 to 0.50 and outer cladding features have normalized dimensions in the range 0.5 to 0.9, where the normalization factor is a typical feature spacing . The fiber is further characterized by an inner cladding feature spacing of greater than 2.0 microns.

the WO 0 137 008 A2 relates to an optical fiber waveguide. The fiber optic waveguide includes a core region and a cladding region surrounding the core region. The cladding area includes an inner cladding area and an outer cladding area. The inner cladding region has a latticework of large diameter columns. The lattice of large diameter columns has a diameter d to pitch LAMBDA ratio, defined by the relationship d/LAMBDA, greater than or equal to 0.3. In 6 A fiber optic waveguide is shown with the innermost six columns omitted.

The publication Koplov, JP, Kliner, DAV, Goldberg, L.: Use of bend loss to obtain single-transverse mode operation of a multimode fiber amplifier. In: Conference on Lasers and Electro Optics (CLEO 2000). Technical Digest. Postconference Edition, TOPS Vol.39 (IEEE Cat. No.00CH37088), 2000, p. 286-287. - ISSN DOI: 10.1109/CLEO.2000.907021 relates to higher order mode filtering via fiber bending losses.

The publication Fermann, M.E.: Single-mode excitation of multimode fibers with ultrashort pulses. In: Optics Letters, Vol. 23, January 1, 1998, no. 1, p. 52-54 relates to single-mode excitation of step-index fibers that can propagate multiple modes.

U.S. 2003/165313 A1 relates to an optical fiber, the fiber comprising at least one core surrounded by a first outer cladding region, the first outer cladding region being surrounded by a second outer cladding region, the first outer cladding region having, in cross section, a number of first outer cladding features with a lower refractive index than any material surrounding the first outer cladding features, wherein for a plurality of the first outer cladding features the minimum distance between two nearest adjacent first outer cladding features is less than 1.0 µm or less than an optical wavelength of the light im Use is passed through the fiber.

SUMMARY

The invention is defined by the patent claims.

character list

• 1A and 1B Figure 12 shows schematically an exemplary hollow fiber having a cladding region defined by a plurality of "holes" or cladding features that facilitate the propagation of only a single optical mode through a relatively large core;
• 2A and 2 B show schematically a photonic crystal fiber (PCF) having a plurality of holes or cladding features arranged in a two-dimensional representation of an exemplary simple cubic array, showing the size of the holes, the spacing between the holes, and the number of layers or rows of holes and/or composition of holes can be varied to allow propagation of only a single optical mode in a relatively large core;
• 3A and 3B show schematically a PCF having a plurality of holes or cladding features arranged in a two-dimensional representation of an exemplary hexagonal close-packed array, wherein the dimension of the holes, the distance between the holes, the number of layers or rows of holes and/or the Hole composition can be varied to allow propagation of a single optical mode through a relatively large core;
• 4A-C show schematically various hexagonal close-packed arrangements of the 3A and 3B , which have various numbers of "rings" (i.e., rows or layers) of hole/cladding features distributed in an array around a relatively large core, where the number of rings can be used to estimate the leakage losses of different optical modes of an optical affect signal;
• 5A-C show schematically how the hole size can be varied to affect the leakage of different modes;
• 6A-C show schematically how the hole spacing can be varied to affect the leakage of different modes;
• 7 Figure 12 shows schematically a PCF having a relatively large core suitable for supporting propagation of only a single optical mode and further comprising an outer cladding region formed by a plurality of holes for supporting multiple modes;
• 8th Figure 12 shows schematically a PCF having a relatively large doped core region suitable for single mode propagation;
• 9A-F Figure 12 shows schematically various hollow fibers having asymmetric hole arrangements distributed around a relatively large core to facilitate polarization-maintaining (PM) single-mode propagation;
• 10A-D show schematically how, according to the invention, an effective core dimension of a PCF can be changed by "removing" a different number of inner "rings" or layers ten, such a core dimension can be used to allow propagation of selected modes through a relatively large core;
• 11 Fig. 12 shows schematically a PCF having a doped ring or annular portion placed around a hollow region of the core to provide gain;
• 12 shows loss as a function of the number of layers for the fundamental and first order modes;
• 13 shows loss as a function of inverse bend radius for fundamental and first order modes;
• 14 Figure 12 shows effective modal index as a function of inverse bend radius for fundamental and first order modes;
• 15 Figure 12 shows schematically two exemplary holes or cladding features of a PCF, where a quantity d represents the diameter of each hole and A represents the distance between the holes;
• 16A-C are schematic representations showing the operating characteristics of a PCF wherein in a plot of λ/Λ versus d/A singlemode and multimode propagation regimes are separated by curves representing different numbers of layers;
• 17 FIG. 12 schematically shows two exemplary hole features having different shapes and dimensions, and
• 18A and 18B are schematic representations showing the operating characteristics of a hollow fiber, in a plot of λ/Λ _avg versus d _avg /Λ _{avg singlemode} and multimode regimes are separated by curves representing different effective thickness for cladding regions.

DETAILED DESCRIPTION OF SOME EMBODIMENTS

These and other aspects, advantages, and novel features of the present teachings will become apparent from the following detailed description and the accompanying drawings. In the drawings, similar elements have similar reference numbers.

The present teachings relate generally to hollow fibers, photonic crystal fibers, and photonic bandgap fibers. For purposes of description herein, and consistent with current usage of the terminology, a hollow fiber is an optical fiber in which optical guidance is provided essentially by regularly or irregularly arranged cladding features or "holes". As used herein, holes refer to features located in a backbone material that have different optical properties, such as index of refraction, compared to the backbone material. The holes may be open, evacuated, or filled with air, gas, liquid, or other material, transparent or non-transparent, such as glass, a polymer, or other substance or material. A photonic crystal fiber (PCF) is a hollow fiber with a substantially regular array of holes. A photonic bandgap fiber (PBF) is a PCF with appropriate hole spacing and arrangement or cladding features that satisfies the Bragg condition at the operating wavelength and essentially provides optical guidance through photonic bandgap effects.

1A shows a segment of a hollow fiber 100 while 1B FIG. 1 schematically shows an exemplary cross-section of hollow fiber 100 along line 1B-1B. As shown in cross-section, the hollow fiber 100 includes a cladding region 102, which may be defined as an annular or annular region having an average inner radius ρ and an average outer radius R . The cladding region 102 includes a plurality of holes or cladding features 110.

The inner radius ρ further defines a core 106 of the hollow fiber 100. The core 106 may or may not be formed of the same material as that of the outer region 112 and/or the spaces between the holes 110 (the matrix in which the holes are formed will). In various embodiments, the core 106, the spaces between the holes 110 and the outer region 112 are formed of glass having a first index of refraction n ₁ .

In certain embodiments, the holes 110 may be formed from materials having a second index of refraction n ₂ less than n ₁ , including but not limited to vacuum, a gas, a mixture of gases, a second type of glass, or other solid or liquid material. (On the contrary As an alternative to this, various embodiments of the photonic bandgap fiber may include holes or cladding features 110 formed from materials having a second refractive index n ₂ that is greater than the refractive index of the core 106 or the surrounding substance.) As in FIG 1B As illustrated, the holes 110 can have various dimensions (exemplified by dimension di). Furthermore, the distances between the holes 110 can be different (exemplified by a distance Λ _j ). In addition, although the holes 110 are drawn as round, non-round holes can also be used.

One aspect of the present teachings relates to selectively manipulating a parameter or combination of parameters such as effective cladding region thickness (R-p), average hole size (d), average hole spacing (A), and hole placement. As described above, the holes can be formed from a different material. For purposes of portions of this specification, the holes will be assumed to have an index of refraction less than that of the core, although the holes need not be so limited and may have an index of refraction greater than that described above. By manipulating the above properties of a hollow fiber, one can selectively control the leakage of the different modes of an optical signal propagating through the core. For example, in various embodiments, one mode is propagated with significantly less leakage while substantially all other modes propagate with significantly greater leakage.

As used herein, singlemode and multimode fiber are defined consistently with the definitions used for traditional non-hollow fiber. For traditional fibers, singlemode and multimode fibers are generally defined in terms of V number, which is equal to π (numerical aperture) (core diameter)/wavelength for step-index fibers. For non-step-index fibers, numerical aperture and core diameter can be calculated using the equivalent values for step-index fibers [see, e.g., Martinez, F., Husey, CD, “(E)ESI Determination from mode-field diameter and refractive index profile measurements on single-mode fibers “ IEEE Proceedings V135, pp. 202-210, (1988)]. For fibers that satisfy the relationship V < 2.4, the fundamental mode power is significantly greater than the optical power of the next higher mode. Alternatively, for fibers where V > 2.4, at least the next mode above the fundamental mode can have significant performance compared to the fundamental mode. Singlemode and multimode traditional fibers are thus specifically defined by the relationship V<2.4 and V>2.4, respectively. V = 2.4 is the truncation of propagation of any mode other than the lowest order mode.

In hollow fibers, the numerical aperture can be determined by the difference in refractive index between the core and cladding. However, a core diameter that is an equivalent value for step-index fibers is difficult to calculate. Various references [see for example (1) Knight et al, "Properties of photonic crystal fiber and the effective index model" J. Opt. Soc. At the. A Vo. 15, pp. 748-752, (1998), and (2) Mortensen et al "Modal cutoff and the V parameter in photonic crystal fibers" Opt. Lett. V. 28, pp. 1879-1881, (2003)] report that if the core diameter is made equal to the distance between the holes Λ, then the V parameter for cutting off the propagation of any mode other than the single mode is equal 2.5 (see, for example, Knight et al.) and π (see, for example, Mortensen et al.). For the various embodiments described herein, it is not critical whether the clipping V is 2.405, 2.5, or π. Various embodiments of the hollow fiber described herein have a much larger core radius than is possible with conventional optical fiber supporting propagation of a single optical mode. Therefore we will use the current research in this technical field, where a multimode fiber is defined when V > π and the core diameter equals the distance or average distance to the fiber. Conversely, a single mode fiber is defined as a fiber where V < π.

As described above, a hollow fiber can be designed to introduce loss for specific modes. For example, the hole size, spacing, and number of holes can be chosen to induce loss in the propagation of higher order modes in a multimode fiber where V > π.

With reducing the number of holes, light in the higher order modes cannot be trapped in the core and can escape from the fiber. Such a loss introduced in a multimode fiber (V > π) is analogous to a traditional non-holey multimode fiber having a V number greater than π, which includes modal filters, for example provided by bending the fiber to introduce loss in the propagation of higher order modes. (Mode filters are described for example in the US patent 5 818 630 , issued to Fermann et al. on October 6, 1998 and entitled "Single-mode Amplifier and Compressors Based on Multi-mode Fibers," which is incorporated herein by reference.) Sufficient bending can be applied to induce losses for any of the modes higher than the fundamental mode is such that the fundamental mode is the only mode propagating through the bent multimode fiber. Similarly, a multimode hollow fiber that has a V number greater than about π can have a design that introduces loss for higher order modes such that the propagation of these higher order modes is attenuated.

the 2A and 2 B and subsequent figures depict various hollow fibers having regularly spaced holes in various configurations. Accordingly, these holey fibers are now referred to as PCFs (Photonic Crystal Fibers), consistent with the definition set out above. It will be understood that, unless otherwise stated, the concepts disclosed with respect to various PCFs also apply to an aperiodic structure, such as the exemplary hollow fiber discussed above with respect to FIG 1A and 1B has been described, as well as asymmetric and other structures that have varying degrees of order. Furthermore, the various geometric arrangements of the holes in the PCFs herein are exemplary and should not be construed to limit the scope of the present teachings, as a wide range of variations in configurations, placement, and feature characteristics are possible.

2A and 2 B Figure 12 shows an exemplary PCF 120 having a plurality of holes 130 arranged in a two-dimensional representation of a simple cubic array. A core 126 can be defined by an area free of holes, which is indicated by the absence of a central hole (distant from 2 B ) is generated in a simple cubic arrangement. Accordingly, 2p is a measure of core size, which corresponds to the diameter of the central hole, which is absent from the assembly. An exemplary single-layer cladding region 122 may be located between an inner ring 124a (radius p) and an outer ring 124b (radius R) surrounding the first layer of holes 130, for example. More layer(s) of holes can be found beyond the first layer that is in 2 B shown can be arranged. Furthermore, a core can be altered by removing more than the central hole (in which case the number of rows or layers in the array would otherwise be greater than one).

The exemplary cladding region 122 provides the core 126 whose diameter 2p is comparable to hole dimension d and hole spacing A . By selecting a relatively large hole dimension, and hence a relatively large hole spacing, the corresponding core diameter can be relatively large. Propagation of only a single optical mode can be achieved by selectively adjusting the values of d, A and/or the number of layers.

3A and 3B FIG. 14 shows an exemplary PCF 140 having a plurality of holes 150 arranged in a two-dimensional representation of an exemplary hexagonal close-packed array. A core 146 may have a region that overlaps with an absent central hole formed in 3B and removed from the hexagonal array. A first layer 142 includes six holes 150 arranged in a hexagonal fashion and defines the core 146 which has a radius ρ. As described below, more layer(s) or rows of holes can be placed beyond the first layer 142, which is in 3B will be shown. Furthermore, a core can be defined by removing more than the central hole (in which case the number of layers or rows in the array would otherwise be greater than one).

The example cladding region 142 provides the core 146 whose diameter 2p is generally comparable to the hole dimension d and hole spacing A. By choosing a relatively large hole dimension, and hence a relatively large hole spacing, the corresponding core diameter can be relatively large. Single mode propagation, i. H. Propagation of only a single optical mode to the exclusion of sustaining the propagation of higher order modes can be achieved by selectively adjusting the values of d, A and/or the number of layers.

The example fiber used in 3B and other hollow fibers disclosed herein can be formed, for example, by stacking glass capillaries to construct a preform. One or more capillaries in the center can be replaced with a glass rod/wires to form the core. In some embodiments, as described below, the wire(s) forming the core may be doped with rare earth ions to provide a gain medium. The preform can then be drawn into similar cross-section fibers in a conventional fiber drawing tower. Other methods of forming hollow fibers can also be used.

In one embodiment, the large core design disclosed herein has a substantially uniform refractive index distribution, particularly across the core, preferably across the cross-section of the fiber (in the cladding region, gap regions between holes). When the core diameter is in the range of about 40 to 100 µm (such as those disclosed herein), this uniformity becomes increasingly important to essentially eliminate any local waveguide effect within the core of the fiber.

4A-C now show various embodiments of a PCF having cladding regions formed by hexagonally arranged holes or cladding features. In the example PCFs provided in the 4A-C are shown, core dimensions are comparable to hole spacing and hole dimensions. Thus, by selecting a relatively large hole size and spacing, a relatively large core size can be achieved.

One way to design such a relatively large core is to begin by determining the desired size of the core. A hexagonal close-packed array of holes or cladding features can then be arranged, where the hole diameter d is substantially equal to the hole spacing A. To create a hole, the center hole (and possibly more holes around the center hole) can be removed. The resulting "shell" region comprises one or more rings or rows of closely packed holes, with adjacent holes being substantially close or touching. Since such dense packing can reduce higher order mode leakage, leakage "paths" can be provided and leakage can be increased by reducing hole size, increasing hole spacing, or a combination thereof so that, for example, the adjacent holes no longer touch or are so close together.

4A FIG. 12 shows a PCF 160 having a cladding region 162 formed by a layer or row (N=1) of holes 170. FIG. An inner portion 164a of cladding region 162 generally defines a core 166 having a radius ρ and an outer portion 164b of cladding region 162 has an outer diameter R. In the PCF 160 shown in FIG 4A As shown, the hole diameter d is smaller than the hole spacing A. In one embodiment of an N=1 PCF, the hole diameter d is about 75 µm and the hole spacing A is about 80 µm, resulting in a core diameter 2p (=2A-d). , which is about 85 µm.

4B FIG. 12 shows a PCF 180 having a cladding region 182 formed by two layers or rows (N=2) of holes 170. FIG. An inner portion 184 of cladding portion 182 generally defines a core 186 having a radius ρ and an outer portion 184b of cladding portion 182 has an outer diameter R. In PCF 180, shown in FIG 4b is shown, the hole diameter d is smaller than the hole spacing A. In one embodiment of an N=2-PCF, the hole diameter d is approximately 35 µm and the hole spacing is approximately 40 µm, resulting in a core diameter 2p (=2A-d) that is approximately is 45 µm.

4C FIG. 12 shows a PCF 200 having a cladding region 202 formed by three layers (N=3) of holes 210. FIG. An inner portion 204a of cladding portion 202 generally defines a core 206 having a radius ρ and an outer portion 204b of cladding portion 202 has an outer diameter R. In the PCF 200 shown in FIG 4C As shown, the hole diameter d is smaller than the hole spacing A. In one embodiment of an N=3 PCF, the hole diameter d is about 15 µm and the hole spacing A is about 20 µm, resulting in a core diameter 2p (=2A-d). , which is about 25 µm.

Table 1, as shown below, summarizes various properties of those referred to above 4A until 4C described PCFs together in comparison with various conventional PCFs described in Philip Russell: "Photonic crystal fibers", Science, Vol. 299, pages 358-62, 2003, Bouwmans et al.: "High-power Er: Yb fiber laser with very high numerical aperture pump-cladding waveguide”, Applied Physics Letters, Vol. 83, pp. 817-18, 2003, Furusawa et al.: “Cladding pumped Ytterbium-doped fiber laser with holey inner and outer cladding”, Optics Express, Vol. 9 , pp. 714-20, 2001, and Wadsworth et al.: "High power air clad photonic crystal fiber laser," Optics Express, Vol. 11, pp. 48-53, 2003. In this table, the "propagated modes" are either a single mode (SM) or multimode (MM), depending on whether only a single optical mode propagation is supported or whether many optical modes are propagated through the fiber, N corresponds to the number of rings or rows (layers), d is the hole diameter, A is the hole spacing, 2p is a measure of the core diameter, which is approximately equal to 2A-d in cases where the core corresponds to a single distant central hole of a hexagonal close-packed array, λ is the operating vacuum wavelength, Δn is the modal spacing (both measured and calculated), NA is the numerical aperture (measured and calculated), and V is the normalized frequency parameter given as (2ρ)(NA)π/λ. can be expressed. Reference [1] corresponds to Philip Russell: "Photonic crystal fibers", Science, Vol. 299, pages 358-62, 2003; Reference [2] corresponds to Bouwmans et al.: "High-power Er: Yb fiber laser with very high numerical aperture pump-cladding waveguide", Applied Physics Letters, vol. 83, pp. 817-18, 2003; Reference [4] corresponds to Furusawa et al.: "Cladding pumped Ytterbium-doped fiber laser with holey inner and outer cladding", Optics Express, Vol. 9, pp. 714-20, 2001; and reference [5] corresponds to Wadsworth et al.: "High power air clad photonic crystal fiber laser," Optics Express, Vol. 11, pages 48-53, 2003, each of which is incorporated herein by reference in its entirety.

In Table 1 one can see that the conventional single-mode PCFs have a relatively small value of d/A, which corresponds to a relatively small core size. Furthermore, their cladding region contains a relatively large number of rows or layers of holes. Thus, in these conventional PCFs that propagate a single optical mode, the cladding region generally resembles the cladding (e.g., a slightly lower refractive index glass) of a traditional non-holey single-mode fiber.

Table 1 further shows details of two exemplary large-core PCFs that propagate only a single optical mode, denoted as " 4B " and " 4A ' in the reference column. These two PCFs involve different designs and techniques, which are described here. Some techniques described below were used to obtain the numerical simulation for these two example PCFs.

The measured index difference Δn _meas can be obtained from the measured numerical aperture (NA) of the fiber output. The precision of such a measurement may be less than ideal. Since this value is a useful parameter for comparison, the index difference was calculated for each case from the given d and A. These calculated values can be more accurate than those obtained by direct measurement.

The calculated An _cal can be obtained by first calculating the equivalent cladding index, which is the effective index of the fundamental mode of a structure having the basic cladding unit cell extending infinitely in all directions. For each case, this calculation was performed using a finite element model with sufficient resolution. The value of An _cal is then determined by calculating the difference in the glass refractive index and the equivalent cladding index.

As stated above, 2p is the measure of an equivalent core diameter and is taken as 2A-d with a single hole (e.g. a removed wire) forming the core. In cases where three absent wires or three holes form the core, the value of the core diameter can be obtained, for example, from Mortensen et al.: "Improved large-mode-area endlessly single-mode photonic crystal fibers", Optics Letters, Volume 28, pages 393-395, 2003. Table 1 ref Spread Fashions N d (µm) Λ (µm) d/Λ (µm) 2ρ (µm) λ (µm) Δn _meas (Δn _cal ) NA _meas (NA _cal ) V _meas (V _cal ) [1] SM 5 2.7 9.7 0.3 16.7 1.06 0.0008 (0.00088) 0.048 (0.0505) 2.38 (2.50) [2] SM 4 2.65 8.8 0.3 15 1 0.0042 (0.001) 0.11 (0.054) 5.2 (2.58) [2] mm 4 5.7 10.4 0.55 15 1 N/A (0.0021) 0.056 (0.078) N/A (3.67) [4] SM 6 2 11.5 0.18 28 1.07 0.00086 (0.00042) 0.05 (0.035) 4.1 (2.87) [5] MM: 3-4 9 12 0.75 17 1.06 0.0017 (0.0051) 0.07 (0.12) 3.5 (6.13) 4C SM 3 15 20 0.75 25 - - - - 4B SM 2 35 40 0.87 5 45 1.55 N/A (0.0025) N/A (0.0854) N/A (7.8) 4A SM 1 75 80 0.93 8 85 1.55 N/A (0.001106) N/A (0.0566) N/A (9.75)

As stated above with reference to Table 1, the V parameter is the normalized frequency parameter (also referred to as the V number), which can be approximated by (2ρ)(NA)π/λ. V _meas and V _cal correspond to measured and calculated V values, respectively. In a traditional fiber, V<2.405 corresponds to the range of a single mode fiber. As can be seen in Table 1, for all of the PCFs in the first five rows that only support single mode propagation, the calculated V numbers are less than 2.405 (V _cal <2.405) or very close to it. In comparison, the calculated V numbers for N=2 ( 4B ) and N=1 ( 4A ) based on the designs and techniques described here and supporting propagation of only a single optical mode, 7.8 and 9.75, respectively. Such large V values would be associated with multimode fiber in traditional non-hollow optical fiber as well as multimode propagation in conventional PCFs.

In all of the single-mode PCFs in the first five rows of Table 1, d/A<0.6, whereas the N=2 and N1=1 embodiments of the PCFs based on designs and techniques described herein have d/A have >0.6. The relative hole diameter of the PCFs described here is therefore much larger than that of conventional PCFs. In some embodiments, the fiber contains a capital A for a large core diameter and λ/Λ<0.3. The hole size d in such a region of interest can also be very large compared to the wavelength λ and the gaps (e.g. glass matrix region) between holes and scattering features.

In one possible operating range, as described above, single-mode operation can be achieved (supporting only a single optical mode and suppressing higher-order modes). High leakage is provided for higher order modes while the fundamental mode leakage is kept very low. A PCF can be designed to allow substantial power loss for the higher order modes over the device length while maintaining an acceptably low transmission loss for the fundamental mode. Some possible ways to optimize the leakage by adjusting the number of layers of holes (N) and the hole properties are described in more detail below.

For the large-core single-mode PCF embodying the designs and techniques described herein, a preferred operating range is d/Λ ≥ 0.35 with N ≤ 3. More preferred d/ Λ ≥ 0.4 or d/ Λ ≥ 0.6 . Even more preferably d/Λ ≥ 0.75. The core sizes can be between 20 to 300 microns, or more preferably between about 30 to 150 microns. In some exemplary embodiments of a hollow fiber or PCF d/Λ>0.6, N<4 and λ/Λ<0.3. In other embodiments of a hollow fiber or PCF e.g. d/ Λ > 0.75, N ≤ 2 and λ / Λ < 0.3. The range of different fiber designs should not be limited to these parameters, however, as wide variation is possible.

5 and 6 now show how, for a given number of layers or holes, leakage can be changed by adjusting hole size and hole spacing. While an N=2 array for describing 5 and 6 is used, it goes without saying that the concept is also valid for other configurations of N.

5A-C show three exemplary PCFs 240, 250, 260, which have a substantially constant hole spacing A. As in 5A As shown, a first exemplary cladding region 244 includes two layers of holes 242 having a first diameter d ₁ given a hole spacing A are arranged. Such a configuration results in a first core 246 having a first core diameter of 2ρ ₁ and a first gap dimension (between two adjacent holes) of Λ-d ₁ .

As in 5B As shown, a second exemplary cladding region 254 includes two layers of holes 252 having a second diameter d ₂ and spaced A apart. Such a configuration results in a second core 256 having a second core diameter of 2ρ ₂ and a second gap dimension (between two adjacent holes) of Λ-d ₂ .

As in 5C As shown, a third exemplary cladding region 264 includes two layers of holes 262 having a third diameter d ₃ and spaced A apart. Such a configuration results in a third core 266 having a third core diameter 2ρ ₃ and a third gap dimension (between two adjacent holes) of Λ-d ₃ .

In the 5A-C it can be seen that as the hole diameter decreases (d ₁ > d ₂ > d ₃ ), the gap between the holes increases (Λ-d ₁ < Λ-d ₂ < Λ-d ₃ ), allowing larger leakage losses. One can also see that the core size increases (ρ ₁ < ρ ₂ < ρ ₃ ) as the hole diameter decreases. Thus, one can choose a desired gap (and possibly its corresponding core size) by choosing a hole size given a hole spacing.

6A-C 12 now show three example PFCs 270, 280, 290 that have a substantially constant hole dimension d. As in 6A As shown, a first exemplary cladding region 274 includes two layers of holes 272 of diameter d, spaced at a first hole spacing Λ ₁ . Such a configuration results in a first core 276 having a first core diameter 2ρ ₁ and a first gap dimension (between two adjacent holes) of Λ ₁ -d.

As in 6B As shown, a second exemplary cladding region 284 includes two layers of holes 282 of diameter d spaced at a second hole spacing Λ ₂ . Such a configuration results in a second core 286 having a second core diameter of 2ρ ₂ and a second gap dimension (between two adjacent holes) of Λ ₂ -d.

As in 6C As shown, a third exemplary cladding region 294 includes two layers of holes 292 of diameter d spaced at a third hole spacing Λ ₃ . Such a configuration results in a third core 296 having a third core diameter 2ρ ₃ and a third gap dimension (between two adjacent holes) of Λ ₃ -d.

From the 6A-C it can be seen that as hole spacing increases (Λ ₁ < Λ ₂ < Λ ₃ ), the gap between holes also increases (Λ-d ₁ < Λ-d ₂ < Λd ₃ ), allowing larger leakage losses. One can also see that the core size increases (ρ ₁ < ρ ₂ < ρ ₃ ) as hole spacing increases. Thus, one can select a desired gap (and possibly its corresponding core size) by choosing a hole spacing given a hole dimension.

The above with reference to 5 and 6 Leakage adjustments described above are made by independently adjusting hole size and hole spacing. One can also adjust leakage by adjusting hole size and hole spacing in combination to achieve a desired leakage property. Furthermore, hole size and spacing may vary from layer to layer to achieve a desired result.

7-9 now show various PCFs containing relatively large cores and suitable for single mode propagation in which only a single mode is propagated and propagation of higher order modes is suppressed. 7 12 shows an exemplary PCF 300 having a cladding region 302 that allows single-mode propagation through a relatively large core 304. FIG. The PCF 300 further includes an outer jacket 306 disposed within an outer portion 312 of the PCF 300 . The exemplary outer shell 306 includes a plurality of holes 310 arranged in a hexagonal shape. In one embodiment, the hexagonal shape of outer shell 306 is substantially concentric with the hexagonal shape of shell region 302.

Such a "double clad" fiber can be used, for example, in fiber amplifiers, lasers, continuous wave and pulsed light sources, or in other devices and systems where the fiber is pumped. An outer cladding formed of holes can provide a number of advantageous features over conventional double-clad fibers. A possible advantage is that a polymer layer with low index of the conventional double-clad fiber can be eliminated. Low-index polymer layers are relatively easily damaged by heat and high optical power, potentially posing reliability and stability issues for high-performance applications. Another possible advantage is that a relatively high numerical aperture can be achieved compared to conventional double-clad fibers. This feature allows for a smaller pump waveguide and consequently improved overlap of pump and laser light, potentially resulting in shorter device length. A shorter device, in addition to being more compact and economical, also reduces non-linear effects that limit the peak power of lasers and amplifiers.

8th FIG. 12 shows another PCF 320 having a cladding region 22 that allows single-mode propagation through a relatively large core region. The PCF 320 further includes a doped core 324 located in the core region. Such a doped core PCF may include an outer cladding formed by a plurality of holes disposed in an outer region 326 around the cladding region 322 . In one embodiment, the doped core 324 comprises a rare earth-doped core that allows the PCF 320 to be used as a gain medium, such as in a fiber laser or a fiber amplifier.

9A-D show various PCFs that have cladding regions that allow single-mode propagation through relatively large core areas. The cladding regions include asymmetries that facilitate polarization maintenance of single polarized mode optical signals propagating through the respective cores.

9A FIG. 3 shows an example PCF 330 having a cladding region 332 that allows single-mode propagation through a relatively large core area. In the exemplary PCF 330, the core area is populated with a doped core 340 to provide a gain medium. The cladding region 322 includes a plurality of first holes 334 forming one or more layers. The cladding region 332 includes an asymmetry introduced through second holes 336 that are distinct from the first holes 334 . In one embodiment, the second holes 336 are smaller than the first holes 334 and are located on opposite sides of the core surface, thereby allowing a preferential direction of the electric field of the optical signal.

9B 12 shows an exemplary PCF 350 having a cladding region 332 (similar to FIG 9A ), allowing single-mode propagation through a relatively large core 360. The cladding region 332 and the asymmetry therein allowed polarization-preserving single-mode propagation through the core 360.

9C FIG. 3 shows an exemplary PCF 370 having a cladding region 372 that allows single-mode propagation through a relatively large core area. Similar to the PCF 330 out 9A , the exemplary PCF 370 has a doped core 380 to provide a gain medium. Asymmetry of the cladding region 372 is achieved by smaller holes 376 (smaller than holes 374) arranged around the core 380, thereby allowing a preferential direction of the electric field of the optical signal.

9D 12 shows an exemplary PCF 390 having a cladding region 372 (similar to FIG 9C ), allowing single-mode propagation through a relatively large core 400. The cladding region 372 and the asymmetry therein allow polarization-preserving single-mode propagation through the core 400.

As in 9E and 9F As shown, the location of the first larger set of holes 334 and the location of the second smaller set of holes 336 can be selected such that the distance (s ₁ ) of the larger holes 334 is approximately equal to the distance (s ₂ ) of the smaller holes 336 ( _s1 ≈ _s2 ). (These distances s ₁ , s ₂ and generally the position of the holes 334, 336 can be measured, for example, from the center of the holes 334, 336.) This configuration provides a more circular fundamental mode, which is desirable in many applications. In various preferred embodiments, the positions of these holes 334, 336 are at lattice points corresponding to a lattice (e.g., a generally hexagonal lattice) having an average spacing A, as described above. Alternatively, the small holes 336 may be closer (see the dashed holes in Fig 9E ), or the large holes 334 (see the dashed hole in 9F ) may be further away. 9E Figure 12 shows the pinholes 336 shifted with respect to the grating positions and particularly towards the center of the fiber. In contrast, shows 9F the large holes 334 shifted with respect to the grating positions and in particular away from the center of the fiber. Other configurations are also possible.

The various PCFs referred to above 2-9 have cores that occupy areas where the corresponding central holes would otherwise be located. It is understood that a relatively large core can also include surface(s) in other areas. For example, two adjacent holes of the first layer can be removed to yield a triangular shaped core that is slightly eccentric to the geometric center of the hexagonally shaped cladding region. In another example, the entire first layer can be removed to provide a larger core that is similar in dimension to the outer edge of the first layer.

Whether such enlarged cores can support single-mode propagation depends on the extent of leakage of the non-fundamental (higher order) modes relative to the loss of the fundamental modes for a given length of the PCF. Various properties of the hollow cladding described above (such as hole size, hole spacing, number of layers, hole composition, etc.) can be adjusted to achieve single mode propagation through the enlarged core. Alternatively, different properties of the hollow cladding can be selected to allow propagation of some selected modes of an optical signal.

As used herein, "optical signal" refers generally to propagated optical energy or power and is not limited to an optical carrier of voice, data, or information. For example, the optical signal may correspond to light propagating through an optical fiber amplifier or the output of a fiber optic laser or other light source.

The claimed invention relates to the propagation of a single-mode signal through a multi-mode supporting fiber. Since "mode selection" in hollow fiber depends on the relative leakage losses of the different modes, the hollow fiber can be configured to allow the propagation of different non-fundamental modes if the cladding region is configured to accommodate those modes higher To "trap" order (ie not to allow substantial loss through leakage). Such a hollow fiber for propagating many modes (e.g. about 20 modes or more) or "few" modes can be made, for example, by removing one or more inner layers, as in 10A and B shown.

In 10A a PCF 410 according to the invention comprises a cladding region 412 which is formed by a hexagonal arrangement of holes 414 .

In accordance with the invention, a core 420 is formed in an area where the center hole and first layer of holes (not shown) would otherwise be in the array. The PCF 410 may further include an outer jacket disposed in an outer region 416 .

In 10B An example PCF 430 includes a cladding region 432 formed by a hexagonal array of holes 434 . In one embodiment, a core 440 is formed by additionally removing a second layer of holes. The example PCF 430 may further include an outer jacket disposed in an outer region 416 . Thus, one can see that the resulting core 440 (and the core 420 from 10A ) can be relatively large to allow handling of high power optical signals without reaching excessive intensity levels causing optical non-linear effects that could impair operation.

10C and 10D now show how a single-mode signal can be propagated over a given length of a PCF that supports few optical modes. In 10C For example, one embodiment of a PCF 450 that supports "few" modes includes two (exemplary) layers 454A, 454B of holes forming a cladding region 452 around a relatively large core 460. FIG. Although an outer mantle 456 in 10C is shown, such a feature may not be desired in some embodiments.

As in 10C As shown, a single-mode signal 464 with no higher-order modes is introduced from a signal source 462 into the core 460 of the PCF 450 . The single mode signal 464 is shown to survive substantially the length L of the PCF 450 to emerge as a single mode signal 466 . Single-mode signal survival can be facilitated by relatively large PCF model spacing. PCFs supporting many optical modes can be designed to have a large modal spacing between the fundamental mode and the next higher mode, greater than about 0.001, while at the same time having a large core (potentially hundreds of microns in diameter). . The large modal spacing (model spacing) reduces coupling between the fashions. Accordingly, from a signal of a fundamental mode (464 in 10C ) introduced into the PCF, which supports propagation of many modes, expects to be robustly guided in that single mode over a given length of the PCF.

A variety of techniques can be used to introduce a single mode into fiber that supports multiple optical modes. For example, an optical coupler can be placed with respect to the core region to couple light into the core region. The optical coupler can be configured to output light having an optical distribution that substantially matches the modal profile of the fundamental mode. More of the light is thereby essentially coupled into the signal mode than into the higher order modes. The optical coupler may, for example, comprise a lens or other optics having appropriate characteristics (e.g. optical power, numerical aperture, etc.) positioned with respect to the core region to focus light down to an extent equal to the size of the mode profile corresponds to the fundamental mode. Alternatively, the optical coupler may include a waveguide that outputs light having a distribution that conforms to the mode profile of the fundamental mode. Other techniques can also be used.

10D FIG. 4 shows a block diagram of a system 470 for propagating an optical signal that reduces a high power multimode signal 472 having, for example, 20 or more optical modes to a single mode signal 486. FIG. One or more segments of a PCF are used that support "few" optical modes, as referred to above 10A-C described. The exemplary signal 472, which has many modes, is shown entering a first PCF 474, which has a first length L ₁ , which supports few optical modes. After traversing the first length L ₁ , the multimode signal 472 is reduced to a sparse mode signal 476 as modes other than the sparse modes are lost through leakage. The "low" mode signal 476 is then shown passing through a filter 480 that filters out non-fundamental modes from the low mode signal 476 . The filter 480 can be configured to substantially eliminate power in the higher order mode(s) while allowing the fundamental mode to pass. Such mode filtering can be implemented by devices and/or techniques such as tapers, bends, gratings such as long period gratings to couple higher order modes into leaky modes, and gain/loss profiling. Mode filters are described in US patent 5 818 630 issued to Fermann et al on October 6, 1998 and entitled "Single-mode Amplifier and Compressors Based on Multi-mode Fibers" (Attorney Docket IM-59), which is incorporated herein by reference.

How further in 10D As shown, a single mode signal 482 emerges from the filter 480. FIG. The single-mode signal 482 can then be passed through a second PCF 484 (length L ₂ ) that supports few-mode propagation to result in the output single-mode signal 486 .

Another aspect of the present teaching relates to a PCF having a plurality of holes arranged in an array or grating that satisfies the Bragg condition for a given signal wavelength. In accordance with the description presented above, such a PCF is referred to as a PBF (photonic bandgap fiber). 11 FIG. 4 shows an exemplary PBF 490 having an array 492 of holes 494 disposed on a fibrous material 476. FIG. The array 492 defines a core area 500 through which the signal propagates.

Because a PBF relies on Bragg reflection or scattering rather than an average refractive index difference, the core can have a low refractive index. Consequently, the core can be evacuated or filled with air or other gases or media that have very small non-linear effects. Such an advantageous feature can lead to a significant reduction in nonlinear effect, and consequently very high optical power can be propagated in these PBFs without suffering from nonlinear distortion.

In one example, the core 500 further includes a doped annular or annular region 502 disposed about the center of the core. There 11 Figure 12 is a cross-sectional view of the fiber, doped region 502 may have a doped portion that is in the shape of a hollow, regular, and circular cylinder. Accordingly, an open interior region 504 of the doped ring 502 can be filled with air or other gases to reduce the non-linear effect of the PBF. Various other configurations are possible. For example, the doped region may be linear and have a square or rectangular cross-section when viewed from the angle described in 11 is presented, is viewed. Alternatively, the cross-sectional view of the doped region may be triangular, pentagonal, hexagonal, octagonal, or any of a wide variety of regular or irregular shapes. This area also need not be contiguous and may include lobes or segmented portions. For example, the doped region may comprise two or more parts that are opposite each other with respect to the hollow core. Still other variations are possible. However, in various embodiments, the doped region 502 in the PBF core can provide gain for lasers and amplifiers, and such lasers and amplifiers can have very high outputs.

12-14 show various properties of hollow fibers and computer simulation results characterizing the fundamental and first-order modes. 12 Figure 5 shows a loss profile 510 for the fundamental and first order modes of a PCF as a function of the number of layers of holes. In this simulation, the hole dimension and the hole spacing were d=35 µm and L=40 µm (accordingly, it had a core diameter of approx. 45 µm). The N=2 data point corresponds to the cladding region configuration discussed above with reference to FIG 4B and the N=1 data point belongs to the cladding region with the second layer removed.

It can be seen that the leakage strongly depends on the number of layers. As expected, decreasing N (N=2 to N=1) results in an increase in power dissipation for both modes. For the fundamental mode, the reduction in N (2 to 1) causes a leakage increase of more than three orders of magnitude. For the first order mode, the same reduction in N causes about a two orders of magnitude increase in leakage.

Because of such a strong dependence on N, one way of designing a PCF for a particular application is to first roughly tune the leakage by choosing a number of layers or rows N . Once the leakage is in the desired general or broad range, fine tuning of the leakage can be made by varying the hole size. As expected and as related to above 5A-C described, smaller holes generally result in greater power loss.

In general, for higher order modes, the increases in power loss are much larger at different values of N as well as for smaller holes. To design a PCF that only propagates a single optical mode, choosing the value of N and the hole size to essentially eliminate the first order mode over a given length ensures that higher modes are also essentially eliminated.

In various applications of PCFs such as fiber lasers, amplifiers and delivery fibers, the fiber length is on the order of a few meters or less. In such applications, N ≤ 3 can achieve single-mode operation for d/Λ ≥ 0.6. Preferably d/Λ ≥ 0.75.

13 FIG. 5 shows a loss profile 520 as a function of bending of an exemplary PCF having two layers of air-filled holes, with d=35 μm and Λ=40 μm (thus having a core diameter of approximately 45 μm). The fundamental mode is largely confined by the air/hole boundary. This property leads to a useful low bend loss, which follows from the fact that the mode in the bend region is significantly reduced as a result of the strong confinement of the air/hole boundary. At very small bend diameters, for example about 25 mm (the inverse bend radius is about 0.04 mm ^-1 ), the fundamental mode loss at a bend can be large, while the first-order mode loss at the bend due to the mode reduction effect , which is described under, can be significantly reduced.

Bend loss is commonly referred to as propagation loss, which pertains to propagation through bends. To the in 13 To obtain the results shown, the mode transition loss from straight fiber modes to bent fiber modes was not considered in the model. Such a transient loss depends on the nature of the transient and may be negligible for a slow adiabatic transient.

It can be seen that the amount of flexing of a PCF can be used to determine how much power in the fundamental and first order modes is lost due to flexing. To help design the bend-induced loss, some of the properties of bend-loss in PCFs described below may be useful.

The bending sensitivity of these types of fibers (e.g. PCFs with large holes) is very low. This result is due to the fact that the modal area is progressively reduced as the diameter of the bends is reduced, which follows from the fact that the modal field is restricted for penetration into the large nearby air holes. This characteristic is significantly different from traditional fibers and conventional PCFs, in which the modal field is off-center, with a much lower level of field perturbation upon bending. As a result, the modal field moves beyond the core-cladding boundary and produces large bending loss in traditional non-hollow fibers and conventional PCFs. This effect is particularly important for large core designs in traditional fibers and conventional PCFs.

14 shows how the effective modal indices of the fundamental and first-order modes look like for the exemplary large-core PCF 13 (with two layers of holes, N=2, d=35 µm and Λ=40 µm, core diameter of 45 µm) varies as a function of inverse bend radius. The difference between the fundamental mode curve and the first-order mode curve corresponds to the modal spacing between the fundamental mode and the first-order mode. It can be seen that in the measured range of bends the modal spacing is greater than about 0.005. Such a modal spacing (0.005) is much larger than that of conventional PCFs and traditional fibers. The large modal spacing is indicative of very low coupling between the modes and is also an indicator of robust single-mode operation. As a result of the high modal spacing (about 0.005), a sub-millimeter period perturbation is needed for sufficient phase matching to allow substantial coupling between the modes, for example in the exemplary N=2 -PCF.

High modal spacing extends even over larger core diameters. For fibers whose core diameter is on the order of about 100 µm (such as the N=1 large-core PCF from 4A ), the modal distance between fundamental and next higher mode is significantly larger than that of traditional non-hollow fibers and conventional PCFs. Again, this feature produces relatively little coupling between the modes.

In general, as the core size is further increased, the modal spacing decreases. This trend, along with increasing bend loss, can eventually place an upper limit on the maximum core diameter. Nevertheless, an order of magnitude increase in core diameter over the largest conventional PCF supporting single-mode propagation (see, for example, reference [4] in Table 1 with a core diameter of 28 µm) may be possible with the techniques and designs described here.

15-16 show a method for characterizing different single-mode propagating PCFs that also have large cores. 15 12 shows an exemplary pair of holes or cladding features 550 showing hole diameter d and hole spacing A. FIG. Such holes, arranged in a hexagonal fashion as described above, can lead to different relationships between λ/Λ and d/Λ, as shown in FIGS 16A-C shown. Other configurations are also possible.

16A FIG. 5 depicts a curve 562 in a plot of λ/Λ versus d/Δ for PCFs, substituting a single-mode (SM) propagation regime, in which only a single optical mode is propagated, from a multi-mode (MM) propagation regime which many optical modes are propagated. Conventional PCFs operating in the SM propagation regime have a relatively small core size as well as a relatively large number of hole layers (N≥4), so the overall cross-sectional profile is generally similar to that of traditional fibers. For such a fiber the value of d/A is about 0.35 in the limit when λ/Λ approaches zero. Accordingly, an area to the right of the curve causes 562 in 16A (ie increasing hole size relative to spacing, d/A), at which d/Λ ≥ 0.4, the PCF becomes an MM fiber, thereby limiting the possible core size for SM operation.

16B 12 shows a group 582 for the N=2 embodiment referred to above with reference to FIG 4B and Table 1. In that embodiment, the value of (d/Λ) ₂ is approximately 0.88 when λ/Λ is approximately equal to 0.1 (for 10 meters of fiber, while for a longer fiber this value would be larger). Thus, one can see that a useful SM propagation regime is advantageously extended from curve 562 to curve 582 as indicated by arrow 584. FIG. A larger hole size relative to pitch can be used while maintaining SM operation. By selecting a large Depending on the hole size, a corresponding hole spacing can result in a similarly large core size, as seen above with reference to Table 1.

16C FIG. 5 shows a curve 592 for the N=1 embodiment referred to above in FIG 4A and Table 1. In that embodiment, the value of (d/Λ) ₁ is about 0.94 when λ/Λ equals 0.1 (for 10 meters of fiber, while for a longer fiber this value would be larger). Thus, it can be seen that a usable SM propagation regime is advantageously increased even further from curve 562 to curve 592 as indicated by arrow 594, thereby providing a larger core size for SM operation.

17-18 show a generalized concept of augmentation of the SM propagation regime as described above in 15-16 in the context of hollow fibers. As in 17 shown and described above with reference to FIG 1A and 1B , a hollow fiber having an advantageously large core suitable for SM may have an at least partially irregular or disordered arrangement of holes. Furthermore, the holes do not necessarily have to be of uniform size or shape. Such unevenness of holes becomes in 17 shown where two exemplary holes 600 and 602 are circular and square shaped, respectively. Furthermore, the circular hole 600 is shown as having a dimension of d _i and the square hole 602 is shown as having a dimension of d _j . An example distance between the two example holes 600 and 602 is denoted by Λ _k .

For hollow fibers whose cladding region is formed by such non-uniform holes, the hole spacing parameter and the hole size parameter can be expressed as average values d _avg and Λ _avg . Using such average parameters, a similar λ/Λ versus d/A relationship (like those in 16 ) are expressed. 18A Figure 6 shows a curve 612 separating the SM propagation regime from the MM propagation regime in λ/Λ _avg - d _avg /Λ _avg space. The value (d _avg /Λ _avg ) _min represents the value of d _avg /Λ _avg in the limit when λ/Λ _avg reaches zero. Similar to the threshold of d/A, which is approximately 0.35 for conventional PCFs, (d _{avg /} Λ _avg ) _min may represent a threshold for conventional hollow fibers having a cladding region that is a relatively thick average thickness around a core which has a relatively small average dimension.

18B shows a concept for extending the SM propagation regime in a manner similar to that related to FIG 16B-C is described above. A useful SM propagation regime is advantageously extended from curve 612 to curve 622 as indicated by arrow 624, thereby providing a larger average core size for SM operation. By choosing a large average hole size, a corresponding average hole spacing can provide a similarly large average core size.

For example, the various fibers disclosed herein can be used in gain fibers, amplifiers and lasers, as well as in fiber amplifiers and fiber lasers, and in other systems as well. For example, these fibers can be applied to construct pulsed laser systems and continuous wave laser systems or high power light sources. These systems may include pumps, a length of fiber to provide a gain medium, and a pair of mirrors at either end of the fiber to form a resonator. Polarizing components can be added to achieve single polarization mode operation where only a single polarization mode is propagated. Dispersing elements can also be added for intra-cavity dispersion control. The fibers disclosed herein can also form part of the gain medium in pulsed or cw optical amplifiers, comprising one or more optical pumps, pump couplers and possibly isolators. The fibers disclosed herein can be used to form part of a fiber delivery system, in which the output of a laser is introduced into the fiber through a focusing element or splice, while the output of the fiber is fed through additional optics a goal is directed. Other configurations and uses are possible.

The fibers disclosed herein can also be used in telecommunications applications to replace MM telecom fibers for short-distance data transmission, where large core size leads to low cost and robust connections, while the single-mode nature of the fiber can improve the bandwidth of the transmission system . In such applications, a second glass can be used in place of the holes. Still other applications are possible.

Although the above-disclosed embodiments of the present invention have shown, described and clarified the fundamental novel features of the invention applied to the above-disclosed embodiments, it should be understood that various omissions, substitutions and changes in form of detail of the devices, systems and/or the methods shown can be made by those skilled in the art without departing from the scope of the present invention. Accordingly, the scope of the invention should not be limited by the foregoing description, but should be defined by the appended claims.

Claims (16)

A multimode optical fiber for propagating the fundamental mode and a plurality of non-fundamental low-order modes of the multimode optical fiber, comprising: a cladding region having cladding features arranged in a matrix, the cladding features having a substantially constant spacing Λ and an average size d, and the cladding features substantially in a plurality of rows N and in a uniform hexagonal arrangement arranged; a core region surrounded by the cladding region, the core region being formed in the matrix in place of a central cladding feature and in place of one or more inner rows of the cladding features of the hexagonal array, so that the multimode optical fiber allows the propagation of the plurality of non-fundamental low-order modes; and wherein the cladding region and the core region provide an effective index of refraction for the fundamental mode and the plurality of low-order non-fundamental modes; wherein an effective refractive index difference between the fundamental mode and the low-order non-fundamental modes is established such that the mode spacing between the fundamental mode and the first non-fundamental mode is large enough to reduce mode coupling in the multimode optical fiber, so that a single-mode input signal in the fundamental mode, after propagation along the multimode optical fiber, is output as a substantially single-mode signal in the fundamental mode. Multimode optical fiber according to claim 1 , where λ/Λ is less than or equal to 0.3. Optical multimode fiber according to any one of Claims 1 - 2 , where the cladding features comprise a gas, a mixture of gases, or a glass. Optical multimode fiber according to any one of Claims 1 - 3 wherein the core region has a refractive index that is uniform to within 0.001 as measured across a lateral cross-section perpendicular to the longitudinal direction of the fiber length. Optical multimode fiber according to any one of Claims 1 - 4 , where the V number is greater than 2.4. Optical multimode fiber according to any one of Claims 1 - 5 , further comprising dopants that provide gain. Optical multimode fiber according to any one of Claims 1 - 6 wherein the core region has a doped region, and wherein the multimode optical fiber has an outer cladding with a plurality of holes. Optical multimode fiber according to any one of Claims 1 - 7 , wherein the cladding region comprises a first and a second set of cladding features, the first set of cladding features being arranged to support the propagation of the fundamental mode, and the second set of cladding features being arranged to it supports the propagation of multiple modes for propagating pump light. Optical multimode fiber according to any one of Claims 1 - 8th , where the arrangement of the cladding features is asymmetrical. Optical multimode fiber according to any one of Claims 1 - 9 , where the arrangement of the cladding features has a two-fold symmetry to preserve the polarization of light propagating through the core region. Optical multimode fiber according to any one of Claims 1 - 10 wherein the matrix comprises a first material having a first index of refraction, and wherein at least a portion of the cladding features comprises a second material having a second index of refraction less than the first index of refraction. Optical multimode fiber according to any one of Claims 1 - 11 , wherein the cladding features comprise a first plurality of larger sized cladding features spaced apart a distance s1 and a second plurality of smaller sized cladding features spaced apart a distance s2. Multimode optical fiber according to claim 12 , where s1 and s2 are essentially identical. Optical multimode fiber according to any one of Claims 1 - 13 , wherein the effective core size is at least 30 microns and less than 150 microns. Optical multimode fiber according to any one of Claims 1 - 14 , where d/Λ is greater than 0.35. A cladding-pumped fiber amplifier or fiber laser, comprising: the multimode optical fiber according to any one of Claims 1 - 15 ; and a pump source optically connected to the optical fiber.

Images