WO2025019512A1

WO2025019512A1 - Reduction of thermal sensitivity in active optical fibers

Info

Publication number: WO2025019512A1
Application number: PCT/US2024/038239
Authority: WO
Inventors: Rasmus V.S. JENSEN; Poul Kristensen; Jeffrey W. Nicholson; Simona OVTAR; Miranda MITROVICH
Original assignee: Ofs Fitel, Llc
Priority date: 2023-07-20
Filing date: 2024-07-16
Publication date: 2025-01-23

Abstract

A rare earth-doped optical fiber (active fiber) is proposed that exhibits a reduction in the core region thermo-optic coefficient (dn / dT) with respect to the dn/dT of the surrounding cladding. The reduction of core region dn/dT has been found to reduce the effect of transverse mode instability (TMI) in the presence of high levels of pump power (or other conditions that may also increase the heat load present in the core region). Particularly, reducing core region dn/dT by as little as 10% with respect to cladding dn/dT has been found sufficient to extend the temperature range over which active fibers remain in single mode operation. The reduction in dn/dT may be provided by modifying the dopants introduced into the core region where, for example, the introduction of boron into Yb-doped fiber is known to reduce the dn/dT of the core region.

Description

REDUCTION OF THERMAL SENSITIVITY IN ACTIVE OPTICAL FIBERS

Cross-Reference to Related Applications

This application claims the benefit of U.S. Provisional Application No. 63/527,976, filed July 20, 2023 and herein incorporated by reference.

Technical Field

This disclosure relates to designs of rare-earth doped fibers (referred to at times as “active fibers”) used in high-power applications and, more particularly, to the control of the thermo-optic coefficient of the fiber’s core with respect to the cladding to reduce the possibility of triggering of transverse mode instability (TMI) at elevated operating temperatures (e.g., in the presence of high levels of pump power) .

Background of the Invention

During the operation of fiber-based lasers and amplifiers, the temperature of the core composition glass (which includes a rare-earth dopant) will increase relative to the temperature of the cladding glass, where this in turn modifies the refractive index of each segment of the fiber, and for conventional laser fibers ultimately results in the fiber becoming multi-moded and decreasing its ability to suppress higher-order modes (HOMs). The loss of single mode operation eventually leads to transverse mode instability (TMI), which is considered as one of the dominating mechanisms limiting the available output power from fiber lasers and amplifiers.

In particular, TMI results when a thermally-induced refractive index grating within an active fiber reaches a level where the propagating fundamental mode (e.g., LP01) couples to HOMs. Generally, the LP11 mode is the dominant HOM of concern. As the modes couple together, the output from the active fiber randomly fluctuates with kHz frequency between the desired fundamental mode (e.g., LP01) and the unwanted HOM (typically, the LP11 mode). The random coupling between the modes causes significant noise in the generated output and a reduction in optical efficiency.

One mitigation strategy for reducing TMI is to modify the structure of the active fiber to include enhanced waveguides with even greater HOM suppression. While useful, these designs may degrade in the presence of increasing pump powers, as the refractive index profile is perturbed by thermal effects.

Another mitigation strategy is to reduce the thermo-optic coefficient dn/dT of the fiber core to essentially zero, since this is theoretically predicted to eliminate TMI. However, such a strategy is challenging from a materials perspective, particularly when working with the silica-based materials used in optical fiber composition. Moreover, as fiber laser power is scaled, the need for larger effective area fibers to suppress nonlinear effects exacerbates this issue, as such fibers are inherently more sensitive to small index perturbations induced by thermal effects.

Summary of the Invention

The needs remaining in the art are addressed by the present invention, which relates to a rare-earth doped (active) fiber that is configured to reduce its sensitivity to thermal fluctuations by introducing a limited reduction in the thermo-optic coefficient (dn / dT) of the core region with respect to the dn / dT of the surrounding cladding.

In particular, it has been discovered that reducing the core region dn/dT by about only about 10% (with respect to the cladding dn/dT) has been found sufficient to provide predictable operating characteristics across a wide range of operating conditions. Preferably, a reduction in the range of about 10-25% is proposed to extend the temperature range over which active fibers remain in single mode operation, allowing for the use of kW level of pump beams to generate high power output signals. Said another way, it has been found that modifying the composition of the core region of an active fiber in a manner that reduces the core region thermo-optic coefficient by a minimum of about 10% with respect to the surrounding cladding layer allows the operating range of high power lasers and amplifiers to be extended over that of the prior art. In one example embodiment when using a Ytterbium-doped (Yb-doped) active fiber, it is proposed to use boron instead of fluorine as an index-lowering dopant, where the boron has been found to also reduce the dn / dT of the core with respect to the dn/dT of the cladding. Other materials (and combinations of the materials) may be preferred for use with different rare-earth dopants. More particularly, it is to be understood that the principles of the present invention extend to other known types of rare-earth doped fibers used in lasers and amplifiers (such as erbium (Er)-doped, Er-Yb doped, thulium (Tm)-doped, and the like).

For some particular types of fiber (particularly, large mode area fiber) and/or applications requiring the use of extremely high power pumps, it is contemplated that core region thermo-optic coefficient reductions in the range of 25-60% may be required. As the percentage increases (that is, as the dn/dT of the core region decreases), additional fiber modifications to maintain single mode operation may be required, such as the inclusion of rings or trenches in the cladding, or using a pedestal-type core configuration. In some embodiments, the dn/dT of these additional features may be adjusted as well to further improve the single mode operation of the device in the presence of temperature fluctuations (in particular, in the presence of very high core temperatures).

The benefits of using a reduced dn/dT for the core region is equally applicable to step-index fiber designs and graded-index fiber designs.

In one case, a fiber may take the form of an active fiber configured to generate amplification in the presence of a pump beam, where the active fiber includes a core region including a rare-earth dopant, a cladding layer surrounding the core region and a coating material surrounding the cladding layer. In accordance with the principles of the present invention, the core region includes at least one dopant selected to reduce a core region thermooptic coefficient (dn/dT) with respect to the dn/dT of the cladding layer.

Other and further aspects and principles of the present invention will become apparent during the course of the following discussion and by reference to the accompanying drawings.

Brief Description of the Drawings

Referring now to the drawings, where like numerals represent like elements in several views:

FIG. 1 contains a simplified cut-away view of an active fiber, as well as a plot of the temperature gradient along the radial direction of the fiber;

FIG. 2 illustrates this change in refractive index of the core when the active fiber is being used (i.e., when a pump beam is injected into the core of the fiber);

FIG. 3 contains data depicting the dependence of the transverse mode instability (TMI) threshold on both HOM bend loss and cladding absorption, where plot A shows the change in TMI threshold (measured in kW) associated with a conventional fiber and data points I and II are associated with two fiber designs formed in accordance with the principles of the present invention;

FIG 4 illustrates the simulated HOM bend loss for a set of four different fibers formed in accordance with the principles of the present invention, as well as the HOM bend loss for a conventional prior art fiber design (both room temperature and heated), for the sake of comparison;

FIG. 5 contains a plot of simulated bend loss for the fundamental LPO 1 mode for the set of fibers associated with the data plotted in FIG. 4;

FIG. 6 illustrates an exemplary ring configuration of a large area fiber as well as its associated refractive index profile;

FIG. 7 is a graph depicting the LP11 loss (at room temperature) for the ring-assisted fiber configuration of FIG. 6; and FIG. 8 contains a set of three graphs associated LP11 for the ring- assisted structure of FIG. 6, where FIG. 8(a) is associated with the high temperature operation of a conventional fiber, FIG. 8(b) shows the improvement in LP11 loss for a ring-assisted fiber formed in accordance with the present invention (with the dn/dT of the also adjusted), and FIG. 8(c) is associated with an alternative configuration of the ring-assisted fiber of the present invention, again having a reduction in the dn/dT of the core region, and the dn/dT adjusted to be slightly less than the value associated with the plot of FIG. 8(b).

Detailed Description

Transverse mode instability (TMI) is one of the dominating mechanisms limiting the level of output power that may be achieved with high-power fiber lasers and amplifiers. TMI occurs when the fiber is no longer able to maintain single mode operation, rendering higher-order mode (HOM) suppression ineffective. As mentioned above, TMI results when a thermally-induced refractive index grating within an active fiber reaches a level where the propagating fundamental mode (e.g., LP01) couples to HOMs. As the modes couple together, the output from the active fiber randomly fluctuates with kHz frequency between the desired fundamental mode (e.g., LP01) and the unwanted HOM (typically, the LP11 mode). The random coupling between the modes causes significant noise in the generated output and a reduction in optical efficiency.

Reduction of the thermo-optic coefficient dn/dT in the core material of active optical fibers, relative to the cladding material, is proposed here to minimize the effects of high core temperature on the waveguiding properties of the fiber in a manner that maintains single-mode operation as the heat load in the core region increases. As will be discussed below, the increased temperature may be attributed in large part to the substantial increases in pump power (e.g., kW level pump beam) required to provide the desired high- power amplified output signal. Reducing the dn/dT of the core may permit an active fiber to exhibit improved HOM suppression with respect to conventional fiber compositions, enabling further power scaling than possible with the current state of the art.

The heat load present in an optical fiber as used in a fiber-based optical amplifier (or laser) will typically be largest in the core region, since the heat is generated as a result of both quantum defect and absorption losses. FIG. 1 contains a simplified cut-away view of an active fiber 10, as well as a plot of the temperature gradient along the radial direction of fiber 10. Here, fiber 10 comprises a relatively small core region 12, surrounded by a cladding layer 14, which is itself surrounded by a coating (jacket) material 16.

The temperature gradient is shown as a function of fiber radius, with the center of core region 12 being aligned with the midpoint of the gradient. Indeed, as mentioned above, the heat load in active fibers used for amplification will be the greatest in core region 12, since the majority of the pump power resides in this region. In contrast, the fiber’s temperature drops off significantly in the outer areas of coating 16. The temperature gradient as depicted in FIG. 1 in turn modifies the refractive index of the fiber in a manner that is proportional to both the temperature and the thermo-optic coefficient (dn/dT) of the material, as characterized by the following relation:

where TH is the refractive index of the i^th region of the fiber, no, as the refractive index at a reference temperature To and T is the actual temperature in the region, and dni/dT is the thermo-optic coefficient of the i^th region of the fiber. The third term in this defined relation is used to take into account the stress (e) induced by the expansion or contraction of the region due to temperature change. In most cases, this third term is considered as a small perturbation as compared to the second term.

For the core compositions used in active fibers, the refractive index of the core will increase relative to the cladding during the operation of the device including the active fiber (i.e., either a fiber laser or a fiber amplifier). FIG. 2 illustrates this change in refractive index of the core. As mentioned above, this change in refractive index results in suppressing the HOM loss of the fiber, which will eventually lead to TMI as the heat load increases.

In accordance with the principles of the present invention, it is proposed to reduce the core region thermo-optic coefficient (dn/dT) by at least 10% with respect to the dn / dT of the surrounding cladding layer by modifying the composition of the core region to incorporate a dopant known to reduce dn/dT, where the concentration of the incorporated dopant is selected such that at least a 10% reduction with respect to the cladding layer dn/dT is achieved. This level of reduction is in contrast to the conventional teaching that dn / dT needed to be essentially eliminated (i.e., no change in refractive index in the presence of temperature fluctuations) in order to provide a suitable TMI threshold for high power /temperature applications.

It has been found that utilizing a core composition that provides a reduction of at least 10% in the core region dn/dT )with respect to the cladding dn/dT), and preferably, in the range of 10-25%, results in a standard active fiber remaining single-moded over a much wider power range than comparable fibers with traditional core compositions. Moreover, an active fiber formed to exhibit this level of reduction in core region dn/dT and associated suppression of HOMs has been found to allow for stable operation without phase-related effects from any residual HOM content in the output.

In an exemplary embodiment of the present invention, the composition of a Yb-doped optical fiber is modified to utilize boron instead of fluorine as the index-lowering dopant in the core region, since boron will also reduce the dn / dT of the core relative to the cladding. In particular, the percentage reduction of dn/dT will be a function of the amount of boron that is introduced. In one particular embodiment, both GeCh and B2O3 may be introduced into the core, where GeC>2 is used to control the actual refractive index of the core material itself. The resultant core composition may be found to have comparable performance to conventional fibers (in terms of lasing efficiency and absence of photodarkening effects, for example), while allowing for higher power performance, while still maintaining single mode operation. For exemplary applications of Yb-doped active fibers, it has been proposed to find designs having a dn = lx IO ³ relative to silica, assuming an index contribution of dn- lxlO ³ from Yb. Typical preform recipes aimed to achieve low thermo-optic coefficients that are manufacturable by MCVD utilize a combination of B2O3, P2O5, AI2O3, GeCb and fluorine. It has been found that by replacing the fluorine with additional B O3 (and also perhaps GeO₂), the value of the refractive index may be preserved, while reducing dn/dT by at least 10% and more typically in the range of about 10% - 25%. The dn/dT may be tailored, within a relevant range, by the addition of both B O3 and GeO2 in a manner that is readily manufacturable.

Compositions suitable for higher Yb concentrations may be realized by increasing the AIPO4 and either AI2O3 or P2O5 concentrations, which are considered to have negligible effect on the overall dn/dT (in the case of AIPO4) or lead to only a relatively small reduction in dn/dT (for P2O5). The value of dn/dT for different glass regions of the active fiber (such as core, cladding, trench, ring, and the like) may be calculated based on the concentrations of individual dopants and their values of dn/dT. One reference is found to offer the following values:

The dn/dT may also be determined by measuring optical properties, such as bend loss of different optical modes, as a function of temperature, and inferring the change in index profile needed to produce the observed changes.

FIG. 3 contains data depicting the dependence of the TMI threshold on both HOM bend loss and cladding absorption. Plot A shows the change in TMI threshold (measured in kW) as the HOM bend loss and cladding absorption of a conventional fiber increase. It is clear that a TMI threshold of greater than 4kW is unlikely to be obtained with a conventional fiber. Also shown in FIG. 3 are data points I, II, associated with active fibers designed in accordance with the principles of the present invention to exhibit a reduction of core dn / dT with respect to the cladding dn/dT. The core region of the prior art plot A may exhibit a dn/dT on the order of about 9.3e-6, where the estimated values for the inventive fiber have been determined to fall within the range of 8.5 - 8.3e-6 (in both cases, a conventional silica cladding with a dn/dT of 10e-6 is used).

The increase in TMI threshold obtained by a fiber having the characteristics shown in data points I, II is evident; indeed, an increase in TMI threshold of about 50% is observed with respect to the conventional values of plot A. In accordance with the principles of the present invention, the improvement in TMI threshold is associated with the increased ability of the inventive fiber to preserve strong HOM suppression, even as the pump power is increased.

This conclusion regarding the performance improvements of a fiber formed in accordance with the present invention to exhibit a reduction in core region dn/dT with respect to the cladding dn/dT may be further supported by simulations that have been developed, taking into account the guiding properties of the power under high power operation. FIG. 4 is a plot of simulated HOM bend losses associated with a high-power pump operating at 4.2 kW. For the sake of comparison, the HOM bend loss for a conventional fiber (having a core dn/dT of 9.3e-6) is shown, for both room temperature (RT) operation and elevated temperature operation in the presence of a 4.2kW pump power. As expected, the HOM suppression is significantly degraded for a conventional fiber operating in the presence of a high power pump (compare curve 1 to curve 2).

The simulated HOM bend loss for a set of four different fibers formed in accordance with the principles of the present invention is shown as well in FIG. 4, ranging from a dn/dT of 8.5e-6 (associated with curve 3) to 7.5e-6 (associated with curve 6). As expected, all four curves confirm a significant improvement in HOM suppression when compared to a prior art fiber operating at the same temperature (i.e., as shown in curve 2). This simulation is considered to affirm the presumption that the reduction of core dn/dT in association with the principles of the present invention sufficiently increases the threshold at which TMI is triggered that single mode operation may be maintained. It is to be understood that the range of dn/dT analyzed in this simulation is exemplary only, and that values outside of this range may also be acceptable for use in improving HOM suppression.

A related consideration is associated with an inventive fiber with reduced dn/dT will be able to adequately guide the fundamental mode (LP01) under a heat load. Indeed, as the core dn/dT would approach zero, it is known that the fundamental mode becomes extremely lossy under high power operation, eventually leading to device failure. It has been found, however that as long as the reduction in dn/dT is maintained within a reasonable percentage, the core region remains able to support the propagation of the fundamental mode. FIG. 5 contains simulation data supporting this conclusion, which shows that in most situations, the loss of the fundamental mode is minimal. Again, the signal loss associated with a conventional prior art fiber is shown for both an RT situation (curve 1A) and an elevated temperature situation (curve 2A). In many applications, the loss of the fundamental mode approaching 0.40 associated with a core dn/dT of 7.5e-6 (curve 6A) remains a concern and may unacceptably degrade device performance. In such situations, the fiber design may include other features, as discussed below, to maintain confinement of the fundamental mode within the core region.

As the core size and required mode field diameter (MFD) of a doped gain fiber become larger and larger, reducing the core region dn/dT may not be sufficient to maintain single mode operation at elevated operational temperatures. Core sizes can become as large as 50 qm (or more), with an MFD greater than 35 qm. High energy pulse fiber lasers, for example, require extremely large cores for suppression of nonlinear impairments such as selfphase modulation, stimulated Raman scattering, and stimulated Brilloiun scattering. At these larger diameters, maintaining a fundamental mode output, and diffraction-limited, single-mode performance (even at room temperature or relatively low pump powers) is considered to be exceedingly difficult.

Thus it is contemplated that the reduction of core region dn/dT as described above may be insufficient to maintain single mode operation. In this situation, therefore, it is proposed to combine the teaching of reduced core dn/dT with other types of HOM suppression known in the art. For example, the inclusion of features such as rings or trenches in the cladding, or utilizing a pedestal core structure may need to be used, in combination with a core region having a reduced dn/dT, in order to maintain acceptable high-power performance with suitable HOM suppression.

Further, it is contemplated that the performance of these large area fibers may be further improved by reducing the dn / dT of the other features in the fiber structure. For example, in a so-called ring-assisted design, it may be beneficial to also modify the dn/dT of the ring and, perhaps, the cladding material as well. By decreasing the dn/dT of the core and optimizing it for the cladding and the ring, effective coupling and, as a result higher order mode suppression may be achieved under very high pump power levels. FIG. 6 illustrates an exemplary ring configuration of a large area fiber 60 as well as its associated refractive index profile. Fiber 60 is shown as comprising a core region 62, an inner cladding layer 64, a ring 66, and an outer cladding layer 68.

FIG. 7 is a plot of LP11 (HOM) loss at room temperature for an exemplary ring-assisted large area fiber configuration as shown in FIG. 6. In particular, this plot is associated with high absorption, ring-assisted Yb-doped active fiber operated in a counter-pumped configuration. It is known that counter-pumped configurations may be veiy prone to dramatically reduced HOM losses at the output end, as the heat load is relatively high at the output termination of the fiber, while the bend radius is not severe enough to provide high HOM losses. By optimizing the dn/dT of core region 62, inner cladding layer 64, and ring 66, efficient coupling between core 62 and ring 66 may be preserved, even at pump power levels of 4kW (or higher) . In comparison, FIG. 8 contains plots of LP11 loss for a heated fiber (e.g., under high pump power conditions), where the plot of FIG. 8(a) is associated with a conventional ring-assisted fiber where the dn/dT of the core is 9.3e-6 (with the dn/dT of the inner cladding and ring both being 1.0e-5. FIG. 8(b) is a plot of LP11 loss for a ring-assisted fiber formed in accordance with the principles of the present invention, where the dn/dT of the core region is reduced to 8.3e-6. In this example, the dn/dT of the ring is increased (relative to the prior art) to a value of 1.1 dn/dT. Reducing the dn/dT to a value of 1.02e-5 results in the plot of FIG. 8(c), where it is to be noted that the scaling on the y-axis of FIG. 8(c) is different than that of the others, here extending between 400 and 800 dB/m (as opposed to the 1 - 1000 dB/m scale of FIGs. 8 (a) and (b).

Designs with either step index or graded index cores may be combined with a surrounding trench, with the trench used to assist in the separation of the fundamental mode and the HOMs. By including dn/dT tailoring the inner cladding and trench, in combination reduction of the core dn/dT, the performance of the resultant fiber under variations in operation/ thermal load conditions may be optimized as well.

The use of pedestal design is somewhat different, since the pedestal relies on the index difference between the core and an extended, raised inner cladding (which may or may not have additional cladding features) . Reductions in the core region dn/dT in this case may be of particular interest to applications using any of the known rare-earth dopants, where the designs of these fibers tend to use optimal glass compositions that are inherently high index (with respect to silica) and without a reduction in the core dn/dT may be difficult to maintain in single mode operation in their large mode area form.

While the foregoing invention has been described in terms of the embodiments discussed above, numerous variations are possible. Indeed, while specific reference has been made to creating a Yb-doped active fiber, the premise of reducing the dn / dT of the active fiber core region is equally applicable to active fibers formed to include other rare-earth dopants (or combinations thereof). Accordingly, modifications and changes such as those suggested above, but not limited thereto, are considered to be within the scope of the following claims.

Claims

What is claimed is:

1. An optical fiber configured to generate amplification in the presence of a pump beam, comprising a core region including a rare-earth dopant, a cladding layer surrounding the core region; and a coating material surrounding the cladding layer, wherein the core region includes at least one dopant selected to reduce a core region thermooptic coefficient (dn/dT) with respect to the dn/dT of the cladding layer.

2. The optical fiber of claim 1 wherein the at least one dopant is selected to reduce the core region dn/dT by at least about 10% with respect to the cladding layer dn/dT.

3. The optical fiber of claim 2 wherein the core region dn/dT is reduced to a value within a range of 10% - 25% of the cladding layer dn/dT.

4. The optical fiber of claim 2 wherein the core region dn/dT is reduced by no more than 60% with respect to the cladding layer dn/dT.

5. The optical fiber of claim 1 wherein the core region includes Ytterbium (Yb) as the rare-earth dopant, with boron included as the at least one dopant for reduction of dn/dT.

6. The optical fiber of claim 4 wherein the addition of boron is controlled to provide a value of dn/dT within the range of 8.5 - 8.0e-6.

7. The optical fiber of claim 1, wherein the fiber further comprises at least one additional feature for controlling suppression of higher order modes.

8. The optical fiber of claim 7, where the dn/dT of the at least one additional feature is modified to improve separation between a propagating fundamental mode and unwanted higher-order modes.