Fiber designs for higher power lasers

Ming-Jun Li, Xin Chen, Ji Wang, Anping Liu, Stuart Gray, Donnell T. Walton, A. Boh Ruffin, Jeffrey Demeritt, Luis Zenteno

Science and Technology Division, Corning Incorporated, SP-AR-02-2, Corning, NY 14831

Tel: 607-974-3099, E-mail: lim@corning.com

ABSTRACT

This paper reviews different fiber design approaches for high power lasers. First, we discuss the conventional step index profile design and methods for achieving single mode operation in high power lasers such as bending, helical core fibers and Yb dopant profile designs. Then we present new design approaches for reducing the SBS through profile and glass composition designs. Finally, we describe fiber designs to achieve single polarization and at the same time to mitigate the SRS effect.

Keywords: Fiber laser, fiber design, stimulated Brillouin scattering, stimulated Raman scattering, large mode area fiber

1. INTRODUCTION There have been significant efforts on developing high power fiber lasers with narrow line-width and good beam quality [1]. Such high power lasers pose significant challenges on fiber designs to reduce nonlinear effects such as stimulated Brillouin scattering (SBS) and stimulated Raman scattering (SRS), and achieve simultaneously single mode operation. Neither SBS nor SRS are desirable for high power lasers because they limit the amount of output optical power.

One well known way to reduce the SBS and SRS effects is to make fibers with large mode area (LMA) [2]. However, when the core size is too large, the fiber becomes multimoded. Modal discrimination techniques, such as bending, are then required to achieve single mode operation. But for large core size, bending deforms the mode field distribution and reduces the mode area, which puts a limit on the benefit of using large core.

Nonlinear effects can also be managed through design approaches other than increasing the effective area. It has been found that SBS can be reduced by decreasing the overlap between the optical and acoustic fields [3-5]. This is achieved through fiber profiles designs or fiber material designs to confine the optical and acoustic filed in different regions in the core. SRS effect can be mitigated by using fibers that have a cutoff wavelength for the fundamental mode [6], for example single polarization fibers [7-8].

In this paper, we review design concepts for achieving single mode operation in high fiber power lasers. We also discuss different fiber design approaches for managing SBS and SRS effects in high power fiber lasers.

2. FIBER DESINGS FOR SINGLE MODE LASER OPERATION To reduce the nonlinear effects, it is desirable to design a fiber with an effective area as large as possible. In a step index profile, the mode effective area is increased by lowering the core refractive index and increasing the core diameter. This can be seen in Figure 1, which shows how the effective area changes with the core diameter for different numerical aperture (NA) values of a step index fiber. For example, by lowering the NA to 0.05, and increasing the core diameter to 30 µm, the effective area is increased by a factor of 10 compared to standard single mode fiber (SMF) with NA of 0.12 and core diameter of 8 µm. However, for a given NA value, if the core diameter is too large, the fiber becomes multimoded. This is illustrated in Figure 2 where the fiber core diameter is plotted as a function of the NA for the cutoff wavelength of 1060 nm. With the core size larger than 20 µm, the fiber becomes multimoded even with NA as low as 0.05. A modal discrimination technique is then required to strip the higher order modes.

Invited Paper

Optical Components and Materials IV, edited by Shibin Jiang, Michel J. F. Digonnet, Proc. of SPIE Vol. 6469, 64690H, (2007) · 0277-786X/07/$18 · doi: 10.1117/12.714018

Proc. of SPIE Vol. 6469 64690H-1

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 08/05/2014 Terms of Use: http://spiedl.org/terms

0

200

400

600

800

1000

1200

1400

0 10 20 30 40 50 60

Core Diameter (µm)

Effe

ctiv

e Ar

ea ( µ

m2 )

NA=0.05NA=0.08NA=0.12

Figure 1. Effective area changes with core diameter for

different NA values.

Figure 2. Core diameter as a function of NA for cutoff

wavelength of 1060 nm. One way to strip the higher order modes is to bend the fiber [9]. When a fiber is under bending conditions, the bend losses of the higher order modes increase more rapidly than the fundamental mode. By choosing a right bend radius, it is possible to have the differential bend losses between the higher order modes and the fundamental mode high enough such that the fiber behaves like a single mode fiber. However, this approach has its limitations. For a given NA, the larger the core size the smaller the bend radius is required to strip the higher order modes. This is illustrated in Figure 3, where the bending losses of LP01 and LP11 modes are plotted versus the bend radius for 20 and 30 µm core fibers. The NA value for both fibers is 0.06. To create a bend loss of several dB/m, the bend radius is about 10 cm for the 20 µm core fiber, and 5 cm for the 30 µm core fiber, respectively. For a fiber with a core larger than 30 µm, the bend radius is even smaller than 5 cm. One drawback of using a small bend radius is that the mode field is deformed and the effective area is reduced. To illustrate this phenomenon, Figure 4 shows the effective area changes as a function of bend radius for the fibers with 20 and 30 µm core diameters. For the fiber with a 20 µm core, the effective area change is negligible for the bend radius down to 5 cm. On the other hand, for the fiber with a 30 µm core, the effective area drops quickly when bend radius decreases. For the bend radius of 5 cm, the effective area is reduced by more than 25%. For a core size larger than 30 µm, the effective area is expected to drop even quicker and the benefit of larger core size disappears. Bases on the results in Figures 3 and 4, we conclude that, for LMA fibers, the largest core size that can be used with the bending method to achieve single mode operation is about 25 µm.

1.E-12

1.E-10

1.E-08

1.E-06

1.E-04

1.E-02

1.E+00

1.E+02

1.E+04

0 2.5 5 7.5 10 12.5 15Bend Radius (cm)

Ben

d Lo

ss (d

B/m

)

20 um Core LP0120 um Core LP1130 um Core LP0130 um Core LP11

Figure 3. Bending losses of LP01 and LP11 modes for the

fibers with 20 and 30 µm core diameters.

0

50

100

150

200

250

300

350

400

450

500

0 5 10 15 20Bend Radius (cm)

Effe

ctiv

e A

rea

( µm

2 )

20 um Core30 um Core

Figure 4. Effective area changes with the bend radius for

the fibers with 20 and 30 mm core. Another approach to strip the higher order modes is to make a fiber with helical a core [10-11]. A helical core fiber in a pump cladding is shown schematically in Figure 5. The structure can be described by two parameters: one is the offset from the center Doff, the other one is the spin pitch, Λ, defined as the distance along the length of the fiber that completes one full cycle of the rotation. The radius of curvature of the helix can be calculated from the following equation

off

off

DD

R22 ))2/(( πΛ+

= (1)

Single mode

Multimode

0

10

20

30

40

50

60

70

80

90

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14NA

Cor

e D

iam

eter

( µm

)

Proc. of SPIE Vol. 6469 64690H-2

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 08/05/2014 Terms of Use: http://spiedl.org/terms

The helical structure has a constant curvature so that it has a similar effect to bending a fiber. The advantage of a helical core fiber is that the bending is built into the fiber when the fiber is made and no bending is required in the fiber deployment. By changing the two parameters of the offset and the spin pitch, different bending radii can be obtained. Figure 6 shows calculated effective bending radii as a function of spin pitch for different offset values. By replacing the bend radius with the curvature radius of the helix, the bending loss of the helical fiber can be obtained from the conventional bending loss methods. Single mode operation on a 30 µm diameter core and a numerical aperture of 0.087 has been demonstrated yielding 60.4 W of output at 1043 nm in a beam with M2<1.4. However, we believe that similar to the bending approach, the helical-core approach based on a conventional LMA fiber will not be able to scale up to a larger core as the modal discrimination between the fundamental mode LP01 and first higher order mode, typically LP11 will be reduced and tighter bending is required. In addition, helical-core fiber has one further disadvantage: the axis of the core is not parallel to the fiber axis. In practice, this means that after cleaving and/or polishing in a standard way, the emitted beam will be asymmetric and distorted, thus reducing its beam quality.

Figure 5. Schematic diagram of a helical-core in a pump cladding

0

20

40

60

80

100

120

0 5 10 15 20 25 30

Pitch (mm)

Effe

ctiv

e B

endi

ng R

adiu

s (m

m)

Offset=0.1mmOffset=0.2mmOffset=0.4mmOffset=0.6mmOffset=0.8mm

Figure 6. Calculated effective bending radii for different

values of offset and pitch for helical core fibers

Figure 7. Fiber design with a reduced Yb doping region in

the core

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0 0.2 0.4 0.6 0.8 1 1.2RYb/Rcore

Pow

er fr

actio

n in

fund

amen

tal m

ode

Figure 8. Power fraction as a function of the ratio of radius

of Yb doping region to the radius of the core A different approach to modal discrimination without using bending is to design the Yb doping profile such that its overlap with the fundamental mode is optimized while its overlap with higher order modes is reduced [13-14]. In this way, the fundamental mode will experience higher gain and the gains of higher order modes are suppressed. One simple design is to have Yb doping region less than the core as shown in Figure 7. By changing the radius of the doping region, the differential gain between the fundamental mode and higher order modes is maximized. As a result, the fraction of power in the fundamental mode is increased. This technique has been successfully used in a 33 mm core fiber with 0.06 NA to produce nearly diffraction limited output. We believe that it can enable single mode operation with even larger core size. As an example, we modeled a fiber with a NA of 0.1 and a core diameter of 70 µm. Figure 8 shows the power fraction change in the fundamental mode after 6 m long fiber when the radius of the doping region changes. It can be seen that the power fraction in the fundamental mode reaches maximum in when the ratio RYb/Rcore is around 0.6. Note

RYb Rcore

Doff

∆Λ

Proc. of SPIE Vol. 6469 64690H-3

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 08/05/2014 Terms of Use: http://spiedl.org/terms

that the fiber in the modeling is not bent. By using the dopant profile design together with the bending method, it is possible to have single mode laser operation with a larger core area than conventional step index fibers.

3. FIBER DESIGNS FOR REDUCING SBS EFFECT In addition to increasing the effective area, the SBS effect can be reduced by designing a fiber with reduced overlap between the optical and acoustic fields. To understand the factors that affect the SBS, we have developed a model using the coupled mode formalism. The threshold of SBS is affected by the following factors [3]: (2) where αu is the acoustic attenuation coefficient for the acoustic mode of order u, ∆β is the phase mismatch, Aeff is the optical effective mode area, G(νmax) is the effective gain coefficient at the peak frequency, and K is the polarization factor, and ao

uI is the normalized overlap integral between the electric and acoustic fields, (3)

The above equation indicates that, in addition to the mode area, the SBS threshold can be increased by increasing the acoustic loss or by decreasing the overlap integral and the maximum gain coefficient. The overlap integral can be controlled by fiber refractive index profile design and acoustic velocity profile design. The acoustic loss can be changed by glass composition design. To reduce the overlap between the optical and acoustic fields, the simple step index profile made of GeO2 doping is not suitable because the fundamental optical and acoustic modes have very similar field distributions and the overlap integral is about 1. However, it is possible to use more sophisticated profile designs to change the optical and acoustic fields. The basic concept in designing a fiber with low SBS is to locate the optical field in a different region of the core than the acoustic field to minimize their interaction. Figure 9 shows schematics of profiles designs that can reduce the overlap. One is to confine the acoustic field in the central portion of the core as shown in Figure 9(a), and the other is to push the acoustic filed towards the edge of the core as shown in Figure 9(b).

Figure 9. Profile designs to reduce the overlap between the optical and acoustic fields

Table 1. Trend of optical and acoustic refractive index change of different dopants in silica

GeO2 P2O3 TiO2 B2O3 F2 Al2O3 Optical refractive index Acoustic Refractive index

Optical

Acoustic

(a) (b)

( )( ) θρρθ

θρ

rdrdrdrdEE

rdrdEEI uao

u

∫∫∫=

*2*00

2**00

aou

ueffth ILG

KAP

),( maxν

α∝

Proc. of SPIE Vol. 6469 64690H-4

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 08/05/2014 Terms of Use: http://spiedl.org/terms

Figure 10. Dopant profile designs for reducing the overlap between the optical and acoustic fields

The SBS can be also reduced by placing the right dopants in the core and the cladding because different dopants have different effects on optical and acoustic properties. Table 1 lists some common dopants that can be used for making silica glass based fibers and their effects on optical and acoustic refractive indices. The acoustic refractive index is defined as )()( _ rVVrn LSilicaLa = , where VL(r) is the longitudinal acoustic velocity in the core region, and VL_Silica is the longitudinal acoustic velocity of pure silica glass. Among the dopants listed in Table 1, the first three dopants increase both the optical and acoustic indices, while the last three dopants have opposite effects on the optical and acoustic indices. Two approaches can be used to manipulate dopants for reducing the SBS. One is to design a fiber structure using different dopants that guides the optical wave but anti-guides the acoustic wave. This is accomplished, for example, by choosing a dopant in the core such as Al2O3 to increase the optical index but decrease the acoustic index or by choosing a dopant in the cladding such as F to decrease the optical index but increase the acoustic index. The resultant optical and acoustic refractive index profiles are shown schematically in Figure 10(a). Because the acoustic wave is not guided in the core region, the interaction between the optical and acoustic waves is reduced. SBS improvement using this approach has been demonstrated [14]. The second approach is to use different dopants in the core to change the optical and acoustic field distributions to reduce their overlap. A design example of two doping regions in the core is shown in Figure 10(b). In this example the first core region is doped with GeO2 and the second core region is doped with Al2O3. The optical refractive index profile is selected to be a step, but the acoustic index profile becomes a W-shape. The optical field is confined in the composite core region whereas the acoustic field is confined in the first core region only. As a result, the overlap integral is reduced significantly. Numerical modeling indicates that SBS threshold increase from 3-7 dB is possible by the two design approaches.

Figure 11. Fiber design schematic of a SBS improved

fiber.

0

5

10

15

20

25

30

35

0

5

10

15

20

25

30

35

0.0 50 100 150 200 250 3000.0 50 100 150 200 250 300Output power (W)

Bac

kwar

d po

wer

(mW

)

Step Index

Low SBS

SBS threshold is improved by 6 dB

Dual pump configuration

Figure 12. Measured back-scattered power for a step index

fiber and a SBS improved fiber

Al

Ge Ge

B/F

Yb

Optical Index

Acoustic Index

Optical

Acoustic

Al2O3 doped

(a)

Al2O3 doped

(b)

GeO2 doped

Proc. of SPIE Vol. 6469 64690H-5

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 08/05/2014 Terms of Use: http://spiedl.org/terms

To demonstrate the design concept by reducing the overlap using different dopants, a fiber was made using Ge and Al co-doping. Figure 11 shows fiber optical and acoustic refractive index profiles and Ge and Al dopant profiles used in the core. In this design, the Ge and Al have counter-graded profiles rather than step changes. This Al/Ge-counter-graded composition profile design eliminates the physical interface within the core, which reduces scattering loss due to interface imperfections. In fiber fabrication, the flat optical-index profile can be conveniently achieved by up-doping Germania-content and at the same time down-grading Alumina-concentration in a manner that they are compensated one another in optical refractive-index contribution. On the other hand, linearly down-grading Alumina and up-grading Germania from the center of the core form a V-shaped acoustic profile, causing the acoustic wave to be primarily guided or distributed towards the edge of the core. As a result, the overlap between the optical and the acoustic field is reduced. Modeling efforts have indicated an increased SBS threshold of more than 6 dB. Figure 12 shows the measured SBS power of the fiber in comparison of a step index fiber made with GeO2 only. The SBS threshold is increased by about 6 dB, which agrees well with the theoretical prediction.

4. SINGLE POLARIZATION FIBERS Linearly polarized laser output is required for many applications, for example phase-matched nonlinear conversion configurations to achieve a wider range of operating wavelengths, and coherent beam combination for scaling laser output to even higher powers. Single polarization (SP) optical fibers are attractive for making fiber lasers with polarized output. Unlike a polarization maintaining (PM) fiber that supports two orthogonal polarization modes, a single polarization fiber guides one polarization mode only while the other polarization mode is eliminated. There are several advantages of using SP fibers compared to polarization maintaining fibers including high extinction ratio, no polarization mode dispersion and polarization dependent loss issues, and low sensitivity to polarization alignment. Another advantage of SP fiber for high power lasers is that it can suppress SRS effect. An important property of single polarization fiber is the cutoff wavelength for the fundamental mode. Through proper design this cutoff wavelength for each polarization mode can be separated, thus providing the means to achieve single-polarization propagation. The spectral range between the cutoff wavelengths of each of the two orthogonal polarization modes defines the single-polarization bandwidth. We have designed and fabricated dual-hole-assisted single polarization fibers [7-8]. Figure 13 shows a dual-hole-assisted single polarization fiber design. The fiber has dual air holes next to a central core that can be circular or elliptical. In this design, the two air holes are used to produce large birefringence and to create differential cutoff wavelengths. Fibers with over 50 nm single polarization bandwidth have been demonstrated in the 1310-1550 nm. Single polarization fibers operating at 1060 nm have been also demonstrated. Figure 14 shows a measured transmission spectrum of a fiber at the 1060 nm window. The single-polarization bandwidth is about 30 nm.

Figure 13. Dual-hole single polarization fiber schematic.

Wavelength (nm)1020 1040 1060 1080 1100

Spe

ctra

l Sig

nal (

dB)

-85

-80

-75

-70

-65

-60

1060 window

Figure 14. Measured transmission spectrum of a single

polarization fiber designed for 1060 nm window.

Core Air hole

Cladding

Proc. of SPIE Vol. 6469 64690H-6

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 08/05/2014 Terms of Use: http://spiedl.org/terms

Air holeYb-doped core

cladding

120

100

80

o 600-

40

o 20

0

0 50 100 150

Coupled Pump PowerlW)

y 0.5242x - 6.2754

200 250

Linear single polarization fiber lasers have been demonstrated using dual-hole single polarization fibers [15-16]. In the experiment reported in reference [16], an ytterbium-doped, single polarization double clad fiber was used. The double clad fiber was made using the dual-hole elliptical core single polarization fiber design doped with Yb. Figure 15 is schematic of the single polarization double clad fiber. The inner cladding was machined to a D shape in order to enhance the pump field overlap with the core. A low index polymer coating served as the outer cladding. In the laser experiment, each end of the fiber was pumped with a fiber coupled diode system. One end was pumped with a 70-W diode at 976 nm (LIMO). The laser performance is shown in Figure 16. A maximum output power of 104 W was measured for 210 W of coupled pump power. The polarization extinction ratio was 17 dB. The slope efficiency and laser threshold were 52% and 13 W, respectively. The 52% slope efficiency is less than the 77% that we have typically measured in our standard ytterbium-doped double clad fibers. We calculated that this decrease in slope efficiency corresponded to an excess 5dB of background loss in our gain fiber. We expect lower background loss and improved performance as our fiber processing is optimized.

Figure 15. Schematic of ytterbium-doped, double clad,

single polarization fiber cross section.

Figure 16. Output power vs. coupled pump power.

0

20

40

60

80

100

120

140

160

0 1 2 3

Core ∆ (%)

Band

wid

th (n

m)

Figure 17. Single polarization bandwidth as a function of

core delta for dual-hole single polarization fibers

Figure 18. Single polarization fiber design with two air

holes and two stress rods.

To design a fiber with a large bandwidth for hole assisted single polarization fiber, the relative refractive index (delta) in the core must be high. Figure 17 plot the single polarization bandwidth versus the core delta for dual-hole single polarization fibers. It is clear that the bandwidth increases linearly with the core delta. For bandwidth larger than 30 nm, the core delta has to be greater than 0.5%. One draw back of using high delta is that the effective area becomes small.

Core Air hole

CladdingStress rod

Proc. of SPIE Vol. 6469 64690H-7

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 08/05/2014 Terms of Use: http://spiedl.org/terms

For high power lasers, large effective area is desirable. To increase the bandwidth without using high delta core, we propose to add stress rods to the dual-hole design as shown in Figure 18. In this design, the stress rods create a large birefringence, which does not depends on the core delta. The two air holes are used to create cutoffs for the two fundamental polarizations modes. As a result, large mode area single polarization fiber can be realized with high bandwidth. With this design, single polarization fibers with a core of more than 20 µm and bandwidth of more than 100 nm are possible.

In addition to the advantage of linear single polarization output, a single polarization fiber has the benefit of reducing the SRS in high power fiber lasers. Light propagating at wavelengths longer than the fundamental cutoff wavelength in hole-assisted fiber is strongly coupled to cladding-modes, thus experiences a high distributed loss and reduced overlap with the fundamental. This property can be used to make single-transverse-mode fiber devices with virtual suppression of Raman scattering.

Wavelength (nm)1420 1440 1460 1480 1500 1520

OSA

Sig

nals

(dB

)

-90

-85

-80

-75

-70

-65ID: R04-309-3E

Wavelength (nm)1420 1440 1460 1480 1500 1520

OSA

Sig

nals

(dB

)

-90

-85

-80

-75

-70

-65ID: R04-309-3E

Figure 19. Transmission spectrum of a single polarization

fiber used in the CP-RASE experiment.

-95.00

-90.00

-85.00

-80.00

-75.00

-70.00

-65.00

-60.00

-55.00

-50.00

1480 1500 1520 1540 1560 1580 1600 1620

Wavelength (nm)

Pow

er (d

Bm

)

SSS

-95.00

-90.00

-85.00

-80.00

-75.00

-70.00

-65.00

-60.00

-55.00

-50.00

1480 1500 1520 1540 1560 1580 1600 1620

Wavelength (nm)

Pow

er (d

Bm

)

SSS

Figure 20. Measured CP-RASE power for 300 m single

polarization fiber (bottom), 50 m SMF (middle) and 500 m SMF (top).

Reduction of Raman gain by several orders of magnitude has been demonstrated compared to single-mode fiber SMF-28 in measurements of counter-propagating Raman amplified spontaneous emission (CP-RASE) in a single-mode dual-hole-assisted fiber [6]. The measured cutoff wavelengths near the end of this fiber are shown in Figure 19. The fiber has a single-polarization bandwidth of about 40 nm between 1465 nm and 1505 nm. Loss at the Stokes wavelength λs=1592 nm is more than 25 dB. The hole-assisted single-polarization fiber with length L=300m was excited with about 1W of power at the pump wavelength λp=1486 nm. The bottom trace in Figure 20 shows the measured CP-RASE power for this fiber. For comparison the top two traces in the same figure show the measured CP-RASE power from 50 and 500 m of SMF-28. Clearly, the 300 m of hole-assisted fiber has a power level at least 23 dB below that of 500 m of SMF, which demonstrates the effectiveness of using single polarization fiber to suppress Raman scattering.

5. CONCLUSIONS There has been significant progress in fiber designs for single mode fiber laser applications. For step index fiber designs, single mode fibers with large mode area require low NA in the core. For multimode large mode area fibers, mode discrimination techniques are needed to strip higher order modes to produce effective single mode operation. For NA of 0.06, the largest core diameter is limited to about 25 µm if bending method is used to strip the higher order modes due to the mode area reduction with the bending. Yb doping profile designs can be used in conjunction with bending to further improve the effective area. In addition to effective area, the SBS effect can be reduced by designing fibers with reduced overlap between the optical and acoustic fields. Single polarization fibers are useful for making linear polarized fiber lasers, and are effective in suppressing the SRS effect. It is expected that, with the advanced fiber designs, linearly

Proc. of SPIE Vol. 6469 64690H-8

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 08/05/2014 Terms of Use: http://spiedl.org/terms

polarized single mode fiber lasers with power more than 2 kW and line-width of a few nanometers will become reality in the near future.

REFERENCES

1. Stuart Gray, Donnell Walton, Ji Wang, Ming-Jun Li, Xin Chen, Boh Ruffin, Jeffrey Demeritt, Luis Zenteno, High power, narrow linewidth fiber amplifiers, Optical Amplifiers and Applications 2006, Paper OSu B1, Whistler, BC, June 25-28, 2006.

2. N. G. R. Broderick, H. L. Offerhaus, D. J. Richardson, R. A. Sammut, J. Caplen, and L. Dong, Large Mode Area Fibers for High Power Applications, Optical Fiber Technology 5(2), pp.185-196, 1999. A. Eisenberg, Guide to Technical Editing, Oxford University, New York, 1992.

3. A. Boh Ruffin, M.J. Li, X. Chen, A. Kobyakov, and F. Annunziata, Brillouin Gain Analysis for Fibers with Different Refractive Indices, Opt. Lett. 30 (23), pp. 3123-3125, 2005.

4. A. Kobyakov, S. Kumar, D. Q. Chowdhury, A. B. Ruffin, M. Sauer, S. R. Bickham, and R. Mishra, Design concept for optical fibers with enhanced SBS threshold, Opt. Exp. 13 (14), pp. 5338-5346, 2005.

5. Y. Koyamada, S. Sato, S. Nakamura, H. Sotobayashi, and W. Chujo, Simulating and designing Brillouin gain spectrum in single mode fibers, J. Lightwave Technol. 22(2), pp. 631-639, 2004.

6. L. Zenteno, J. Wang, D. Walton, B. Ruffin, M. Li, S. Gray, A. Crowley, and X. Chen, "Suppression of Raman gain in single-transverse-mode dual-hole-assisted fiber," Opt. Express 13, 8921-8926 (2005).

7. Daniel A. Nolan, George E. Berkey, Ming-Jun Li, Xin Chen, William A. Wood, Luis A. Zenteno, “Single-polarization fiber with a high extinction ratio”, Optics Letters, Vol.29, pp. 1855-1857, 2004.

8. Ming-Jun Li, D.A. Nolan, G.E. Berkey, Xin Chen, J. Koh, D.T. Walton, J. Wang W.A. Wood and L.A. Zenteno, “High performance single-polarization optical fibers”, invited talk presented at APOC 2004, Proceedings of SPIE vol. 5623, pp. 612-621, 2005.

9. J.P. Koplow, D.A.V. Kilner, and L. Goldberg, “Single-mode operation of a coiled multimode fiber amplifier”, Opt. Lett. 25, 442-444, 2000.

10. P. Wang, L. J. Cooper, R. B. Williams, J. K. Sahu and W. A. Clarkson, “Helical-dore ytterbiuj-doped fibre laser”, Electronics Letters 40, (21) 1325-1326, (2004)

11. P. Wang, L. J. Cooper, J. K. Sahu, and W. A. Clarkson, “Efficient single-mode operation of a cladding-pumped ytterbium-doped helical-core fiber laser”, Optics Letters 31 (2) 226-228 (2006)

12. J.M. Sousa and O.G. Okhotnikov, “Multimode Er-doped fiber for single-transverse-mode amplification,” Appl. Phys. Lett. 74(11), pp. 1528-1530, 1999.

13. Pierre Laperle, Claude Paré, Huimin Zheng, André Croteau, and Yves Taillon, “Yb-Doped LMA Triple-Clad Fiber Laser”, Photonics North 2006, edited by Pierre Mathieu, Proc. of SPIE Vol. 6343, 63430X, (2006)

14. P.D. Dragic, C. H. Liu, G. C. Papen , and A. Galvanauskas, Optical Fiber With an Acoustic Guiding Layer for Stimulated Brillouin Scattering Suppression, CLEO’2005, paper CThZ3, 2005.

15. D.T. Walton, S. Gray, L. Zenteno, X. Chen, M. Li, D. Nolan, G. Berkey, J. Wang, J. Koh, W. Wood and C. Tennent, “High Power, Linearly Polarized Yb-doped Fiber Laser”, ASSP 2004.

16. Venkata A. Bhagavatula, Rajaram J. Bhat, George E. Berkey, Xin Chen, Matthew J. Dejneka, Michael T. Gallagher, L. C. Hughes, Karl W. Koch, J. Koh, Ming-Jun Li, Victor X. Liu, Daniel A. Nolan, H. M. Rasmussen, Christine L. Tennent, Natesan Venkataraman, Donnell T. Walton, Jing Wang, Chung-En Zah, and Luis A. Zenteno, “Progress in high-power fiber lasers”, Proceedings of SPIE, Volume 5335 pp. 210-216, 2004.

Proc. of SPIE Vol. 6469 64690H-9

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 08/05/2014 Terms of Use: http://spiedl.org/terms