Difficult to say without more information about the fibre.  Do you have any information about absorption or attenuation at other wavelengths?

Is it single mode or multimode?  Polymer core, polymer clad or all-glass?  What is the core composition and numerical aperture?

In high purity all-glass silica-core fibres, attenuation around 1100 nm is dominated by Rayleigh scatter loss.  This is typically around 1 dB/km at 1000 nm, and decreases inversely as the 4th power of wavelength.  If you know the attenuation at 1000 nm, the attenuation at 1117 nm will be approximately 0.64 times lower.

If you are interested in pure absorption rather than scattering and other attenuation mechanisms, it may be difficult to measure absorption accurately.

In flame hydrolysed silica, often used where short wavelength (e.g. UV) transmission is important, there are very strong hydroxyl absorption bands close to 1384 nm, 1250 nm and  950 nm, with significant absorption at 1117 nm.  In low OH synthetic silica there can be absorption peaks in the visible around 630 nm which are sensitive to fibre drawing conditions.

In polymer clad fibres, there are strong C-H absorption bands in the 1100-1200 nm region.

In all-glass doped silica fibres, hydroxyl and other impurity absorption is generally quite low, so that Rayleigh scattering dominates around 1100 nm.  The scattering coefficient increases with germania doping, so attenuation is generally greater in high numerical aperture fibres.

Beyond 1600 nm, there is an exponentially increasing phonon absorption edge.  For short visible and UV wavelengths there is increasing absorption associated with electronic transitions.

Impurities such as dissolved hydrogen, encountered in some cable environments, or transition metal impurities incorporated during manufacture, have characteristic absorption spectra.

Standard single mode fibres can show excess loss around 1100 nm as both LP01 fundamental and LP11 higher order modes are quided, but the LP11 mode is only weakly guided and is highly sensitive to bending losses.

Hope this helps.

http://www.photonics.com/Article.aspx?AID=55205

http://www.jb.man.ac.uk/research/fibre/intro2fibre.htm

http://www.vialite.com/electro-optical_conversion_process.php

Thanks Mr Alan

It was realy helpful. I want to simulated the effect of Stimulated Raman Scattering in high power fiber lasers. The steady state rate equation are writen in following article and I dont know the exact value of absorption coefficient for Stockes wave(Raman)  and my simulation run when the absorption coefficient is high i.e 0.05 . I apply the RK4 method to simulate it.

I suppose that the value of absorption coefficient for pump, signal and stokes wave are 0.003, 0.005 and 0.05 respectively. Two frist value is correct from some articles but I dont know the third value is correct.

Please guide me

Best regard

Article Analysis of Raman and thermal effects in kilowatt fiber lasers

I can't access a copy of the Wang paper, so I can't comment on the rate equations.  I assume they include Yb absorption and stimulated emission, in addition to SRS effects.

What units are your absorption coefficients? m-1? dB m-1? 

Are you studying SRS in the Yb doped laser fibre?  What are your pump and signal wavelengths?  What is the Yb concentration?

Are these background losses which exclude the much larger saturable absorption due to rare earth (Yb) doping?  Saturable absorption will dominate a conventional low power attenuation measurement, so background loss is typically estimated from measurements outside the rare earth absorption band.

Background loss in rare-earth doped fibres depends on how the fibre is made, and can vary considerably between different fibres. 

Yelen et. al ("Experimentally Verified Modeling of Erbium–Ytterbium Co-Doped DFB Fiber Lasers", J. Lightwave Technology, p 1380, vol 23,  March 2005) measured 0.65 dB/m background loss around 1550 nm for an Er-Yb co-doped fibre.  Losses at 1100 nm could be higher or lower.  It is quite possible that significantly lower background losses are achieved in the Yb-doped kW laser fibres you are studying.

Codemard in his PhD thesis (page 68 in link below) reports background losses between 0.05 and 0.8 dB/m for Yb-Er co-doped fibres measured at 1300 nm.

I assume your signal and pump wavelengths are in the wavelength range 900-1050 nm.  In the absence of additional information, I would assume a similar background loss at signal and Stokes wavelengths.

Note that the effective background loss for a pump propagating predominantly in the inner cladding of a double clad fibre is likely to be much lower than for the same wavelength propagating as a core-guided mode.  Background losses are typically significantly greater in a highly doped core.  The clad-guided properties will be important in determining the Yb population inversion, but the core-guided properties may be relevant for SRS interactions.

Regards,

Alan.

http://www.orc.soton.ac.uk/publications/theses/3810_cac/3810_cac.pdf

http://eprints.soton.ac.uk/38449/

Hi, thanks for your attention. 

The units of absorption coefficent is dBm-1 and I know how change it to m-1.

alpha (dBm-1)= 4.34 alpha (m-1).

Yes I study the effect of nonlinear process in fiber lasers that restrict output power.

The pump and signal wavelength are 915 and 1065 nm, respectively and the raman wavelength is 1117 nm based on a 440 cm-1 frequency shift in germanosilicate fibers. I dont know exact value of absorption coefficient of pump , signal and Stokes wave.

I send you the wang article.

best regard

may this can help

Scattering losses in optic fiber materials. II. Numerical estimates

http://dx.doi.org/10.1063/1.332995

Scattering losses in optic fiber materials. I. A new parametrization  http://dx.doi.org/10.1063/1.332994

If you want to calculate the absorption coefficient by simulating the optical fiber structure, you can use comsol multiphysics. You have to apply PMLs surrounding the fiber geometry. From the imaginary part of simulated values of effective mode index, you can calculate absorption coefficient.

thanks Luo for your comment

thank you Mr. Vishwatosh  

Hey guys,

Do you guys know how to convert the absorption coefficient in (dB/m) to concentration in parts per million (ppm-wt) for erbium doped fibres?

@Asavela Sigonya: in general, the saturable absorption coefficient in erbium-doped fibres depends on:

• The absorption cross-section at the wavelength of interest, σ(λ)
• The erbium-ion concentration and the radial distribution of ions within the core and cladding ρ(r)
• The mode field distribution of the fibre mode ψ(r, λ)

Absorption coefficient α = ∫∫ ψ2(r,φ) ρ(r,φ) σ r dr dφ / ∫∫ ψ2(r,φ) r dr dφ

where the integrals are with respect to area over the the fibre cross-section, here expressed in cylindrical coordinates (r,φ).

If the Er doping is confined to a cylindrical cross-section, concentric with the core, and doping concentration ρ0 is constant within this region then:

α(λ) = ρ0 σ(λ) Γ(λ)

If the uniformly doped region also coincides with the core of a step-index fibre, then the overlap integral, Γ(λ), is simply the fraction of power within the core. It depends only on the normalised frequency (V-number) of the fibre via the wavelength.

More generally you need to know both dopant profile and mode field distributions.

The erbium cross-section varies with both wavelength and with the composition of the host glass. Alumina-doped silica fibres have a different absorption spectrum to germania-doped fibres. Desurvire (1994) cites peak values for absorption cross-section near 1530 nm of 4.7∙10-25 m2 for Ge-Al doped silica and 8∙10-25 m2 for germania-silica fibres.

Given the overlap integral, average concentration: ρ0 = α(λ) / σ(λ) Γ(λ)

Conversion from number density to ppm-wt is straightforward, given appropriate attention to whether you require ppm-wt Er3+, or ppm-wt Er2O3.

There is a useful discussion in Giles & Desurvire, "Modeling erbium-doped fiber amplifiers", J. Lightwave Tech., vol 9, no 2, p 271, Feb 1991. https://www.semanticscholar.org/paper/Modeling-erbium-doped-fiber-amplifiers-Giles-Desurvire/4ce33073a28b9ac2471b2d8ace8db21521513253

More details in "Erbium-doped fiber amplifiers", Emmanuel Desurvire, Wiley, 1994. https://www.wiley.com/en-gb/Erbium+Doped+Fiber+Amplifiers%3A+Principles+and+Applications-p-9780471589778

Thank you so much, much appreciated Alan Robinson Alan Robinson