Erbium-doped fibre can be pumped at various visible wavelengths, including 510 nm, 532 nm and 665 nm. Usually, 980 nm or 1480 nm bands are generally preferred to minimise excited state absorption at the pump wavelength https://www.fibercore.com/expertise/fiberpaedia/erbium-er

Article Operation of erbium-doped fiber amplifiers and lasers pumped...

I am not familiar with much of the fibre laser literature, but it is possible to achieve better than 10-15% QE in a fibre laser - see for example: Article 75 W 40% efficiency single-mode all-fiber erbium-doped laser...

Good quantum efficiency is possible at lower output powers using more conventional (smaller mode field diameter) erbium-doped fibres.

What output power and what wavelengths do you need from your diode-pumped fibre laser? Is there a reason (such as limits on electrical power) why you require high QE, rather than simply specify output power?

@ Alan Robinson:

Great to know you. So much appreciated with your good guidance to me.

I may share you with a few points you replied to me and asked being:

- Currently, output power is not the main concern in my Lab works. One among those my future works is to maximize the QE by using LED pump diodes to replace Laser diodes so as to save diode's source power dissipations if possible.

- The workable wavelengths of my Lab works are commonly ranged in GREEN (520 - 540 nm) and RED (635 - 650 nm) solid-state diodes.

Depending on how much power you want you could try Yb/Pr co-doped upconversion fiber lasers which emit in the red (635 nm), orange (605 nm), green (520 nm), or blue (491 nm) spectral region. https://www.rp-photonics.com/upconversion_lasers.html

The QE of a Yb:fiber laser can be well over 50% but the wavelength (1 - 1.1 microns) might not be right for you. You could always make the fundamental and Raman shift (which can be done with very high efficiency) to get to other (longer IR) wavelengths. The problem with LED pumping is going to be pump source brightness. You may have to use a novel pumping architecture like the Hamamatsu disc fiberConference Paper 1kW CW output from fiber-embedded disk lasers

Hi John F Black,

Thanks indeed for your great teaching and opening the possibility for my future works.

As John F Black points out, LED pumping of single transverse mode fibre lasers is problematic. Coupling efficiency into the core will be low from a spatially incoherent LED source.

Yb co-doping of Er-doped fibre is widely used in a clad-pumped configuration, in which the laser/signal is confined to the core, but pump light is guided in an inner cladding region. This allows pumping by 900-980 nm broad stripe diode lasers which are efficient, but difficult to couple to a single mode waveguide. High output powers and high efficiency are possible.

https://www.fiberlabs.com/glossary/erbium-doped-fiber-amplifier/

https://www.rp-photonics.com/double_clad_fibers.html

If your intent is to minimise power dissipation in the diode source, optical quantum efficiency of the laser may not be the best metric. You also need to consider quantum efficiency and wavelength of the diode pump, and coupling efficiency to the laser fibre or cavity. Note that for the same QE and drive current, a 530 nm diode will generally dissipate more power than a 980 nm diode, due to the higher band gap and hence a higher voltage drop.

Hi Alan Robinson,

Having read your kindly feedbacks to me, I am indeed curious and wanted to try my future efforts on improving the total system efficiency (e.g. pump diode QE, total system power and more).