Article Taming of random lasers
Dear Colleagues,
The recent publication of “Taming random lasers” in Nat. Photonics encouraged me to start an informal discussion about random lasers. I consider this paper an excellent work and the following discussion is general to the topic, simply inspired by this publication.
Usually RLs are treated as exotic devices and their understanding requires complex theoretical approaches that put this laser sources away from classical optical resonators.
Is this assumption unavoidable? I would like to comment with you the attached figure. Consider two classical laser structures, the Fabry Perot (FP) and the Distributed Feedback DFB laser. The disordered counterpart of the DFB is the commonly used architecture of a RL: instead of a periodically modulated refractive index in gain medium, we have a randomly varying refractive index.
If we change the mirrors of a FP laser for two different scattering surfaces, we observe random lasing with the same random spectral signature exiting the two different scattering surfaces. Our group has demonstrated this in different devices. I don’t want to make publicity to our work here and I am not using this discussion for this purpose. I would like to share with you the analogy with classical lasers, which is resumed in the attached figure, and I would be very glad to receive your comments and ideas.
Hope to start an interesting and fruitful debate with you.
Antonio Consoli,
well, of course, random lasers are conceptually the same as the other types that you listed -- in all cases, one needs an eigenmode of the system (mathematically speaking, a pole of the S-matrix in the complex frequency plane), and a gain medium. Then, the physics of lasing (i.e., the interaction Hamiltonian) is identical in all systems.
What's different, is the nature of the eigenmodes. And in random lasers they are particularly challenging to find and to control, hence random lasers often require more complex theoretical approaches to find these modes and to tailor their characteristics.
https://www.oca.eu/IMG/ppt/WGuerin_RandomLaser_HBTmeeting.ppt.
Dear Denis G. Baranov ,
Dear Denis,
Thank you very much for your reply. Your analysis well applies to FP and DFB. In these cases, the cavity is well defined, eigenfrequencies are easily found and lasing emission is understood, with longitudinal and transverse modes.
As you say, in RLs, eigenmodes are hard to find and the physical interpretation of random lasing is far from the well understood view of FP and DFB. RLs are usually systems with distributed feedback (lower row on the right). Consider now a system in which gain medium is placed between two scattering media, e.g. two different random distributions of scattering particles.
This system emits the same random frequencies from both scattering ensembles. If we change the scattering pair, emission also changes, but it is still the same spectral signature from both. This recalls a FP laser in which mirrors have been changed with scattering volumes (same results are obtained with surfaces).
Why frequencies are the same if the scattering ensembles are different? In RLs with distributed feedback the spectral signature is specific to the random distribution considered. In the matrix based approach you resume, a set of eigenmodes is associated to a specific random spatial distribution of refractive index.
The interpretation we gave is the following.
Light travels and is amplified along the gain medium, when it reaches a scattering surface it is back scattered again into the gain medium. Consider that the phase contribution of the scattering surface at each wavelength is unpredictable (would be a 2pi in a mirror for all frequencies). Light travels again in the gain medium and reaches the other scattering surface, it is back scattered again and re-enters active material, closing a round trip. If in this loop the total phase accumulated is an integer number at a specific wavelength, then we have an allowed mode in the cavity at that wavelength.
In other words, we model the scattering surfaces as mirrors with frequency varying phase with unknown spectral profiles. At each wavelength, there is the possibility that the phase contribution from both scatters allows a mode in the cavity. It is clear that some back reflections would be directed away from the amplifying path of the gain material, however some light reaches both cavity ends, and we can observe the same spectrum from both.
More experimental and modeling details can be found here: www.nature.com/articles/srep16848
Doufnoune RACHIDA Rachida indeed a good presentation, however I consider the debate is still open. From slide #5 " The nature of the ‘modes’ has been a long debate in the last years… " and so the understanding of random lasers
Absolutely, the paper you cite from 1967 shows a similar structure. However, there was no evidence of resonant random lasing, which came with Cao, H. et al. Appl. Phys. Lett. 73, 3656–3658 (1998).
Nicolai Morten Hoinka
, I don't know if some advance has been made in understanding modes in the diffusive case. The definition of modes for random lasers is sometimes different from classical cavities, which in my opinion is a confusing approach.Basically, the question is: why we have narrow resonances at random frequency positions ? I invite you to take a look at https://www.nature.com/articles/srep16848
and share any comment/criticism
- Badges
- Science topic
- Similar topics
- Engineering
- Electrical Engineering