Sensitivity and spectral control of network lasers

Dhruv Saxena, Alexis Arnaudon, Oscar Cipolato, Michele Gaio, Alain Quentel, Sophia N. Yaliraki, Dario Pisignano, Andrea Camposeo, Mauricio Barahona, Riccardo Sapienza

Abstract

Recently, random lasing in complex networks has shown efficient lasing over more than 50 localised modes, promoted by multiple scattering over the underlying graph. If controlled, these network lasers can lead to fast-switching multifunctional light sources with synthesised spectrum. Here, we observe both in experiment and theory high sensitivity of the network laser to the spatial shape of the pump profile, with mode intensity variation of up to 280% for a non-homogeneous 7% pump decrease. We solve the nonlinear equations within the steady state ab-initio laser theory (SALT) approximation over a graph and we show selective lasing of around 90% of the top modes, effectively programming the spectrum of the lasing networks. In our experiments with polymer networks, this high sensitivity enables control of the lasing spectrum through non-uniform pump patterns. We propose the underlying complexity of the network modes as the key element behind efficient spectral control opening the way for the development of optical devices with wide impact for on-chip photonics for communication, sensing and computation.

Introduction

Standard lasers are designed for stability, typically favoring a single, well-defined frequency. In contrast, "random lasers"—which utilize scattering media instead of rigid cavities—offer high-power, low-coherence light but are notoriously difficult to control. Their spectra often consist of multiple fluctuating frequencies, limiting their utility in signal processing, optical computing, or precise sensing. The central challenge lies in achieving spectral selection—picking exactly which frequency lases—within a fundamentally disordered system.

Programming Light via netSALT

To address this, researchers combined experimental nanofiber fabrication with a new mathematical model called netSALT. This framework extends Steady-state Ab-initio Laser Theory (SALT) to complex graphs. Unlike previous models that only considered "passive" modes, netSALT accounts for the nonlinear competition between modes for energy and "spatial hole burning," where one mode exhausts local gain and prevents others from lasing.

The experimental setup utilizes planar networks of dye-doped polymer nanofibers. Control is achieved via a Digital Micromirror Device (DMD), which acts as a high-speed projector to cast specific, non-uniform optical pump patterns onto the network.

Experimental and simulated network laser spectra.

Complexity as a Control Mechanism

A key insight of this work is that the inherent sensitivity of network lasers is a feature, not a bug. The team found that a mere 7% decrease in pump power in a localized area could lead to a 280% change in a specific mode's intensity. Counter-intuitively, more complex networks are easier to control. The vast number of overlapping modes in a complex network provides a rich parameter space that optimization algorithms can exploit.

By strategically withholding light from certain edges (mode reshaping), the researchers can make specific modes "lossy," effectively clearing the stage for a target frequency. Using a greedy, derivative-free iterative algorithm, they searched a massive configuration space ($2^{160}$ patterns) to find optimal pump profiles.

Experimental Validation and Results

Numerically, the netSALT model predicted that ~90% of the top 50 modes in a complex network could be selectively lased. Experimentally, the team successfully demonstrated single-mode lasing at specific wavelengths (e.g., 583 nm or 586 nm) and even bi-modal lasing, where two specific frequencies were amplified simultaneously while others were suppressed. The optimization process was remarkably efficient, typically converging in about 100 iterations.

Experimental spectral control using a Digital Micromirror Device (DMD) and optimization.

Conclusion and Future Directions

While the current netSALT model assumes a steady state and does not yet account for ultra-fast turn-on dynamics, it provides a powerful foundation for programmable light sources. Future work may overcome current resolution limits imposed by DMD hardware and nanofiber size. This approach paves the way for on-chip photonic devices with electronically switchable spectra and potential applications in neuromorphic computing, where nonlinear network interactions could perform complex optical calculations.

Summary updated

Abstract

Recently, random lasing in complex networks has shown efficient lasing over more than 50 localised modes, promoted by multiple scattering over the underlying graph. If controlled, these network lasers can lead to fast-switching multifunctional light sources with synthesised spectrum. Here, we observe both in experiment and theory high sensitivity of the network laser to the spatial shape of the pump profile, with mode intensity variation of up to 280% for a non-homogeneous 7% pump decrease. We solve the nonlinear equations within the steady state ab-initio laser theory (SALT) approximation over a graph and we show selective lasing of around 90% of the top modes, effectively programming the spectrum of the lasing networks. In our experiments with polymer networks, this high sensitivity enables control of the lasing spectrum through non-uniform pump patterns. We propose the underlying complexity of the network modes as the key element behind efficient spectral control opening the way for the development of optical devices with wide impact for on-chip photonics for communication, sensing and computation.
BibTeX
@inproceedings{saxena2022sensitivity,
  author = {Dhruv Saxena and Alexis Arnaudon and Oscar Cipolato and Michele Gaio and Alain Quentel and Sophia N. Yaliraki and Dario Pisignano and Andrea Camposeo and Mauricio Barahona and Riccardo Sapienza},
  title = {{Sensitivity and spectral control of network lasers}},
  booktitle = {Nature Communications},
  year = {2022},
  doi = {10.1038/s41467-022-34073-3},
}
Figure