Electrically pumped network random lasers

Project Information

EPNRL

Grant agreement ID: 800410

DOI

10.3030/800410

Project closed

EC signature date 1 March 2018

Start date 16 July 2018

End date 15 July 2020

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 195 454,80

EU contribution

€ 195 454,80

Coordinated by

IMPERIAL COLLEGE OF SCIENCE TECHNOLOGY AND MEDICINE
United Kingdom

Periodic Reporting for period 1 - EPNRL (Electrically pumped network random lasers)

Reporting period: 2018-07-16 to 2020-07-15

Summary of the context and overall objectives of the project

Random lasers, based on disordered photonic materials with optical gain, are alternative means to generate laser light. They are typically cheaper and simpler to fabricate than conventional lasers, and can be potentially controlled by non-uniform pumping to emit light at a particular frequency or direction, which can be exploited to make complex integrated light sources. However, most random lasers demonstrated are optically pumped, whereas electrical pumping is needed for practical use in technology.

The ambition of this MSCA Individual Fellowship (EPNRL - Electrically pumped network random lasers) was to develop electrically pumped random lasers made of semiconductor gain medium that could be integrated on-chip and be dynamically controlled by inhomogeneous pumping. Such a device would be able to switch between different lasing modes at fast speeds and route laser light at different wavelengths out into different waveguides. This functionality would be particularly useful for information processing and help advance the development of optical computing, enabling faster and more energy efficient processors.

The proposed idea in this project was to construct novel lasers from a network of III-V semiconductor waveguides. Optical feedback in a network would be obtained by light propagation and interference over multiple paths and recurrent loops. As a prototype, we first demonstrated a network random laser in a self-assembled network made from polymer nanofibers with dye. The objective of this fellowship was to develop network lasers further, fabricating them out of semiconductor gain materials so that they would be more robust (longer device lifetime) and compatible with established photonic chip technology (for electrical injection and integration with passive waveguides). Another objective of the project was to develop a theoretical model using graph theory, which would describe lasing in complex networks, and use this model to design the network lasers by investigating different network topologies. As the semiconductor networks were to be fabricated by lithography, the design of the random lasers could be optimised and enable novel characteristics that could be predicted through the modelling and tested experimentally by inhomogeneous pumping.

The fellowship was indeed successful in developing Indium Phosphide network lasers (optically pumped) and showing that network lasers could be controlled to lase in one, two or more arbitrary modes via inhomogeneous optical pumping (both experimentally and theoretically). While further research and engineering is required to make network lasers electrically pumped, the network approach certainly provides a flexible architecture to design, build and realise complex lasers from very simple elements, such as 1D waveguides.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

There were three key aspects to EPNRL: design, fabrication and characterisation.

Design involved optimising both the individual components (waveguides) as well as the network structure to reduce the threshold of network lasers. For the waveguide design, we used finite-difference time-domain (FDTD) simulations to optimise the waveguide dimensions to reduce scattering losses at network nodes, which are the main source of loss in the network. For the network design, we used a home-built code based on solving Maxwell’s Equations on a graph, to model the passive modes in networks and estimate their Q factors. This code, however, did not include gain and mode competition, which is necessary for modelling lasing. So later on in the project, through a collaboration, we extended the graph model to include gain (netSALT), and used it for modelling lasing in network lasers with non-uniform pumping and for comparing lasing in different network structures.

The fabrication of semiconductor networks involved developing the process and optimising the steps to achieve high quality structures. We focused on making networks with Indium Phosphide (InP) because of its low surface recombination and thus high optical quality. During the project, we developed the process for fabricating high-quality InP networks, optimising the lithography and reactive ion etching steps to minimise the waveguide roughness and reduce ion damage. This part of the project took longer than expected, as it needed access to facilities (affected by closures and equipment downtimes) and materials (purchasing of wafers, resist and resins). During the project, we also collaborated with many external groups, with established III-V fabrication labs, to get advice and help with the fabrication. These collaborations were fruitful in obtaining first working samples, which exhibited optically pumped room temperature lasing (manuscript under preparation).

We characterised lasing in InP networks by optical pumping using a fs-pulsed laser equipped to a microscope. During the fabrication process development, we used micro-photoluminescence to assess the optical quality of etched structures and scanning electron beam to measure the sidewall roughness and dimensions of fabricates structures. We also performed experiments to illuminate the network lasers with a spatially non-uniform pump profile, using a digital micromirror device (DMD), and demonstrated spectral control in the polymer nanofiber network lasers (manuscript under preparation).

A summary of the lasing results from InP networks is shown in the attached figure. Panel a shows a top view image of a fabricated InP network with the intensity profile of a network mode overlaid on top. Schematic of the InP waveguide cross-section is given in the inset. Panel b shows the optical image of the InP network laser above threshold and panels c-d show the lasing spectrum and total intensity from the network as a function of pump fluence, respectively.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The success of EPNRL, beyond its scientific achievements, were the fruitful collaborations that it started. EPNRL brought together experts in network theory, photonics, and semiconductor devices and this collaboration resulted in two other successful grants, CORAL (MSCA-ITN) and ‘Semiconductor lasers on a graph’ (EPSCRC), funding two PhD students and two postdocs to build upon the results from this fellowship and develop it even further. Research in network lasers is still at its infancy and there is great scope for advancing both fundamental knowledge as well as the technology.

One of the significant outputs from the collaborations established during the project was netSALT, which is an adaptation of steady-state ab initio laser theory (SALT) on a graph. netSALT was used in the project to model lasing in network lasers under both uniform and non-uniform pumping, and demonstrate the capability of single-mode lasing in network lasers through optimised illumination. In addition, with netSALT we could better understand the rich physics in network lasers, such localisation of modes, high sensitivity to local perturbations in the pump and relationship between network topology and spectral control. We anticipate that netSALT will be useful for the wider laser community as a modelling tool and will make the code available on GitHub once the project results are published.

Room-temperature lasing from Indium Phosphide networks demonstrated by optical pumping.