Connecting Theory and Practice in Optoelectronics
NUSOD 2017 Preview: Theory and simulations of self-pulsing in photonic crystal Fano lasers
2017-05-16Posted by on
Pulsed lasers are utilised in a wide variety of applications, especially optical communication systems, and in particular lasers that support passive pulse generation (self-pulsing). The phenomenon of self-pulsing was exclusively found in macroscopic lasers until recently, where self-pulsing in a microscopic photonic crystal Fano laser was reported .
The Fano laser (fig. 1) consists of a line-defect waveguide in a 2D photonic crystal membrane coupled to a nearby point-defect, with active material embedded directly in the membrane. This coupling yields a strong, narrowband suppression of transmission, due to the interference of the continuous waveguide modes with the discrete mode of the nanocavity, effectively forming the right-most laser mirror at the symmetry line, with the left formed by termination of the waveguide .
The system is modelled by a combination of coupled-mode theory and conventional rate equations, in order to describe the complex interference between the nanocavity and waveguide fields, while also accounting for the laser dynamics of the waveguide field and the saturable absorption of the nanocavity.
The laser has either continuous wave (CW) or pulsed output, depending on both passive system parameters and the driving current, as shown in fig. 2, where yellow represents self-pulsing, blue is CW output, and insets show the temporal output power evolution.
Due to the active material in the nanocavity the reflection coefficient depends upon the nanocavity free carrier density, meaning that a spike in laser intensity leads to a decrease in nanocavity absorption due to saturation, which in turn yields a larger reflection coefficient, allowing the laser field to increase, further saturating the nanocavity absorption. In this way the nanocavity functions as a highly-dispersive semiconductor saturable absorber mirror, which forms a positive feedback loop for the laser field, allowing for passive pulse generation and existence of a dynamical lasing equilibrium with pulse repetition rates on the order of 10 GHz and pulsewidths around 10 ps.
The stability of this feedback loop depends sensitively on several system parameters, which define the phase space of self-pulsing operation, as in fig. 2. Furthermore, dynamical perturbations of the system, e.g. tuning of the nanocavity resonance frequency or varying the drive current, can result in phase transitions between CW and pulsed output (figs. 3 and 4), so that the type of laser output can be controlled dynamically.
Additional details of the theoretical model, the self-pulsing mechanism and the dynamical phase transitions will be presented at NUSOD 2017, as well as some perspectives for applications such as sub-picosecond pulse generation and transistor-like operation (paper TuA1).
: Y. Yu, W. Xue, E. Semenova, K. Yvind, and J. Mork, Nat. Photon. 11(2), 81–84 (2017), Letter.
: J. Mork, Y. Chen, and M. Heuck, Phys. Rev. Lett. 113(Oct), 163901 (2014).