Titolo della tesi: Designs and dynamics of miniaturized optical frequency comb sources
This PhD thesis aims to propose suitable solutions for the state of the art reasearch of optical
frequency comb (OFC) technology. In particular, I set up a strategy for answering to the constant
demand of lowering the operation power thresholds of passive nonlinear devices, towards the on-chip
technology integration. From one side, I individuate photonics crystals (PhC)s as the smallest resonators
suitable for the OFC demonstration. From the other side, I elaborate novel OFC generation
schemes based on a well known nonlinear mean eld model, namely the Lugiato-Lefever equation
(LLE).
While a recent demonstration of PhC optical parametric oscillator (OPO) paved the route towards
the demonstration of PhC-OFC, there are still some open problems highly motivating the
research presented in this manuscript. The PhC-OPO design is conceptually novel and it results in
a quantum harmonic oscillator like system, with a comb of resonant Bloch eigenmodes. In the seek
of generalization, I set up a design technique capable of tailoring dispersive eects for a wide class
of PhC resonators. The problem of tailoring dispersion in PhCs at the design stage results from
the lack, in literature, of a model directly and accurately treating dispersive terms. I do that by
considering a simple one dimensional approximation of eventually complex PhC geometries. The
high computational efficiency of the resulting solver is a critical requirement for the development
of optimization design algorithms, calling the cost function hundreds or thousands of times. The
optimized geometries I propose exhibits a at dispersion within an error of just few GHz over a
modal distribution of seven modes. I believe that this optimization design technique is suitable for
the OFC-PhC demonstration.
Once the dispersion of resonators is treated and atten out, I proceed with a theoretical study of the
mean field Lugiato-Lefever equation in a novel generation scheme of a chirped pulse pumping. Interestingly,
I found out how the kerr soliton dynamics can be completely controlled by an appropriate
tuning of the chirp parameter. Moreover, a proper chirping could result in a fast convergence
towards stable dynamical attractors of single or multi-soliton states with a wide spectral band.