Ultrafast Coherent Dynamics of a Photonic Crystal All-Optical Switch

Pierre Colman, Per Lunnemann, Yi Yu, and Jesper Mørk

Phys. Rev. Lett. 117, 233901 – Published 29 November 2016

Abstract

We present pump-probe measurements of an all-optical photonic crystal switch based on a nanocavity, resolving fast coherent temporal dynamics. The measurements demonstrate the importance of coherent effects typically neglected when considering nanocavity dynamics. In particular, we report the observation of an idler pulse and more than 10 dB parametric gain. The measurements are in good agreement with a theoretical model that ascribes the observation to oscillations of the free-carrier population in the nanocavity. The effect opens perspectives for the realization of new all-optical photonic crystal switches with unprecedented switching contrast.

Received 21 April 2016

DOI:https://doi.org/10.1103/PhysRevLett.117.233901

Physics Subject Headings (PhySH)

Light propagation, transmission & absorption Nonlinear waveguides Optical microcavities Optical parametric oscillators & amplifiers Photonic crystals Second order nonlinear optical processes Semiconductor microcavities Ultrafast phenomena

Atomic, Molecular & OpticalNonlinear Dynamics

Authors & Affiliations

Pierre Colman

DTU Fotonik, Technical University of Denmark, DK-2800 Kongens Lyngby, Denmark and IEF Institut d’Electronique Fondamentale, Université Paris-Sud, F 91405 Orsay, France

Per Lunnemann, Yi Yu, and Jesper Mørk^*

DTU Fotonik, Technical University of Denmark, DK-2800 Kongens Lyngby, Denmark

^*jesm@fotonik.dtu.dk

Article Text (Subscription Required)

Click to Expand

Supplemental Material (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 117, Iss. 23 — 2 December 2016

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
(a) Principle of operation of a PhC cavity all-optical switch. (b) SEM image of the $H_{0}$ cavity and its access waveguides. (c) Schematic of the experiment. The picosecond pump and probe are carved from the same 80 fs reference pulse. Heterodyne detection using short reference pulses allows characterizing pulses with energy as low as $40 aJ / pulse$ and achieving 150 fs deconvolution-free temporal resolution.
Reuse & Permissions
Figure 2
Switch-off dynamics. (a) Measured probe trace as a function of the pump-probe delay. $Y$ axis (“Time”) corresponds to the reference-probe delay: The ultrashort duration of the reference pulse implies near-instantaneous gating. The temporal sequences of the reference (gray), pump (red), and probe (blue) pulses at points I–IV are illustrated. (b) Measured differential probe transmission in decibels (transmission with the pump on divided by transmission with the pump off). Pump and probe pulses are set on resonance with the cold cavity and have temporal widths of about 1 and 2 ps, respectively.
Reuse & Permissions
Figure 3
(a) Measured envelope of the idler pulse, obtained by monitoring the signal at the beating frequency associated with the idler, as a function of the pump-probe delay and reference delay. Dashed line indicates a coincidence of the pump and reference pulse peaks. (b) Total idler energy versus pump-probe delay.
Reuse & Permissions
Figure 4
(a) Measured envelope of the idler pulse as a function of the coupled pump pulse energy and reference delay (Time). The pump-probe delay was set at $- 2 ps$ . Red line indicates the temporal mean (first moment). (b) Corresponding theoretical results. (c) Idler energy relative to the input probe pulse energy, obtained by integration of results depicted in (a) and (b) along the time axis. Red circle markers indicate experimental data, and the solid blue line presents the calculated result. The dashed line is a guide to the eye indicating a third-order power scaling ( $y = a x^{3}$ ).
Reuse & Permissions
Figure 5
Measured temporal traces of the idler (red), as well as the probe with (blue) and without (gray) the presence of the pump. The pump power is about $120 fJ / pulse$ , and the pump-probe delay is set to $- 3 ps$ .
Reuse & Permissions
Figure 6
(a) Color map of the probe temporal trace for various pump-probe delays. The pump power is about $130 fJ / pulse$ . (a) Experimental map. (b) Simulation wherein the FCD due to the fast free-carrier plasma is neglected. (c) Simulation wherein the Kerr effect is turned off. (d) Full model simulation including both FCD and Kerr effects. When no nonlinear effects are present, the probe is a trace that looks like the one seen at large negative pump-probe delay in (a)–(d).
Reuse & Permissions

Physical Review Letters