Time-Wavelength Reflectance Maps of Photonic Crystal Waveguides: A New View on Disorder-Induced Scattering

Reducing disorder-induced losses for slow light photonic crystal waveguides through Bloch mode engineering

“…In the slow light regime, the Bloch modes feel these scatterers much more due to their increased optical path length and an increase in the local density of states. Thus, the slow light regime exhibits large scattering losses that have been observed experimentally [9,10] and explained theoretically [8,[11][12][13][14]. For a fixed group velocity, the underlying Bloch mode affects the amount of scattering loss [8].…”

Light transport regimes in slow light photonic crystal waveguides

“…[37][38][39] A striking features in the evolution of the dispersive part of the field is its steep extinction. This contrast with the progressive linewidth broadening at the band edge expected when the disordered potential and the physical origin of the band edge are uncorrelated.…”

Slow Light Propagation and Disorder-Induced Localization in Photonic Crystal Waveguides

“…In practice, scattering losses are observed that are attributed to unavoidable structural imperfections, collectively termed "disorder." Indeed, it is now well established that slow-light PC waveguides suffer from significant losses attributed to scattering at disordered surfaces and other device imperfections (Kuramochi et al 2005;Parini et al 2008). Rigorous modeling of this generally undesired extrinsic scattering phenomena is essential for understanding the underlying physics of measurements and for eventually producing functional devices.…”

“…(Kuramochi et al 2005;O'Faolain et al 2007;Engelen et al 2008), except that they naturally break down at extremely low group velocities where using the ideal band structure is no longer a good approximation (Pedersen et al 2008). Recent experimental measurements have reported interesting features such as narrow-band resonances near the band edge (Topolancik et al 2007;Parini et al 2008) that are not explained at all by the incoherent scattering theory, and simple v g scaling rules do not make sense. Also the previous loss formulas employ an expectation-value calculation and, as such, they should be compared with an ensemble average of loss measured on a set of nominally identical structures and omit zero-mean features visible in individual samples.…”

“…Properties such as group velocity and complex frequency-dependent reflectance can be extracted from the interference pattern. Here, we also present a powerful and recently developed analytic technique: time-frequency reflectance maps (TFRMs) (Parini et al 2008), that can be used to visualize the frequency-dependent impulse response and reveal a number of interestOur formalism also provides fresh insights into the long standing question of light localization in PC waveguides. It was proposed by John (1984) and Anderson (1985), that strong localization may be observable in PCs.…”