Title: How microcavities change nonlinear optics
  
   By: Steven G. Johnson(Massachusetts Institute of Technology(MIT) USA)

 Date: May 27, 2008

   
    Abstract:For better or for worse, nonlinearities in electromagnetism are weak effects: typically, changes in the refractive index are much less than 1%. In order to increase the strength of these effects, a common strategy is to confine the light in a small volume for a long time --in a microcavity -- which both increases the intensity of the fields for a given input power and (because of the narrowband resonance) makes the system more sensitive to small changes in the material. But a microcavity can do much more than increase the strength of nonlinear- optical phenomena: it can fundamentally change the consequences of the nonlinearity, introducing qualitatively new phenomena. This was, perhaps, first observed over 30 years ago in the phenomenon of optical bistability for transmission through a nonlinear cavity. I begin by reviewing this well-known effect, recasting its analysis in terms of temporal coupled-mode theory, and mentioning some recent low-power realizations employing photonic crystals. Bistability utilizes nonlinear effects at only a single frequency (self-phase modulation), but I will then describe how even more unusual phenomena can occur when multiple resonant frequencies are coupled nonlinearly via harmonic generation. Not only do doubly-resonant cavities coupled by harmonic generation have a surprising solution where 100% harmonic conversion is possible at low powers (in the absence of dissipation, but including nonlinear downconversion), but even more interesting effects occur when one considers the dynamics. We have found the existence of multistable solutions, bifurcations, transitions between stable and unstable behavior, and even limit cycles (stable optical oscillators that arise spontaneously from a continuous-wave input source). I discuss these behaviors, the phase diagram of the system, ways to excite the different solutions of the coupled nonlinear oscillators, and potential applications of these surprising effects.

    Steven G. Johnson is a professor of Applied Mathematics in MIT. He received three B.S. degrees, in Physics, Mathematics, and Computer Science, in 1995 and a PhD in Physics in 2001 from MIT. Steven G. Johnson joined the Faculty of Mathematics at MIT as Assistant Professor of Applied Mathematics in 2004. He was co-recipient of the 1999 J. H. Wilkinson Prize for Numerical Software for his work on FFTW, a widely used and influential package to compute the fast Fourier transform. He is also the author of other popular software packages in scientific computation, such as his MPB software for modeling photonic crystals (which has received nearly 200 citations in the last three years and averages 400 downloads/month). Besides his efforts in high-performance computation, his work has ranged from the fundamental understanding of incomplete-bandgap systems and bandgap fibers, to the development of new semi-analytical and numerical methods for electromagnetism in high-contrast and periodic systems, to authoring some of the few analytical theorems about the behavior of photonic cyrstals, to the design of integrated optical devices. He is the author or co-author of more than 40 journal articles and holds five issued U.S. patents on nanostructured materials and devices.