Research Interests

Polarization properties of crystalline WGM resonators

Currently most WGM resonators are fabricated fully symmetric to their rotational axis. In WGM resonators fabricated from uniaxial crystals this symmetry axis then coincides with the optic axis, such that the modes are either parallel or perpendicular polarized. If the optic axis is however tilted with respect to the symmetry axis the polarization of the modes changes dramatically. A complex type of mode is observed that can be fully coupled (decoupled) with a right (left) hand circular polarized beam of light. Furthermore, the polarization properties at diff erent outcoupling positions, determined via full Stokes measurements, are recorded and show a continuous complex change in ellipticity. We present the experimental results. Understanding the polarization behavior in an off -axis, birefringent WGM resonator may o ffer a new way for phase-matching in non-linear $\chi^{(2)}$ materials ArXiv (2013).

Sub-kHz linewidth laser

WGM between two prisms Standard fiber ring lasers are cheap, but suffer from unstable lasing behavior and a relative large linewidth. Utilizing a millimeter sized diamond turned CaF2 WGM resonator as a passive filtering element in an erbium-doped fiber ring laser we achieve single mode lasing. This system sustains single mode lasing without the usage of active stabilization techniques which we characterized with a three-cornered-hat measurement to have a linewidth below 13kHz, Optics Letters (2010). These measurements were limited by the reference lasers used. Currently we have build an identical WGM system and together with a commercial system have measures the relative frequency stability to be below a 700 Hz ArXiV (2012).

Microcavity stabilized Fiber Ring Laser

Fiber Ring LaserIn an earlier experiment we utilized a quartz microsphere of 100 μm diameter. This sphere was molten by a CO2 laser and featured Q factors around 107. Evanescent coupling to the sphere was achieved by a tapered fiber and the out-coupling via an angle polished fiber. The setup was not as stable as the prism setup but nevertheless reduced the linewidth by five orders of magnitude to just below 170 kHz, Optics Letters (2009).

Coupled Microresonators and Emission Properties in the THz domain

two coupled dielectric disks Single dielectric whispering gallery mode resonators have been thoroughly studied. They have an interesting mode spectrum, quite similar to the energy levels of an single atom. Bringing two atoms close to each other, perturbs their mode-field and induces due to the symmetry breaking a mode-splitting. Up to now such a mode splitting had not been observed on two identical whispering gallery mode resonators in the optical domain, as the requirements on the shape are stringent. We have fabricated two sets of identical resonators in the THz regime, which due to the wavelength of THz radiation (30 mu - 3 mm), was possible. Clear mode splitting was observed and reported in Optics Express (2008).

Furthermore we study the emission directionality of polymer WGM resonators with finite scatterers, yielding surprisingly high Q factors and directional emission properties ArXiV (2013)

Wave-chaos in Microresonators

Chaotic Wavefunction in the Stadium Billiard Wave-chaos studies the behavior of systems, with a chaotic ray-dynamics in the short wavelength limit. We have studied the emission properties of slightly deformed polymer microcavity lasers and found that the lasing directionality was in perfect agreement with the short time properties of the phase space flow in ray picture. This flow is given by the unstable manifolds of the unstable fixed-points in the ray-dynamical map and we were able to directly relate it to experimental near and farfield data. This work has been published in JOSA (2004)

Numerical methods for dielectric resonators

 Resonance in six random positioned dielectric disks Finding resonances in open, dielectric resonators is time and memory consuming with the standard FDTD approach, which has to discretize the whole space. Boundary methods avoid such problems and only need to discretize the problem on the boundary, lead however to a complex root search. We have devised a method of formulating the equations in term of an eigenvalue problem and by observing the behavior of the eigenvalues with respect to small changes of the wavenumber, we have devised an efficient root-search algorithm. I have worked on this with close collaboration with Hakan Türeci at the ETH Zürich. Recently I started a collaboration with Chris Poulton at the UTS in Australia to investigate multiple randomly positioned dielectric discs, recently published in Optics Express (2009).

Goos-Hänchen Shift

I am interested in the fundamental properties of the reflection of light at a dielectric interface. When light from a denser dielectric material impinges on a less dense material, there is an angle beyond which light is totally internally reflected. Newton already proposed in his Opticks (1704) that light impinging under such an angle would show a shift with respect to the specularly reflected light. In 1947, Goos and Hänchen were the first to measure this shift. However, as the shift is only of the order of the wavelength, they used multiple reflections to obtain a result they could measure.

Our group has devised an experiment that allows the shift to be directly measured in the optical regime after only one reflection following an idea similar to the one behind Goos and Hänchen's original work, and we have published it in Optics Letters (2008), you can also get this pdf.

Further experiments will investigate the reflection properties of a non-linear medium instead of air.

Setup for the Goos-Hänchen Shift

Some nice Pictures from my Ph.D.