Publications

We demonstrate the ultrahigh extinction operation of a silicon photonic (SiP) amplitude modulator (AM) employing a cascaded Mach-Zehnder interferometer. By carrying out optimization sweeps without significantly degrading the extinction, the SiP AM is robust to environment changes and maintained >52 dB extinction for >6 hrs.

Measurement uncertainties in the techniques used to characterize loss in photonic waveguides becomes a significant issue as waveguide loss is reduced through improved fabrication technology. Typical loss measurement techniques involve environmentally unknown parameters such as facet reflectivity or varying coupling efficiencies, which directly contribute to the uncertainty of the measurement. We present a loss measurement technique, which takes advantage of the differential loss between multiple paths in an arrayed waveguide structure, in which we are able to gather statistics on propagation loss from several waveguides in a single measurement. This arrayed waveguide structure is characterized using a swept-wavelength interferometer, enabling the analysis of the arrayed waveguide transmission as a function of group delay between waveguides. Loss extraction is only dependent on the differential path length between arrayed waveguides and is therefore extracted independently from on and off-chip coupling efficiencies, which proves to be an accurate and reliable method of loss characterization. This method is applied to characterize the loss of the silicon photonic platform at Sandia Labs with an uncertainty of less than 0.06 dB/cm.

Abstract not provided.

Abstract not provided.

Volumetric imaging of focusing waveguide grating coupler emission with high spatial resolution in the visible (λ = 637.3 nm) is demonstrated using a scanning near-field optical microscope with long z-axis travel range. Stacks of 2-D images recorded at fixed distance from the device are compiled to yield 3-D visualization of the light emission pattern and enable extraction of parameters, such as spot size, angle of emission, and focal height. Measurements of such parameters are not prevalent in the literature yet are necessary for efficacious design and integration. As a result, it is observed that finite-difference time-domain simulations based on fabrication layout files do not perfectly predict in-hand device behavior, underscoring the merit of experimental validation, particularly for critical application.

Abstract not provided.