Martin Schreivogel

Comparison of fabrication methods for microstructured deep UV multimode waveguides based on fused silica

The suitability of materials for deep ultraviolet (DUV) waveguides concerning transmittance, fabrication, and coupling properties is investigated and a fused silica core/ambient air cladding waveguide system is presented. This high refractive index contrast system has far better coupling efficiency especially for divergent light sources like LEDs and also a significantly smaller critical bending radius compared to conventional waveguide systems, as simulated by ray-tracing simulations. For the fabrication of 300-ffm-thick multimode waveguides a hydrouoric (HF) acid based wet etch process is compared to selective laser etching (SLE). In order to fabricate thick waveguides out of 300-ffm-thick silica wafers by HF etching, two masking materials, LPCVD silicon nitride and LPCVD poly silicon, are investigated. Due to thermal stress, the silicon nitride deposited wafers show cracks and even break. Using poly silicon as a masking material, no cracks are observed and deep etching in 50 wt% HF acid up to 180 min is performed. While the masked and unmasked silica surface is almost unchanged in terms of roughness, notching defects occur at the remaining polysilicon edge leading to jagged sidewalls. Using SLE, waveguides with high contour accuracy are fabricated and the DUV guiding properties are successfully demonstrated with propagation losses between 0.6 and 0:8 dB=mm. These values are currently limited by sidewall scattering losses.

Modeling of ion drift in 4H-SiC-based chemical MOSFET sensors

The effect of mobile ions on electrical performance in ion-sensitive metal–oxide–semiconductor field effect transistor fabricated on 4H silicon carbide for the application as chemical fluid and gas sensors in harsh environments was investigated. The drift and diffusion of these mobile ions in the dielectric gate stack were identified as the source for a change in the sensor signal. The movement of the ions and the resulting electrical properties were successfully modeled using a novel drift–diffusion model implemented in tcad simulation software. The diffusion coefficient and activation energy for drift and diffusion of sodium through an amorphous silicon nitride layer were estimated from these simulations.