Microscale Infrared Technologies for Spectral Filtering and Wireless Neural Dust

Pivotal technologies such as optical computing, autonomous vehicles, and brain-machine interfaces motivate microscale infrared devices. This thesis explores two examples: subwavelength grating (SWG) filters and wireless neural dust (WND). SWGs possess unique imaging and sensing properties, while WND promises advanced neural recording and therapeutic interventions.

First, this work extends the capabilities of SWG infrared filters. Novel grating geometries access quasi-bound states in the continuum. By coupling to these fascinating states, asymmetric SWG geometries enable exceptionally narrowband filtering at configurations that are otherwise inaccessible. Experimental bandpass filters exhibit some of the narrowest linewidths observed in mid-wave infrared SWGs. Fully polarization-independent SWG filters are also demonstrated.

Via near-infrared (NIR) power and data transfer, WND avoids invasive percutaneous wires by leveragingbiological tissue’s high NIR transparency. Digital microLED pulses uplink recorded neural signal. The proposed devices meet ambitious targets for pulse detection efficiency (>99%) and package size (200x170x150 μm^3). Efficient pulse detection is achieved with a unique optical setup and by modeling NIR propagation in tissue. Advanced microfabrication techniques heterogeneously integrate GaAs optoelectronics chips, Si CMOS circuitry, and carbon fiber probes within a microscale package.  Developments include GaAs through-wafer vias, thinned wafer processing, and microscale heterogeneous bonding.

Co-Chairs: Jamie Phillips and David Blaauw