One-dimensional photonic crystals for eliminating cross-talk in mid-IR photonics-based respiratory gas sensing

摘要

Mid-IR carbon dioxide (CO2) gas sensing is critical for monitoring in respiratory care, and is finding increasing importance in surgical anaesthetics where nitrous oxide (N_20) induced cross-talk is a major obstacle to accurate C0_2 monitoring. In this work, a novel, solid state mid-IR photonics based C0_2 gas sensor is described, and the role that 1-dimensional photonic crystals, often referred to as multilayer thin film optical coatings [1], play in boosting the sensor's capability of gas discrimination is discussed. Filter performance in isolating C0_2IR absorption is tested on an optical filter test bed and a theoretical gas sensor model is developed, with the inclusion of a modelled multilayer optical filter to analyse the efficacy of optical filtering on eliminating N_20 induced cross-talk for this particular gas sensor architecture. Future possible in-house optical filter fabrication techniques are discussed. As the actual gas sensor configuration is small, it would be challenging to manufacture a filter of the correct size; dismantling the sensor and mounting a new filter for different optical coating designs each time would prove to be laborious. For this reason, an optical filter testbed set-up is described and, using a commercial optical filter, it is demonstrated that cross-talk can be considerably reduced; cross-talk is minimal even for very high concentrations of N_20, which are unlikely to be encountered in exhaled surgical anaesthetic patient breath profiles. A completely new and versatile system for breath emulation is described and the capability it has for producing realistic human exhaled C0_2 vs. time waveforms is shown. The cross-talk inducing effect that N_20 has on realistic emulated C0_2 vs. time waveforms as measured using the NDIR gas sensing technique is demonstrated and the effect that optical filtering will have on said cross-talk is discussed.