Detectors and light-sources for optical spectrometry: from a 3D-printed light-source to a self-powered sensor fabricated on a flexible polymeric substrate, and from there on to an IoT-enabled "smart" system

We are developing detectors to sense the visible part of the spectrum. We are also developing light-sources that generate spectral signals from micro-samples introduced (for compatibility reasons) into micro-plasmas. Our battery- operated microplasmas are coupled to a portable, fiber-optic spectrometer and this combination (or system) can be thought of as a "multi parameter" or "multi-element sensor" for the UV and the Vis parts of the spectrum. Initially, our Micro Plasma Devices (MPDs) were fabricated using technologies borrowed from the semiconductor industry (e.g., microfluidics, micromachining) [1] . To reduce fabrication costs and to enable rapid prototyping [2] , we fabricated Micro Plasma Devices using 3D-printing of polymeric materials [3] . We also fabricated (and continue to characterize) a relatively- inexpensive self-powered detector on a flexible polymeric substrate [4] . The detector responds to light from the visible part of the spectrum. To enable portability for chemical measurements on-site (i.e., in the field) we often used a smartphone for data acquisition and signal processing, thus enabling a sensor-system to be placed on the Internet of Things (IoT) and potentially, to be employed in Society 5.0 applications [5] . To further facilitate use on-site (i.e., in the field), portable optical spectrometers with a short focal length must be used. But as focal length decreases, spectral overlaps (often called spectral interference effects) arise. To address them, we employed Artificial Intelligence (AI) methods using Artificial Neural Networks (ANNs) and Deep Learning approaches, thus (in many respects) making sensor-systems smarter [6] . In this paper (due to space limitations), emphasis will be will be placed on recent developments.