Sensor Fusion: Comprehensive Real-Time, On-Line Monitoring for Process Control via Visible, Near-Infrared, and Raman Spectroscopy.

Autor: Lines AM; Energy and Environment Directorate, Pacific Northwest National Laboratory, Richland, Washington 99352, United States., Hall GB; Energy and Environment Directorate, Pacific Northwest National Laboratory, Richland, Washington 99352, United States., Asmussen S; Energy and Environment Directorate, Pacific Northwest National Laboratory, Richland, Washington 99352, United States., Allred J; Energy and Environment Directorate, Pacific Northwest National Laboratory, Richland, Washington 99352, United States., Sinkov S; Energy and Environment Directorate, Pacific Northwest National Laboratory, Richland, Washington 99352, United States., Heller F; Energy and Environment Directorate, Pacific Northwest National Laboratory, Richland, Washington 99352, United States., Gallagher N; Eigenvector Research, Manson, Washington 98831, United States., Lumetta GJ; Energy and Environment Directorate, Pacific Northwest National Laboratory, Richland, Washington 99352, United States., Bryan SA; Energy and Environment Directorate, Pacific Northwest National Laboratory, Richland, Washington 99352, United States.
Jazyk: angličtina
Zdroj: ACS sensors [ACS Sens] 2020 Aug 28; Vol. 5 (8), pp. 2467-2475. Date of Electronic Publication: 2020 Jul 27.
DOI: 10.1021/acssensors.0c00659
Abstrakt: On-line monitoring based on optical spectroscopy provides unprecedented insight into the chemical composition of process streams or batches. Amplifying this approach through utilizing multiple forms of optical spectroscopy in sensor fusion can greatly expand the number and type of chemical species that can be identified and quantified. This is demonstrated herein, on the analysis of used nuclear fuel recycling streams: highly complex processes with multiple target and interfering analytes. The optical techniques of visible absorbance, near-infrared absorbance, and Raman spectroscopy were combined to quantify plutonium(III, IV, VI), uranium(IV, VI), neptunium(IV, V, VI), and nitric acid. Chemometric modeling was used to quantify analytes in process streams in real time, and results were successfully used to enable immediate process control and generation of a product stream at a set composition ratio. This represents a significant step forward in the ability to monitor and control complex chemical processes occurring in harsh chemical environments.
Databáze: MEDLINE