Desarrollo de un sensor DOAS UV para el monitoreo de la concentración de dióxido de azufre en un reactor de lecho fluidizado.
Loading...
Date
2025
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad de Concepción
Abstract
La preocupación global por el impacto ambiental de las emisiones industriales ha impulsado el desarrollo de nuevas tecnologías de monitoreo de gases contaminantes. Entre estos, el dióxido de azufre (SO2) es especialmente relevante debido a sus efectos nocivos sobre la salud humana y su contribución a fenómenos como la lluvia ácida. En procesos metalúrgicos, particularmente en la fusión de concentrado de cobre, el SO2 se genera en altas concentraciones y su correcta medición resulta fundamental tanto para el control operacional como para el cumplimiento de normativas ambientales. Actualmente, el seguimiento de este gas en plantas industriales suele realizarse mediante instrumentación compleja o equipos de alto costo, lo que motiva la búsqueda de soluciones alternativas basadas en técnicas ópticas. Este proyecto presenta el desarrollo e implementación de un sistema experimental para la medición en línea de SO2 utilizando espectroscopía de absorción óptica diferencial en el ultravioleta (DOAS). El sistema se compone de una fuente de luz UV basada en LEDs, una celda de medición de 30 cm, una fibra óptica de acoplamiento y un espectrómetro HR4000 de Ocean Optics conectado a un computador. Se diseñó un algoritmo en MATLAB para adquirir espectros en tiempo real, calcular absorbancia y estimar concentración de gas aplicando la ley de Lambert–Beer. El diseño del sensor y del algoritmo fue validado experimentalmente mediante tres etapas: pruebas preliminares con colorantes líquidos, una prueba funcional en un reactor de lecho fluidizado y ensayos controlados con cilindros presurizados de SO2. Los resultados demostraron que el sistema es capaz de detectar cambios en la concentración del gas, evidenciando la atenuación progresiva del espectro UV conforme aumenta la presencia de SO2. Sin embargo, se observó saturación óptica en concentraciones superiores al 20 %, debido a que el trayecto óptico de 30 cm resultó excesivo para el rango de medición, provocando una absorción total de luz. Adicionalmente, se comprobó que la sección transversal del SO2 varía con la longitud de onda, lo que limita la aplicación directa del modelo teórico de Lambert–Beer. Para corregir estas discrepancias, se implementó un modelo empírico basado en interpolación utilizando datos de calibración con concentraciones conocidas, mejorando significativamente la correspondencia entre los valores medidos y los reales. Los experimentos realizados permiten concluir que el sistema DOAS desarrollado es funcional y sensible para la detección de SO2, validando el principio de medición óptica en tiempo real. Si bien el sistema no alcanzó precisión cuantitativa en todo el rango de interés, establece una base sólida para futuras mejoras en el diseño óptico y en el procesamiento espectral.
The global concern regarding the environmental impact of industrial emissions has driven the development of new technologies for monitoring atmospheric pollutants. Among these, sulfur dioxide (SO2) is particularly relevant due to its harmful effects on human health and its contribution to acid rain. In copper smelting processes, high concentrations of SO2 are released, making its measurement crucial for operational control and environmental compliance. Traditional monitoring systems often rely on complex or high-cost equipment, motivating the exploration of alternative optical approaches. This project presents the development and implementation of an experimental system for real-time measurement of SO2 using ultraviolet Differential Optical Absorption Spectroscopy (DOAS). The system consists of a UV LED light source, a 30 cm gas measurement cell, a coupling optical fiber, and a spectrometer connected to a computer. A MATLAB algorithm was developed to acquire spectra in real time, compute absorbance, and estimate gas concentration based on the Lambert–Beer law. The sensor and processing algorithm were experimentally validated in three stages: preliminary liquid-phase tests, a functional test on a fluidized-bed reactor, and controlled tests using pressurized cylinders of SO2. Experimental results showed that the system successfully detected variations in SO2 concentration, evidenced by a progressive attenuation of the UV spectrum as gas concentration increased. However, optical saturation occurred at concentrations above 20 %, due to the excessive 30 cm optical path length, leading to total light absorption. Furthermore, it was confirmed that the molecular absorption cross-section of SO2 varies with wavelength, limiting the direct application of the theoretical Lambert–Beer model. To address these discrepancies, a nonlinear empirical interpolation model was iii Abstract implemented using calibration data, yielding a better agreement between measured and real concentrations. The results validate the functionality and sensitivity of the developed DOAS system for real-time detection of SO2. Although quantitative accuracy was not achieved across the full concentration range, the work establishes a solid foundation for future improvements in optical design and spectral processing, advancing toward a robust and low-cost sensor for industrial environmental monitoring applications.
The global concern regarding the environmental impact of industrial emissions has driven the development of new technologies for monitoring atmospheric pollutants. Among these, sulfur dioxide (SO2) is particularly relevant due to its harmful effects on human health and its contribution to acid rain. In copper smelting processes, high concentrations of SO2 are released, making its measurement crucial for operational control and environmental compliance. Traditional monitoring systems often rely on complex or high-cost equipment, motivating the exploration of alternative optical approaches. This project presents the development and implementation of an experimental system for real-time measurement of SO2 using ultraviolet Differential Optical Absorption Spectroscopy (DOAS). The system consists of a UV LED light source, a 30 cm gas measurement cell, a coupling optical fiber, and a spectrometer connected to a computer. A MATLAB algorithm was developed to acquire spectra in real time, compute absorbance, and estimate gas concentration based on the Lambert–Beer law. The sensor and processing algorithm were experimentally validated in three stages: preliminary liquid-phase tests, a functional test on a fluidized-bed reactor, and controlled tests using pressurized cylinders of SO2. Experimental results showed that the system successfully detected variations in SO2 concentration, evidenced by a progressive attenuation of the UV spectrum as gas concentration increased. However, optical saturation occurred at concentrations above 20 %, due to the excessive 30 cm optical path length, leading to total light absorption. Furthermore, it was confirmed that the molecular absorption cross-section of SO2 varies with wavelength, limiting the direct application of the theoretical Lambert–Beer model. To address these discrepancies, a nonlinear empirical interpolation model was iii Abstract implemented using calibration data, yielding a better agreement between measured and real concentrations. The results validate the functionality and sensitivity of the developed DOAS system for real-time detection of SO2. Although quantitative accuracy was not achieved across the full concentration range, the work establishes a solid foundation for future improvements in optical design and spectral processing, advancing toward a robust and low-cost sensor for industrial environmental monitoring applications.
Description
Tesis presentada para optar al título de Ingeniero/a Civil en Telecomunicaciones.
Keywords
Anhídrido sulfuroso, Reactores fluidizados, Análisis del impacto ambiental