|Steven Buckley also uses lasers to study air samples. By combining rapid-fire
laser bursts with laser-induced breakdown spectroscopy, Buckley can
discriminate particles of dust from similar-sized anthrax spores.
Eye surgeons use them to sculpt corneas. Microchip makers use them to etch tiny lines on silicon wafers. MAE professor Steven Buckley would like to use a laser intended for telecommunications to perform continuous real-time analysis of gas molecules in inhospitable industrial processes.
"Our main goal is to develop practical diagnostics to understand combustion physics, emissions or how to optimize industrial processes," Buckley says. "This type of tunable laser can be used to continuously interrogate gas samples in any industrial process without perturbing the physics of the process."
Tunable diode lasers emit a narrow-wavelength band, or color, of light. In telecommunications, rapid modulation of the laser's electrical current— at speeds up to hundreds of millions of times per second—allows the light to carry voice conversations over fiber-optic lines. At a slower modulation rate, Buckley can use the light to monitor gas mixtures.
"We could aim our laser into a steel-making blast furnace and determine the temperature and concentration of carbon dioxide, carbon monoxide, and other gases, and use that data in feed-forward control or as input to computer models to improve process performance," Buckley says. "We could change carbon-steel manufacturing, glass making, and other thermal processes from an art to a science."
A key part of the technology involves the measurement of light absorption by gas species. Buckley's team has developed a mathematical technique that determines the instantaneous rate of change in absorption at predetermined wavelengths. Higher accuracy than traditional techniques, with almost no interference, is achieved when "wavelength modulation spectroscopy" is employed.
"We're designing these devices to be practical," Buckley says. "And although they will include extremely sophisticated technology and algorithms, you won't need a Ph.D. to interpret the results."