Long-Wave Infrared Supercontinuum Source and Sensor for Standoff Sensing and Trace Particle Identification

Abstract

This dissertation is based on the development of a long-wave infrared supercontinuum source, and the utilization of this source in a Fourier transform based standoff optical sensor. The spectra from trace particles deposited on smooth surfaces are measured with the sensor and simulated using a Bobbert-Vlieger model. We create an ultra-broadband, all-fiber supercontinuum source that emits infrared energy from 1.6 - 11μm. We utilize a master oscillator parametric amplifier with Erbium/Ytterbium and Thulium amplifiers to pump a cascade of ZrF4-BaF2-LaF3-AlF3-NaF (ZBLAN), arsenic sulfide, and arsenic selenide fibers. This source is power scalable and can emit up to 417 mW at 800 kHz pulse repetition frequency, with 69 mW beyond 7.5μm. The output from the long-wave infrared supercontinuum source is near-diffraction limited single mode output that can be collimated to a one inch spot.The output of the source is then tested for feasability of use in commercial FTIR based systems. Although not optimized for 1.5 ns pulsed sources, we are able to measure transmission spectra of polystyrene samples, thin films on wafers and 50μL of acetone that has been evaporated in a 10 cm length gas cell and compare to those illuminated with the systems internal globar. We find that even though the input optics are not optimized, the incident energy on the samples is an order of magnitude higher than that of the globar source. We develop a long-wave infrared standoff sensor by coupling the output of the source to a refraction based FTIR interferometer. The modulated energy is then guided to hit targets that are 3.6 m away from the sensor. The system estimates the energy of each pulse and creates an interferogram that is Fourier transformed into resultant spectra. The linearity of the sensor is verified via the measurement of thin films of SiO2 and polyimide on silicon wafers. This sensor is then used for standoff volatile gas and bulk sample scattering measurements. We then focus on the measurement and modeling of trace chemicals that have deposited on smooth substrates. We measure concentrations as low as 6.5 μg/cm2 on glass substrates. Furthermore, we measure the diffuse reflectance of RDX, acetaminophen, and caffeine on glass, aluminum, and silicon substrates. Each of these chemicals exhibit spectral features between 950 and 1800 cm−1 and substrate based dependencies in reflectance spectra. We simulate these effects with a Bobbert-Vlieger model that takes particle size distribution into account. We find that a range of particle sizes smoothens and broadens reflectance features and changes in target orientation and differences in particle shape can strongly impact the spectra between 1800 and 4000 wavenumbers. We use our Bobbert-Vlieger model to create a library of exemplary spectra based on systematically changing the parameters of the particle size distribution. This library is employed to identify unknown powders based on the root mean square error between the second derivative of measured spectra and those in the library.