This thesis deals with the addition of upconversion detection to non-linear mid-infrared laser spectroscopy techniques. The techniques investigated here are degenerate four wave mixing (DFWM) and infrared polarization spectroscopy (IRPS). These techniques are useful for detecting many of the key species in combustion, such as CH4, C2H6, HCl, and HCN. These species can be probed using ro-vibrational transitions in the mid-infrared, but lack transition that would make detection possible in the visible or near-infrared. These techniques allow spatially resolved, non intrusive and sensitive detection of these important species directly in flames.

The main focus has been on investigating the new opportunities that upconversion detection makes possible. It had already been demonstrated that upconversion greatly improved the sensitivity of DFWM. As a part of this thesis work, it was shown that signal to noise ratio of IRPS was also improved by upconversion, although not to the same degree.

DFWM with upconversion was used to quantitatively measure HCN released during the devolatilization phase of burning straw pellets, which was only possible due to the improved sensitivity of the detection system. A major part of quantifying concentration measurements is achieving accurate temperature measurements, as the signal scales with temperature. A method of measuring temperature using the relative strength of transition lines of water had been described before this thesis. The effects of signal saturation for waterline thermometry were investigated as part of this thesis to increase the precision
and reliability of temperature measurements, as this would aid in achieving more accurate concentration measurements.