2.3 Data acquisition and QCL system
Before measuring the QCL spectra, a background spectrum of a roughened gold substrate was obtained. This background spectrum provided a good and smooth reference trace, due to the lack of MIR signals from the employed solid matrix. The samples were placed in the wells of metal holders (2.54 cm in diam., 10 mm deep). The sample surface was flattened to create a smooth surface for accurate measurements. Duplicate spectra were collected at ten (10) different locations on the surface, resulting in a total of twenty (20) spectra per sample. These were used for the calibration and test analyses. This process was used for the neat samples, and each concentration in the test mixes samples. The spectra were obtained in reflectance (R) mode at a distance of approximately 15 cm, using a LaserScan™ MIR pre-dispersive spectrometer (Block Engineering, Marlborough, MA, USA) equipped with three tunable MIR lasers with a tuning range from 990 to 1111, 1111 to 1178 and 1178 to 1600 cm-1, the scan time was 0.5 s approximately on each laser for a total time scan of 1.5 s. The average power typically varied between 0.5 to 10 mW across the 600 cm-1 total tuning range with 100:1 transverse electromagnetic polarization (TEMoo) and beam divergence of < 2.5 mrad in the x-axis and < 5 mrad in the y-axis. The spectrometer had a ZnSe lens of 3 in. in diameter, which was used to focus the MIR beam, to collect the reflected light, and to focus the light onto a thermoelectrically cooled mercury-cadmium-telluride (MCT) detector. The wavelength accuracy and precision were 0.5 cm-1 and 0.2 cm-1, respectively. The spectroscopic system worked best at a laser head to target distance of 15 ± 3 cm. Each laser diode formed an elliptical spot of 4 x 2 mm2 at the focal plane of the ZnSe lens due to the difference of beam divergence in the axes.