White powder identification using broadband coherent light in the molecular fingerprint region

We show that a variety of white powder samples, including painkillers, amino acids, stimulants and sugars are readily discriminated by diffuse reflectance infrared spectroscopy involving no preparation of the sample and no physical contact with it. Eleven powders were investigated by illuminating each sample with broadband coherent light in the 8-9-µm band from an OPGaP femtosecond optical parametric oscillator. The spectra of the scattered light were obtained using Fourier-transform spectroscopy. Similarities between different spectra were quantified using Pearson's correlation coefficient, confirming that spectral features in the 8-9-µm wavelength region were sufficient to discriminate between all eleven powders evaluated in the study, offering a route to simple and automated non-contact chemical detection.