Mineral dust particles are ubiquitous in the atmosphere and can be transported vast distances affecting climate, air quality, and human health on a global scale. Mineralogical composition has a substantial impact on dust properties and their effects. Natural dust samples are both fine-grained and composed of many different minerals. Most commonly, X-ray diffraction (XRD) has been used to characterize dust mineralogy; however, this technique is less effective for identifying poorly crystalline or amorphous phases. We used Fourier Transform Infrared (FTIR) spectroscopy as a complementary method to identify minerals and their abundances. Long wave infrared (LWIR) spectra (2.5 to 25 μm) are sensitive to molecular bonds rather than crystallography providing additional details. We performed both XRD and reflectance spectroscopy to characterize 37 atmospheric dust samples collected in Ilam City, Iran. The dominant minerals in these samples are quartz, feldspar (albite), calcite and clays (illite, montmorillonite, kaolinite). LWIR reflectance is strongly dependent on particle size but published data of pure silicate minerals in the size range of the Ilam samples (0-63 μm) still show characteristic signatures between 8 and 10 µm (Salisbury et al. 1991; Wenrich and Christensen, 1996). Surprisingly, diagnostic silicate features were not observed in any of the samples although carbonate and OH bonds in the clay minerals were readily identified. Past studies have shown that porosity, grain size and packing can reduce the spectral contrast in the LWIR and additional effects include grain coatings or the interaction of multiple minerals. We also identify a peak at 7.8 µm which may be attributed to anomalous dispersion or the interaction of quartz and calcite in this spectral range. In order to understand the absence of Si-O features we made transmission measurements of representative samples in KBr pellets. Transmission is not influenced by multiple scattering and should clearly detect fundamental Si-O absorptions. Transmission spectra show broad features that include contributions from all silicate minerals (quartz, feldspar and clays) both near 10 μm and at longer wavelengths. We are using various spectral modeling techniques and will compare abundances derived from reflectance and transmission measurements.