Diffuse reflectance infrared fourier transform spectroscopy (DRIFTS)[1] is an infrared spectroscopy sampling technique used on powder samples without prior preparation. The sample is added to a sample cup and the data is collected on the bulk sample. The infrared light on a sample is reflected and transmitted at different amounts depending on the bulk properties of the material. Diffuse reflection of the incident light produced by the sample's rough surface reflection in all directions is collected by use of an ellipsoid or paraboloid mirror. Shape, compactness, refractive index, reflectivity and absorption of the particles are all characteristic of the material being analyzed. If the sample is too absorbent, then it can be diluted with a nonabsorbent material such as potassium bromide, potassium chloride, etc. The particle size should be smaller than the wavelength of the incident light in order to minimize Mie scattering, so this would infer that it should be less than 5 µm for mid-infrared spectroscopy. The spectra are plotted in units of log inverse reflectance (log 1/R) versus wavenumber. Alternative plots of Kubelka-Munk units can be used, which relate reflectance to concentration using a scaling factor. A reflectance standard is needed in order to quantify the reflectance of the sample because it cannot be determined directly.[2][3]

Diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) was used to analyze four types of forensically relevant fabrics coated with varying dilutions of blood. The blood was applied in two manners, dip coating with a smooth and uniform layer and drip coating with droplets from pipettes. Spectra of neat and dip coated fabrics were acquired using controlled orientations, and these were compared to spectra collected on samples with random orientations. The improved reproducibility seen in visual inspection of the spectra is confirmed by principal component and linear discriminant projections of the spectra, as well as by statistical hypothesis testing. Principal component regression (PCR), using the regions of the IR spectra associated with the amide A/B, I, II, and III vibrational bands (3500-2800, 1650, 1540, and 1350 cm-1), was employed on the more uniform dip coated spectra to estimate limits of detection for blood on two of the four fabrics - acrylic and nylon. These results demonstrate that detection limits for blood on fabrics can be decreased significantly by controlling for the orientation and face of the fabric samples while collecting spectra. Limits of detection for acrylic and nylon were found to be 196 Ã and 227 Ã diluted blood, respectively.




Diffuse Reflectance Infrared Fourier Transform Spectroscopy Drift