Analyzing Forensic Trace Evidence with IR PiFM

Background

Trace Evidence Analysis is the discipline of forensic science that deals with the minute transfers of materials in a crime scene that cannot be seen with the unaided eye. Trace evidence may provide a link between the victim and a suspect, a victim and a scene, or the suspect and a scene. Some common trace evidence includes fibers, hairs, and other miscellaneous items such as residues of cosmetics.

Given relatively large size of these trace elements, techniques that are coupled to an optical microscope such as Raman and infrared (IR) microscopes should be able to locate single trace element such as fiber. However, due to the probing volume of these techniques, the vibrational spectra acquired will be dominated by the signal from the fiber itself and little from the ultra-thin coatings that may hold the useful forensic information. In this application note, IR PiFM’s surface sensitivity (probing depth of ~20 nm) will be used analyze a protective coating applied to individual fiber samples.

Analysis

Figure 1 shows six PiF-IR spectra acquired on a single cotton knit fiber (A2) that is glued onto a glass substrate by a thin layer of adhesive; the bottom inset figure shows the optical image seen in the instrument with the red square denoting the general area from where the PiFM measurements are taken. The upper inset figure shows the AFM topography along with six locations (300 nm apart from each other) where the PiF-IR spectra were acquired at. Since the fiber is uncoated, it should be quite homogeneous even in the nanoscale, and the similar spectra reflect such homogeneity. The red FTIR spectrum for cotton (gleaned from the web) shares many of the peaks observed in the PiF-IR spectra. All the other fibers analyzed and discussed below were prepared in a similar manner and six spectra were acquired with 300 nm spacing between each other.

Figure 2 shows six PiF-IR spectra acquired on a single cotton knit fiber that is coated with a 3M protective coating (A1). Unlike the uncoated cotton knit fiber (figure 1), the spectra vary more especially in the relative strength of peaks due to the uneven coating thickness. In the inset, two spectra are shown for more clarity and to highlight the effect of uneven coating thickness.

To better understand the spectra in Figure 2, we compare the average spectra of the uncoated and coated fibers with the FTIR spectrum in Figure 3. We see that the coated fiber has the strongest peak at 1237 cm⁻¹ whereas the bare cotton has the strongest peaks at 1030 and 1056 cm⁻¹. Looking at the two spectra in the inset of Figure 2, we can see that the orange and blue spectra have opposing strengths at these wavenumbers, that is, the orange (blue) spectrum has higher (lower) 1237 cm⁻¹ peak strength and lower (higher) 1030 and 1056 cm⁻¹ peak strengths. That is due to the fact that PiFM is a surface sensitive technique whose signal strength is quickly attenuated with any intervening layer however thin. Therefore, the location from where the orange spectrum was acquired is covered by a thin layer of 3M protective coating whereas the location from where the blue spectrum was acquired is most likely exposed (in such a case, the IR signature of the protective coating seen in the blue spectrum is likely from the neighboring regions since PiFM collects signals from about 10 nm region). Alternatively, it also could be that there is a thinner layer of 3M protective coating at that location. In figure 3, the green dotted lines identify the PiF-IR peaks associated with the coating while the black dotted line identify the peaks associated with the cotton.

Figure 4 compares a spectrum from a bare cotton woven fiber (B2) with two spectra from a cotton woven fiber that is coated with the 3M protection coating (B1); a FTIR spectrum of silicone oil is shown along with the PiF-IR spectra to show that these fibers have been also exposed to silicone oil treatment. Even with the strong contribution from the silicone oil peaks, we can make out the 1237 cm⁻¹ peak associated with the 3M protection coating; we also note that the relative strength of the peak around 1030 cm⁻¹ compared to the other silicone oil peaks (such as 1260 and 1100 cm⁻¹) is much higher for PiF-IR spectra compared to the FTIR due to the contribution from the cotton. In a manner like the uneven coating on A1 (figure 2), the two spectra (purple and yellow) on B1 in figure 4 allow us to decipher the coating condition. First, the silicone peaks at 820 and 1020 cm⁻¹ are similar in strength for both the purple and yellow spectra, suggesting a similar silicone oil coverage at both locations. However, the purple spectrum shows a stronger peak at 1048 cm⁻¹ (associated with cotton) but a weaker peak at 1240 cm⁻¹ (associated with 3M coating, 1237 cm⁻¹ from figure 2) compared to the yellow spectrum, suggesting that the location associated with the purple spectrum has lesser amount of 3M coating. We can map the uniformity of the 3M coating by utilizing the nanoscale chemical mapping capability of PiFM. Figure 5 shows PiFM images of the B1 fiber at 1048 cm⁻¹ (for visualizing cotton) and 1241 cm⁻¹ (for visualizing 3M coating) colored green and red, respectively along with the topography and the spectra from two locations, each from a location that shows stronger signal strength. It also includes a combined PiFM image where the green color represents cotton molecules and red color represents the 3M coating molecules. We can see that the coating is not covering the cotton surface uniformly.

Conclusion

In summary, IR PiFM can chemically analyze ultra-thin protective coatings on and the chemical make-up of individual fiber fragments to provide a valuable capability to forensic community to utilize individual fiber fragments as trace evidence.