Secrets Written in Dust
Research chemists at SwRI investigated dust for its ability to retain unique source attribution profiles
Dr. Kristin Favela, a senior research scientist in the Chemistry and Chemical Engineering Division, is an analytical chemist who leads projects and conducts research in areas of demilitarization, homeland security, forensics, environmental chemistry and metabolomics. She is an expert in analytical mass spectrometry and has developed and validated numerous methods now routinely applied.
Samples of dust were collected and extracted, dissolved in liquid and placed in vials, then loaded in a gas chromatography-mass spectrometer for analysis.
Statistical analysis resulted in grouped signatures of three kinds of chemical impurities associated with dust samples containing chemicals similar to those found in chemical warfare agents.
As hard as they might try to hide them, criminals or terrorists may soon find that there is more than one kind of fingerprint that can be left at a crime scene. Substances as common as house dust may contain traces of the unique chemical signature of explosives, nuclear materials, drugs or chemical warfare agents. Investigators can tie an individual to these as well as other items, such as household inks and copier toners, by analyzing their source attribution profile (SAP), which consists of the chemical characteristics of the substance itself, plus any traces characteristic of its origin or its processing.
This analysis relies on a variety of analytical detection techniques that include infrared spectroscopy, scanning electron microscopy, gas chromatography- mass spectrometry, liquid chromatography- mass spectrometry, nuclear magnetic resonance and inductively coupled plasma-mass spectrometry.
Generally, the amount of analytical information collected is too extensive to manually evaluate all of the possible variables and then separate them from unrelated background signals. Multivariate pattern recognition techniques, including principal components analysis (PCA), are useful for source attribution. These techniques can be supervised (looking for an expected pattern) or unsupervised (looking for patterns to emerge). An important feature of the PCA approach is to reduce the dimensionality of the attribution profile from many different variables to three or four independent quantities.
In a recent study, research chemists at Southwest Research Institute (SwRI) investigated dust for its ability to retain source attribution profiles after chemical exposure. The goal was to demonstrate that dust can yield meaningful information for the purpose of source attribution.
Results of the study were published in “Dust as a Collection Media for Contaminant Source Attribution,” in the April 2012 edition of Forensic Science International (Vol. 217, Issues 1-3, 39-49).
Three sources of acephate, a commercially available organophosphate insecticide used on vegetables and ornamental plants, were investigated as a proof-of-concept model and a substitute for phosphonate-containing chemical warfare agents (CWA). In addition, attribution profiles were created and tested using compounds related to CWAs. Dust was collected from a storage shed and from among particles deposited on carpet, then loaded with distinct chemical profiles using an exposure chamber and aerosolizer. After intervals of one hour, 24 hours or 72 hours, the dust was extracted and its SAP analyzed by gas chromatography-mass spectrometry (GC-MS) and/ or liquid chromatographytandem mass spectrometry (LC-MS/MS).
Dust: It’s everywhere
Dust has several advantages as a collector of SAP information. Dust is ubiquitous in the indoor environment, eliminating the need to have a specific collection device in place at the time of chemical exposure. Collecting dust is uncomplicated. In addition, dust is well documented as an efficient collector of the semivolatile and nonvolatile organic chemicals and metals present in the indoor environment. Prevalent concentrations suggest that house dust is the main source of the exposure of young children to allergens, lead and polybrominated diphenyl ethers (PBDEs). Dust is also a major in-home exposure source for pesticides, polyaromatic hydrocarbons, phthalates, alkylphenols and their ethoxylates, arsenic, cadmium, chromium, mold, endotoxin and bacteria.
Carpeting is a common dust reservoir and an efficient pesticide concentrator. Typically, pesticide concentrations in vacuumed house dust are 10 to 100 times higher than those found in outdoor surface soil.
There is also evidence that dust slows the degradation process of many chemicals as compared to exposure in the open environment. Residues from pesticides discontinued long ago in the U.S. are still found in house dust. Chlordane, banned in 1988, could still be detected in 38 percent of homes that were studied, and dichlorodiphenyltrichloroethane (DDT), discontinued in 1972, turned up in 70 percent of house dust samples collected from 1998 to 2001. Another example is the degradation over time of DDT to dichlorodiphenyldichloroethylene (DDE). The DDT-to-DDE ratio typically found in home dust samples is approximately 5 to 1, while the ratio in soil samples typically registers only about 1.5 to 1.
Materials and methods
An exposure chamber consisting of an 18-inch acrylic cube with an open bottom was custom-built for the project. A small fan was affixed to the inside roof of the chamber, and a fixed needle guide was installed through one of the sides of the chamber. The needle guide set the angle of the aerosolizing needle at approximately 40 degrees pointing upward toward the fan. A commercially available microsprayer was customized for the tests.
For exposure experiments approximately 4 milliliters (mL) of the diluted chemical were aerosolized and deposited on a carpet containing residual dust. A fan in the chamber served to direct the aerosol droplets toward the carpet and maintain circulation. The fan, aerosolizing needle, and inside of the exposure chamber were cleaned in between spray solutions.
Dust was collected from two storage sheds located in Pine Bluff, Ark., in quantities of approximately 100 grams from one shed and 300 grams from the other. Dust from the two sheds was blended at a 50:50 ratio for acephate experiments prior to deposition on the carpet. It was blended at a 15:85 (shed 1:shed 2) ratio and deposited onto a carpet purchased from a local hardware store.
SwRI researchers used principal components analysis (PCA) to determine the association of dust exposed to the same and different chemical sources. The results demonstrated that dust samples exposed to distinct chemical sources are clearly differentiated from one another across all collection times. Furthermore, dust samples exposed to the same source can be clearly associated with one another across all collection times. When the CWA-related compounds were subjected to elevated temperature, the signature remained stable at the 1-hour and 24-hour collections. At 72 hours and with elevated temperature, larger deviations were observed for some compounds compared to a control sample. Raising the alkalinity (pH 10) also affected the profile, but to a lesser degree than elevated temperature. Overall, dust was found to be an effective medium for the in situ collection of source attribution profiles.
Bill Williamson, a senior research scientist in the Chemistry and Chemical Enginering Division, prepares the gas chromatography/mass spectrometer instrument for a sample analysis.
Different impurity profiles (numbered) appear after chemical analysis of two different sources of the same chemical, in this case acephate, an organophosphate pesticide. Note that the relative abundances for the impurities differ, and some impurities, such as No. 3, appear in one sample and not the other.
Dust was chosen as a collector medium because it is known to absorb, concentrate and preserve toxic chemicals such as pesticides and polyaromatic hydrocarbons. Results for the systems investigated in this study (acephate and CWA-related compounds) confirm that the fidelity of the SAP is maintained to the extent that, in most cases, PCA correctly grouped the exposed dust according to source.
Extraction methodology was not studied, but developing a universal dust-extraction method that is well-suited for capturing compounds with a range of polarities would improve confidence associated with matching dust extracts to their corresponding source chemicals. If the source chemical is available, extraction efficiency can be measured empirically and the dust extract data corrected accordingly.
The SwRI team demonstrated that the SAP is tolerant to some degree of variability in the analytical data, depending on the extent of underlying differentiation in the sources. In the systems used for the study, most of the variability among the different sources is attributable to three or four individual compounds, and peaks with large relative abundances will influence the PCA results more than those with smaller relative abundances.
A modestly variable analytical response of a compound that has a relatively low abundance, or does not vary significantly among different source chemicals, affects the PCA results only marginally. In a field situation, collecting replicate samples at different time points can be useful to determine which compounds demonstrate a high level of variability as a function of time. Compounds that are determined to be inherently unstable are ultimately not as meaningful and can be excluded from the PCA.
This process of feature selection is important not only in separating the meaningful profiles from those that are variable, but also from those that can be attributed to the background or to other unrelated sources. In practice, background signals can be identified by analyzing a source of dust that has been protected from chemical exposure. Pre-exposure dust, such as that inside a vacuum cleaner found at the scene, is an example of a protected background source.
The SwRI team demonstrated that dust is a promising medium for collecting source-attribution profiles, but some variables still need to be studied. Dust composition and properties can vary widely depending on geographical location and the immediate environment, and the effects of these differences should be studied. Humidity levels also could contribute to the degradation of some chemicals. Furthermore, particle size may play a role in the differential concentration of chemicals. Previous studies have shown that the concentrations of pesticides and polyaromatic hydrocarbons in house dust are much higher among the small inhalable and respirable particles than on larger particles. Despite these remaining areas of study, many variables likely can be mitigated through proper sample collection, storage, preparation and data processing.
The end result is intended to enhance forensic investigations such that samples taken at the scene of a crime or terrorist incident may be collected and analyzed in the laboratory to identify certain unique attributes of the chemicals used to create the incident. Those attributes can then be compared to the chemical signatures of samples of dust collected elsewhere, such as from the home or office of a suspect in the incident. The discovery of significant similarities between the two could be the key to solving what otherwise might be a very difficult criminal investigation.
Questions about this article? Contact Favela at (210) 522-4209 or firstname.lastname@example.org.
The author gratefully acknowledges the contributions to this project of Dr. Jonathan A. Bohmann, a senior research scientist in the Microencapsulation and Nanomaterials Department and William S. Williamson Jr., a senior research scientist in the Environmental and Demilitarization Technology Department, both within the Chemistry and Chemical Engineering Division at Southwest Research Institute. Funding for this project was provided by the Science and Technology Directorate, U.S. Department of Homeland Security.