Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 200
Countering Terrorism: Biological Agents, Transportation Networks, and Energy Systems - Summary of a U.S.-Russian Workshop 20 Raman Spectroscopic Detection of Chemical, Biological, and Explosive Agents1 Russ Zajtchuk, M.D., Professor Emeritus, Rush University Medical Center, and Gary R. Gilbert, Ph.D., Georgetown University Imaging Science and Information Systems Center, Temporarily Assigned to U.S. Army Medical Research and Materiel Command (USAMRMC) Telemedicine and Advanced Technology Research Center (TATRC) INTRODUCTION The real-time detection and identification of biological and chemical warfare agents as well as potential toxic industrial gases and improvised explosive devices (IEDs) is of paramount importance for protecting soldiers and first responders on the battlefield and in counterterrorism response at home. This paper deals with the development of a chemical, biological, and explosive (CBE) detection system based on Raman spectroscopic measurements integrated to a commercially available unmanned ground vehicle (UGV) platform. Raman detection offers clear advantages over immunoassay- and DNA-based biological detection strategies, especially when configured for use on an unmanned vehicle. Raman measurements are reagentless, greatly simplifying the logistics of deployment. In addition, Raman measurements can be used to detect a broad range of CBE threats in a single measurement cycle. By remotely guiding this sensor system to an incident area to assess soil, water, and surface contamination, exposure of personnel to a hazardous environment is prevented until the nature of the threat is fully known. Bringing the sensor to the sample also minimizes problems associated with sampling, such as cross-contamination and preanalysis decontamination, as well as the problems of disposal after the analysis is complete.
OCR for page 201
Countering Terrorism: Biological Agents, Transportation Networks, and Energy Systems - Summary of a U.S.-Russian Workshop RAMAN SPECTROSCOPY FOR CBE DETECTION Raman spectroscopy has been studied and used as a laboratory tool in chemistry for many years. It is now reaching a level of maturity that is transitioning from the laboratory to a variety of field applications. The Raman effect occurs when a photon encounters a molecule, during which time there is a chance that the energy from the scattered photon will be exchanged with vibrational bond energy of the molecule. This energy exchange manifests itself as a shift in frequency (or wavelength) in a small amount of the scattered light. Because each different chemical bond in a material causes a different frequency shift, the pattern of these shifts, known as the Raman spectrum, is unique to that material. The Raman spectrum reveals the molecular composition of materials, including the specific functional groups present in organic and inorganic molecules. The Raman spectrum is a characteristic property of a material, just like its color or melting point, and can be used to determine the presence or absence of the material. The detector will always be measuring an agent spectrum in the presence of the spectrum from the background or from any other material that may be present. Fortunately, in most real-world situations, the ratio of the amount of agent to the background and any other materials has significant spatial variation. This variation in composition leads to slightly different Raman spectra from different areas on the sample surface. These differences in spectra provide enough information for chemometric processing of the data, allowing identification of the agent and the background materials. The Raman Bio Identification (RBI) system computer receives a command initiated by the operator to acquire and analyze a sample from the UGV central processing unit (CPU). Using software previously developed by the ChemImage Corporation of Pittsburgh, Pennsylvania, up to 19 spectra are acquired from the sample. The laser power is typically 12 milliwatts, resulting in a laser power density of 86 watts per square centimeter. The exposure time used to acquire the spectra in testing was 10 seconds, and each measurement is the product of 10 averages (see Figures 20-1 and 20-2). RAMAN BIO IDENTIFICATION (RBI) DETECTOR The overall concept of the RBI robot demonstration system (Wolverine) was to integrate an RBI point sensor (the RBI head) onto a UGV manipulator arm, and then couple it to an instrument package mounted on the main chassis of the UGV. The coupling of the point sensor is accomplished through both electrical and fiber optic cables running along the manipulator structure. The RBI detector is a Raman point sensor or a Raman proximity detector. To operate, it needs to be close but not necessarily touching the surface to be measured. The RBI detector contains subsystems to allow targeting of the head (video camera and fine-positioning system), laser illumination of the sample to induce
OCR for page 202
Countering Terrorism: Biological Agents, Transportation Networks, and Energy Systems - Summary of a U.S.-Russian Workshop FIGURE 20-1 Raman spectra of several chemical warfare agents. SOURCE: Gardner et al., 2007. the Raman effect, optics to collect and focus scattered light, a fiber optic bundle to transport the scattered light to a spectral analyzer (spectrometer subsystem), and a system computer to provide control and communication (see Figure 20-3). Using a single laser illumination spot and a single spectrometer, the RBI detector can produce up to 19 spatially resolved spectra from a sample region of interest. These spatially resolved spectra can be processed using a mixture analysis algorithm coupled with library searching to provide robust identification of threat and nonthreat materials present in complex environmental samples. UGV INTEGRATION The Wolverine is controlled using a radio frequency link between the robot and the operator control unit (OCU). A payload interface allows control and data transmission to and from the RBI system through this wireless interface. The
OCR for page 203
Countering Terrorism: Biological Agents, Transportation Networks, and Energy Systems - Summary of a U.S.-Russian Workshop FIGURE 20-2 Raman spectra of selected biothreat agents. SOURCE: Gardner et al., 2007. operator has access to this video stream and can use it for fine control of the RBI head. The proof of concept consisted of placing a biological toxin simulant, ovalbumin, on a flattened sheet of galvanized iron air-duct material to provide a constant background for the measurement. The operator then moved the UGV close to the sample area and used the manipulator to position the RBI detector head directly over the sample. The fine-adjustment system in the RBI head was used to set the collection lens of the detector at the proper distance from the sample through commands from the OCU. Once the detector was positioned, the operator started the analysis. The analysis consisted of a 3-minute wait period, during which the native fluorescence of the ovalbumin was quenched by the laser excitation. Next, a 1-minute acquisition of the 19 spatially resolved Raman spectra was taken. This set of spectra was preprocessed and analyzed. To confirm proper spectral performance, the spectra were averaged, prepro
OCR for page 204
Countering Terrorism: Biological Agents, Transportation Networks, and Energy Systems - Summary of a U.S.-Russian Workshop FIGURE 20-3 RBI detector mounted on the UGV. SOURCE: Gardner et al., 2007. cessed, and compared against the library spectrum for ovalbumin. There was good agreement between the RBI and library spectra, which confirmed the accuracy of the RBI detector. Once the analysis and reporting of results was complete, the operator had the option of taking another measurement or moving the detector to another sampling location. LASER-INDUCED BREAKDOWN SPECTROSCOPY (LIBS) COMBINED WITH RAMAN SPECTROSCOPY Laser-induced breakdown spectroscopy (LIBS) is a detection method that can be used to identify chemical and biological hazards in bulk and on surfaces. It is relatively straightforward, requires no sample preparation or consumables, is sensitive, uses only a small sample substrate, is fast (subsecond), is field portable, and can be miniaturized.2 A laser is aimed at a target and is used to quickly heat
OCR for page 205
Countering Terrorism: Biological Agents, Transportation Networks, and Energy Systems - Summary of a U.S.-Russian Workshop the target into a plasma plume. The resulting atomic emissions from the plasma are read by a broadband spectrometer. By measuring the relative intensities of emission spectra peaks and their specific pattern, LIBS can monitor all chemical elements present in a sample, at the same time, with a single laser shot. LIBS has been used to identify substances both in their solid and liquid forms and on top of soil samples. There is significant improvement in detection of chemical, biological, and explosive materials when fusing data from two orthogonal technologies, LIBS and Raman, resulting in a dramatic decrease in false positive rates. The joining of LIBS and Raman into a single sensor unit makes great sense, as both techniques can use the same laser system and same spectrometer. Also, the advanced chemometrics for spectral data analyses can be shared. LIBS and Raman are true orthogonal technologies because LIBS keeps track of the elemental composition of the sample, or target (that is, establishes its stoichiometry) exceptionally well, while Raman provides unique molecular signature information, both of which facilitate the determination of whether the target material is hazardous. Moreover, LIBS and Raman are “universal” sensors that can be applied to a very wide range of materials analyses, both hazardous and benign. The ultimate launching of a LIBS–Raman sensor payload on a robotics platform will lead to unprecedented capabilities for field applications in both proximity and standoff sensing modes. PHOTON SYSTEMS DEEP UV RAMAN AND FLUORESCENCE DETECTOR PROJECT Photon Systems, Inc., and the NASA Jet Propulsion Laboratory are collaborating on a project to develop an advanced, miniature, low-power, reagentless, robot-mounted, laser-based instrument for real-time detection and classification of trace concentrations of biological and chemical agents on surfaces. A combined sensor employing deep ultraviolet (UV) laser-induced native fluorescence (UVLINF) with deep UV resonance Raman spectroscopy (UVRRS) was selected for this project. This instrument is a deep UV laser consuming less than 5 watts of battery power that simultaneously generates Raman scattering and excites native fluorophores contained within microorganisms and many organic and inorganic materials. Using an onboard real-time algorithm, the UVLINF and UVRRS data are processed to identify and classify contaminants in less than 1 second. Simultaneous multiband fluorescence and Raman sensor outputs are processed using neural net algorithms to classify contaminant organic and inorganic materials. During Phase I of this effort, Photon Systems successfully designed, fabricated, and demonstrated a capability of solar-blind, standoff detection and classification of trace amounts of biological and chemical contaminants and explosives on surfaces at working distances of 1-3 meters, significantly greater than the proposed instrument standoff goal of 5-30 centimeters. This optical instrument
OCR for page 206
Countering Terrorism: Biological Agents, Transportation Networks, and Energy Systems - Summary of a U.S.-Russian Workshop is not affected by ambient lighting, because of a combination of its operation in the deep ultraviolet range and the use of pulse-gated detection for background reduction. This is a very important feature for use of the instrument under natural or artificial lighting. The entire instrument was integrated into a single, robot-arm-mounted package with a weight of 5 pounds and power consumption of 5 watts, rather than the original two-part instrument with one-half on the robot arm and one-half in the robot body. The instrument includes onboard microprocessors and firmware for controlling the laser as well as each detector and for performing a variety of computational and self-calibration tasks. The overall data processing using chemometric software for detecting and classifying unknown surface contaminants was performed by remote computer via a wireless link. CONCLUSION A mathematical model was constructed to describe the performance of the detector system. This mode was evaluated using a simulant for anthrax, B. thuringiensis (Bt) spores, and a biological toxin simulant, ovalbumin. This modeling confirmed the feasibility of the design for biothreat agent detection. While Raman identification of agents requires a spectra library for patternmatching recognition of molecular structure, it can still identify new, unknown agents (such as recombinant chemical agents or explosive structures or genetically altered organisms) by flagging for further investigation the spectra not already present in the library, especially if they closely resemble the spectra of a known agent or class of known agents. Warning the user of the presence of an unknown substance that could possibly be a threat is of great value to the war fighter or emergency responder. Such warnings could then be incorporated into standard operating procedures for donning mission-oriented protective posture gear or other personal protective equipment. Ultimately, the suspect spectra would be added to the library as an unknown spectra associated with a potential hazard. Work is continuing to refine the RBI detector hardware and software to allow integration on a wider class of UGV platforms and to optimize system operation for field use. ACKNOWLEDGMENTS This work was supported by the U.S. Army Medical Research and Materiel Command under Contract No. W81XWH-06-C-0010.
OCR for page 207
Countering Terrorism: Biological Agents, Transportation Networks, and Energy Systems - Summary of a U.S.-Russian Workshop DISCLAIMER The views, opinions, and findings contained in this paper are those of the authors and should not be construed as an official Department of the Army position. NOTES 1. Gardner, C. W., et al. 2007. Demonstration of a robot-based Raman spectroscopic detector for the identification of CBE threat agents. Manuscript submitted for the Twenty-Fifth Army Science Conference, sponsored by the Assistant Secretary of the Army for Acquisition, Logistics and Technology, November 27-30, 2007, Orlando, Florida. Defense Technical Information Center Report AD-A481010, available online at hdl.handle.net/100.2/ADA481010. 2. Batavia, P., and R. Watts. 2004. Collaborative robots design considerations. Presentation at the National Defense Industrial Association Fourth Annual Intelligent Vehicles Systems Symposium , Traverse City, Michigan, June 2004.