<%BANNER%>

Characterization of Commonly Encountered Explosives Using High-Field Asymmetric Waveform Ion Mobility Spectrometry Coupl...


PAGE 1

CHARACTERIZATION OF COMMONLY ENCOUNTERED EXPLOSIVES USING HIGH-FIELD ASYMMETRIC WAVEFORM ION MOBILITY SPECTROMETRY COUPLED WITH MASS SPECTROMETRY BY JARED J. BOOCK A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007 1

PAGE 2

2007 Jared J. Boock 2

PAGE 3

To my family and friends. 3

PAGE 4

ACKNOWLEDGMENTS I wish to thank my research director, Dr. Ri chard A. Yost, for teaching me a great deal about the fundamentals of mass spectrometry an d providing guidance throughout my research. I would also like to thank Dr. Ma tthew Pollard at the Dr. Herbert H. Hill laboratory at Washington State University for providing two opportunities for me to conduct research on their instruments, and for providing invaluable assistance and conf erring with me on the science. I thank Ion Metrics, Inc., of San Diego, CA for funding these trips. I wo uld like to recognize the members of my committee: Dr. David H. Powell and Dr. John R. Eyler. I would like to thank the US Air Force Institute of Technology Civi lian Institutions program fo r giving me this opportunity. Thanks also go to my colleagues in the Yost group for their support. Lastly, I would like to thank my parents and my friends, without whom I would have gone insane several times during this year and a half. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES ...........................................................................................................................7 LIST OF FIGURES .........................................................................................................................8 LIST OF ABBREVIATIONS ........................................................................................................11 ABSTRACT ...................................................................................................................................12 CHAPTER 1 INTRODUCTION AND INSTRUMENTATION.................................................................13 1.1 Introduction......................................................................................................................13 1.2 Atmospheric Pressure Chemical Ionization .....................................................................15 1.3 High-field Asymmetric-Waveform Ion Mobility Spectrometry ......................................16 1.4 Ion Optics .........................................................................................................................21 1.5 Quadrupole Ion Trap Mass Spectrometry ........................................................................22 1.6 Multidimensional Mass Spectrometry .............................................................................26 1.7 Thesis Overview........................................................................................................... ...26 2 PROPERTIES AND CHARACTE RIZATION OF EXPLOSIVES.......................................27 2.1 Explosives that were Characterized .................................................................................27 2.2 Classification of Explosives ............................................................................................27 2.3 Chemical Properties of Explosives..................................................................................28 2.4 Instrument Settings ..........................................................................................................29 2.5 Compensation Voltage (CV) Scans .................................................................................30 2.6 Characterized Compounds ................................................................................................31 2.6.1 Nitroaromatics.......................................................................................................31 2.6.1.1 (2,4,6 ) Trinitrotoluene (TNT)..................................................................31 2.6.2 Nitramines.............................................................................................................43 2.6.2.1 Cyclotrimethylene trinitramine (RDX) ......................................................43 2.6.2.2 Cyclotetramethylene tetranitramine (HMX)...............................................46 2.6.3 Nitrate Esters.........................................................................................................5 0 2.6.3.1 Nitroglycerin (NG) .....................................................................................50 2.6.3.2 Pentaerythritol te tranitrate (PETN).............................................................51 2.6.3.3 Ammonium Nitrate Fuel Oil (ANFO).........................................................53 3 FAIMS/MS AND IMS/FAIMS/MS OF EXPLOSIVES........................................................57 3.1 Instrumentation ................................................................................................................57 3.1.1 Electrospray Ionization (ESI) Source....................................................................57 5

PAGE 6

3.1.2 Ion Mobility Spectrometer (IMS)..........................................................................59 3.1.3 Orthogonal Dome FAIMS Cell.............................................................................59 3.2 FAIMS Limitations ..........................................................................................................62 3.3 Effects of Addition of Helium to the Carrier Gas ............................................................63 3.4 High-resolution FAIMS (HRFAIMS) .............................................................................63 3.4.1 Separation of Mixtures of Explosives...................................................................64 3.4.2 Resolving Isomers.................................................................................................68 3.5 Dispersion Voltage ..........................................................................................................69 3.6 Effects of Adjustment of Plate Gap .................................................................................70 3.7 Relationship of DV to CV...............................................................................................80 3.8 Sensitivity ........................................................................................................................81 3.9 IMS/FAIMS of Explosives ..............................................................................................81 4 CONCLUSIONS AND FUTURE WORK.............................................................................83 4.1 Conclusions ......................................................................................................................83 4.1.1 Characterization of Explosive Compounds...........................................................84 4.1.2 FAIMS Characteristics..........................................................................................85 4.1.3 IMS/FAIMS/MS....................................................................................................85 4.1.4 High-resolution FAIMS.........................................................................................86 4.1.5 Mixture Separations with FAIMS/MS..................................................................86 4.1.6 Disadvantages of FAIMS......................................................................................86 4.2 Recommendations ............................................................................................................87 4.2.1 Further HRFAIMS Improvements........................................................................87 4.2.2 FAIMS Carrier Gas Experiments..........................................................................87 4.2.3 Practical Separations and Analysis........................................................................88 4.2.4 IMS Gating............................................................................................................88 4.2.5 CV Scanning..........................................................................................................88 4.2.6 Instrument Maintenance........................................................................................89 4.3 Future Work .....................................................................................................................89 4.3.1 Ionization Sources.................................................................................................89 4.3.2 Development of Field Instruments........................................................................91 REFERENCES ..............................................................................................................................92 BIOGRAPHICAL SKETCH .........................................................................................................96 6

PAGE 7

LIST OF TABLES Table page 2-1 Explosives to be Characterized. .........................................................................................27 3-1 Relative Ion Intensities. .....................................................................................................79 3-2 Relative CV Values. ...........................................................................................................79 7

PAGE 8

LIST OF FIGURES Figure page 1-1 Finnigan LCQ APCI source. ..............................................................................................15 1-2 FAIMS waveform and ion mobility behavior. ...................................................................17 1-3 APCI source and FAIMS cell mounted on a Finnigan LCQ. ............................................19 1-4 Ionalytics line-of-sight FAIMS cell geometry. ..................................................................20 1-5 LCQ quadrupole ion trap showing ion trajectory. ............................................................23 1-6 Example of oscillating i on motion according to the Ma thieu equation (let a=1 and q=1/5).. ...............................................................................................................................24 1-7 Mathieu stability diagram..................................................................................................25 2-1 Compensation voltage (CV) scanning. ..............................................................................31 2-3 Negative ion APCI mass spectrum of TNT. ......................................................................33 2-4 TNT total ion count over four 2 min CV scans. ................................................................34 2-5 FAIMS/APCI mass spectrum of TNT. ..............................................................................35 2-6 FAIMS/APCI mass spectrum of TNT: measured for 2 min at CV = 6.6V ......................36 2-7 FAIMS/MS/MS of TNT m/z 197.1 peak at 35% collision energy. ...................................37 2-8 CV spectrum of 2,4-DNT. .................................................................................................38 2-9 FAIMS/MS mass spectrum of 2,4-DNT. ...........................................................................39 2-10 2,4-DNT: 2 min at CV = 11.5V. .......................................................................................40 2-11 CV spectrum of 2,6-DNT. .................................................................................................41 2-12 FAIMS/MS mass spectrum of 2,6-DNT. ...........................................................................42 2-13 Mass spectra of 2,6-DNT: 2 min at CV = 11.6V and 15.3V. ...........................................43 2-14 CV spectrum of RDX. ........................................................................................................44 2-15 FAIMS/MS mass spectrum of RDX. .................................................................................45 2-16 RDX: 2 min at CV = 5.1V. ...............................................................................................46 8

PAGE 9

2-17 CV spectrum of HMX. .......................................................................................................47 2-18 FAIMS/MS mass spectrum of HMX. ................................................................................48 2-19 HMX: 2 min at CV = 1.2V. ..............................................................................................49 2-20 NG (MW 227) ....................................................................................................................50 2-21 FAIMS/MS mass spectrum of NG. ....................................................................................51 2-22 PETN (MW 316)................................................................................................................52 2-23 FAIMS/MS mass spectrum of PETN. ...............................................................................52 2-24 FAIMS/MS mass spectrum of ANFO made with hexane. ................................................55 2-25 FAIMS/MS mass spectrum of ANFO made with nitromethane. .......................................56 3-1 WSU electrospray ionization source. .................................................................................58 3-2 WSU electrospray io nization source photograph. .............................................................58 3-3 Drawing of Ionalytics orthogonal dome FAIMS cell. .......................................................60 3-4 Photograph of Ionalytics orthogonal dome FAIMS cell. ...................................................61 3-5 WSU apparatus. .................................................................................................................62 3-6 CV spectrum of HMX/RDX mixture. ................................................................................65 3-7 Mass spectrum of HMX/RDX mixture. .............................................................................66 3-8 CV spectrum of TNT/RDX mixture.. ................................................................................67 3-9 Mass spectrum of TNT/RDX mixture. ..............................................................................68 3-10 CV spectrum of both isomers of DNT.. .............................................................................69 3-11 RDX CV spectrum at 0.5 mm.. ..........................................................................................71 3-12 RDX mass spectra at 0.5 mm.. ...........................................................................................72 3-13 RDX CV spectrum at 0.75 mm. .........................................................................................73 3-14 RDX mass spectra at 0.5 mm. ............................................................................................74 3-15 RDX CV spectrum at 1.0 mm. ...........................................................................................75 3-16 RDX mass spectra at 1.0 mm. ............................................................................................76 9

PAGE 10

3-17 RDX CV spectrum at 2.0 mm.. ..........................................................................................77 3-18 RDX mass spectra at 2.0 mm. ............................................................................................78 3-19 Relationship between CV and DV. ....................................................................................80 4-1 Distributed plasma ionization source. ................................................................................90 10

PAGE 11

LIST OF ABBREVIATIONS AFRL Air Force Research Laboratory AGC automatic gain control APCI atmospheric pre ssure chemical ionization DARPA Defense Advanced Research Products Research Agency CV compensation voltage DC direct current DV dispersion voltage FAIMS high-field asymmetric-waveform ion mobility spectrometry IED improvised explosive device IMS ion mobility spectrometry LOD limit of detection MS mass spectrometry QITMS quadrupole ion trap mass spectrometer RDECOM Research, Development, and Engineering Command RF radio-frequency SNL Sandia National Laboratory TOF time-of-flight UXO unexploded ordnance VOC volatile organic compounds 11

PAGE 12

Abstract of Thesis Presente d to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science CHARACTERIZATION OF COMMONLY ENCOUNTERED EXPLOSIVES USING HIGHFIELD ASYMMETRIC WAVEFORM ION MOBI LITY SPECTROMETRY COUPLED WITH MASS SPECTROMETRY By Jared J. Boock May 2007 Chair: Richard A. Yost Major: Chemistry The goal of this research was to charact erize explosive compounds using high-field asymmetric waveform ion mobility spectrometry (FAIMS) as an ion separation device interface to a mass spectrometer (MS). FAIMS is a relatively recently devel oped technology that is promising, adding a dimension of separation. This method was employed in such a way as to be conducive to possible future development of fiel d instruments (e.g., use of atmospheric pressure chemical ionization (APCI)). In addition, se veral experiments were conducted with the application of this method, main ly dealing with optimization of the instrument parameters. These experiments led to the development of a new method: high-resolution FAIMS. An ion mobility spectrometer was used as a desolvati on and ion-focusing device to drastically improve FAIMS resolution. This allowed for the separatio n of mixtures that otherwise could not have been separated. The resolution was high enough that isomers of the same explosive were successfully resolved. This can be beneficial to forensics studies. 12

PAGE 13

CHAPTER 1 INTRODUCTION AND INSTRUMENTATION 1.1 Introduction A major threat to the national security of the United States is the continued use of explosives abroad and possible use by terrorists on the domestic front. 1 As a result, there is a great need for improved separation and detectio n technologies for trace amounts of explosive compounds. There are many possible applications for such technologies, such as prevention of terrorist events, military weapons/cache detecti on, interception of explosives trafficking, improvised explosive device (IED)/landmine detec tion and identification, post-event forensics analysis, and unexploded ordnance ( UXO) detection. A significant ch allenge to the detection of minute amounts of explosives is that oftentimes these analytes will be overwhelmed by complicated background matrices. 2 Another is that benchtop-size instruments are inadequate to provide timely analyses to meet the needs of the user. For instance, a military unit searching for landmines or IEDs in a combat zone cannot easi ly deploy a large and cumbersome instrument. To meet these needs, an instrument is required that can both provide se paration of analyte from atmospheric background containing large amounts of volatile organic co mpounds (VOCs). The required limit of detection (LOD) for such an instrument has been determined by many field trials conducted by government and commercial agencies (US Air Force Research Laboratory (AFRL), US Army Research, Development, and Engineering Command (RDECOM) Sandia National Laboratories (SNL), a nd Defense Advanced Research Products Agency (DARPA) are a few of these) to be in th e parts-per-billi on (ppb) range. 3 One of the primary technologies in use for e xplosives detection today (as of 2006) is the ion mobility spectrometer (IMS). 4 An IMS consists of an ionization source, an atmospheric pressure drift tube, and a detector. A sample is ionized and the ions ar e introduced to the drift 13

PAGE 14

tube and are passed through a seri es of ring electrodes held at increasing voltages of polarity opposite to that of the ions, whic h creates a strong electric fiel d. Separation is based on mobility of each ion in the field, which depends on size, shape, and mass of the ion. The IMS has limitations, however. First and foremost is that it is easily susceptible to contamination from background matrix. The IMS as an explosives detect or is primarily used in indoor facilities such as airports or railroad stations, and samples are usually introduced as a swipe rather than sampling vapor from air. 5 Mass spectrometry could address many of the limitations of ion mobility. Research is currently being conducted by many different orga nizations to design and construct a fieldportable time-of-flight (TOF) mass spect rometer (MS) to meet these needs. 6,7 Ions are accelerated through a vacuum drift tube by an elect ric field. The mass-to-charge (m/z) ratio is determined by the amount of time each ion takes to pass through the drift tube. In most cases, a TOF/MS is coupled with a gas chromat ograph, which provides initial separation. 8 Current commercially available fieldportable TOF/MS designs do not have a high enough resolving power to meet the above needs, and they are not sensitive enough to detect levels of explosive vapor that are practical to operational needs. 9 A relatively recent (theorized in the late 1980s and developed in the 1990s and 2000s) and promising technology is high-field asymmetric-w aveform ion mobility spectrometry (FAIMS), 10 which will be discussed in detail. FAIMS is a novel, continuous chemical separation method that has already proven successf ul in clinical, proteomics, and environmental analytical applications. 11 FAIMS does not have a high enough resolv ing power to function as a stand-alone device; 12 however, it is ideally suited to function as an ion separation device (in lieu of a chromatography instrument) that can be coupled to a mass spectrometer. 14

PAGE 15

1.2 Atmospheric Pressure Chemical Ionization Atmospheric pressure chemical ionization (APC I) is a gas-phase ionization technique that uses either a radioactive -emitting source or a corona discharge at a pressure of 1 atm to produce reagent ions and ultimately analyte ions. This method was chosen for this series of experimentation because it most closely resembles an ionization source that is amenable to be used for field instruments. Most other ionization techniques either require vacuum or are large, complex systems containing lasers or high-voltage ; these techniques are less suitable for a manportable field instrument. The APCI method is not identical to the ionization method to be employed projected field instruments, as a di fference arises from sample introduction. The APCI ionization source used for this series of experiments (see figure 1-1) is designed for Figure 1-1. Finnigan LCQ APCI source. [Rep rinted from Thermo-Electron Corp. 2004. Finnigan LCQ Series APPI/APCI Comb ination Probe Operators Manual Rev. A, Thermo-Electron Corp.] 15

PAGE 16

interfacing to liquid chromatogra phy, and, therefore, introduces analyte samples as liquid. The liquid sample is nebulized into droplets, which are passed pneumatically via nitrogen sheath gas into a heated region where they are vaporized. The nitrogen and vaporized solvent then serve as reagent gas as the vapor is pa ssed over a corona discharge wh ich produces reagent ions that ionize the analyte via ion-molecular reactions. 13 There are different mechanisms depending on the type of analyte and whether or not ionizati on is positive or negativ e. Explosive compounds tend to favor negative ionization (this behavior and ionization mechanisms will be discussed further in a later section). 1.3 High-field Asymmetric-Wavefo rm Ion Mobility Spectrometry As stated in the introduction, high-field asymmetric waveform ion mobility spectrometry (FAIMS) is a relatively recently deve loped technology that in this series of experiments serves as an ion separation device. FAIMS cells have ma ny different configurations, though all basically consist of two electrodes separate d by a gap; the simplest confi guration is two parallel metal plates, which, consequently, was the first iteration of this design (Figure 1-2). 15 An asymmetric waveform is applied between these plates, whic h can either be the sum of two sine waves, 12 or a square wave; in this case, a sum of sines was used with a high negative potential of -4000 V for a short time, and a low positive potential for a slightly longer time. This peak voltage is known as the dispersion voltage (DV), and cr eates a strong electric field. Th e rapid change in polarity of the applied alternating potentials causes the ion to oscillate due to attraction/repulsion forces. For this reason, a high negative potential is typically used for negative ions. The basis for separation for a FAIMS device is a change in the mobility of an ion as this field varies, unlike IMS which measures the mobility itself. 16 Ions exhibit a variety of mobility behaviors under high field, thus making FAIMS an ideal separation device. The changes in 16

PAGE 17

Figure 1-2. FAIMS waveform and ion mobility be havior. [Adapted from Guevremont, Roger. 2004. High-Field Asymmetric Waveform Ion Mobility Spectrometry (FAIMS). (p. 2, fig. 1-1). Ionalytics Corp., Ottawa, Canada.] mobility are classified into three different ca tegories: A-type behavior, or exponential increase in mobility proportional to change in field strength, B-type behavior, or exponential increase in mobility followed by exponential deca y as field strength increases, or C-type, which is exponential decay in mobility as field strength increases. The change in mobility affects the direction the ion travels toward or away from the plates (Figure 1-2). If these conditions are maintained, then all i ons will drift toward the plates according to each separate mobility. However, the drift of a pa rticular ion can be halted by the application of a small DC voltage to one of the plates. This compensation voltage (CV) allows a single type of ion to pass through the plates. As mobility ch anges are greater, an increased compensation voltage is needed to bala nce the drift and allow transmission of the ion. The mobility of an ion influenced by an electric field is described by K h (E) = K[1 + f (E)] where f (E) is the dependence of ion mobility (K h ) as a function of electric field. 12 17

PAGE 18

Field strength is describe d by units known as Townsend (Td), for which 1 Td = 10 17 (E/N) where N is the number density of the carrier gas (expressed in atoms/m 3 ) and E is the field strength (in V/cm). Change in mobility is described by K h /K = 1 + (E/N) 2 + (E/N) 4 where K h is the ion mobility at high field and K is the mobility at low field. The two constants and are dependent on the dimensions of the FAIMS cell. The estimated field strength for this series of e xperiments was calculated to be about 80 Td (20,000 V/cm) at 1 atm. To analyze a mixture of analytes, with diffe ring CVs, it is necessary to scan CVs over a range. Therefore, each analyte in the mixture is transmitted through the cell separately and a CV spectrum consisting of ion signal vs. CV can be acquired. A simple FAIMS cell geometry of two parallel plates is not necessa rily ideal. It is understood that when FAIMS is interfaced between an ionization source and mass spectrometer, there will be a degree of ion loss, though it should be noted that there is no loss of sensitivity. This may not make much of a difference; howev er, for even though signal is reduced, noise may be reduced even further. It ha s been proven that diffe rent types of cells such as the cylindrical cell used in this study have an additional ion focusing capability. 12 An Ionalytics (Thermo) Selectra beta II prot otype was used for the FAIMS work described in this thesis, though there were two different types of cell geomet ries used. The line-of-sight FAIMS cell was used for the characterization, and it consists of two concentric cylinders that are separated by a 1 mm gap. In orde r to assemble the cell, a solid stainless steel inner electrode fits into a hollow stainless steel outer electrode, and a PEEK endcap maintains uniform distance between these electrodes. The ends of the i nner and outer cylinders are machined to have 18

PAGE 19

hemispherical dimensions toward the exit aperture. Un like a flat-plate desi gn, the electric field in a cylindrical cell is not unifo rm; this configuration, along with the hemispherical exit area, allows for the focusing of ions toward the exit aperture. A PEEK sleeve prevents arcing and other types of electrical interf erence between the FAIMS cell and the environment. Figure 1-3 shows a photograph of the APCI source and FAIMS cell. APCI Source Curtain Plate Waveform lead FAIMS Cell Curtain Plate lead Figure 1-3. APCI source and FAIMS cell mounted on a Finnigan LCQ. Prior to entering the FAIMS cell, ions pass th rough a hole in a curtain plate, which has a potential of the same polarity as the ion mode of the source, in order to ensure that ions are 19

PAGE 20

repelled into the FAIMS cell. For positive ion mode, the curtain plate voltage is kept constant at +1000V, while negative ion mode curtain plate volta ge is set to V. Figure 1-4. Ionalytics line-o f-sight FAIMS cell geometry. A clean, dry curtain gas is adde d to the FAIMS cell via a side port at a rate of 3 L/min. This curtain gas serves two purposes. First, a portion exits through the orifice in the curtain plate, which assists with desolv ation of the ions and droplets from the ionization source and minimizes the accumulation of solvent neutrals fro m the APCI source into the FAIMS cell. The remaining gas propels the ions through the FAIMS cell and into the heated capillary of the mass spectrometer. This gas can cons ist of a pure gas or a gas mixtur e, typically nitrogen, helium or carbon dioxide; however, nitrogen was chosen for these experiments due to its predominance in 20

PAGE 21

the atmosphere. Any moisture or trace hydrocarbons introduced to the carrier gas will affect the performance of the FAIMS device by reacting with the ions, so charcoal and/or molecular sieves are used, to purify the gas. A waveform generator serves as the power s ource for the asymmetric waveform. A cable connects the generator directly to the FAIMS cell. There is also a connector on the waveform generator that allows the operato r to observe the waveform on an oscilloscope and ensure that the waveform DV and shape are constant. The Selectra prototype FA IMS unit is extremely sensitive, and any slight interferences with th e waveform cable can affect the waveform. There are four different FAIMS modes depe nding on the polarity of the CV and DV polarity required to transmit a particular ion. 12 The determination as to which mode an ion falls under at this juncture is made experimentally, as a library has not yet been fully constructed, which justifies a portion of this thesis. An ion in the N1 qua drant (including many explosives) passes through the FAIMS cell with a negative DV and positive CV; an ion in N2 requires positive DV and positive CV; an ion in P1 re quires positive DV and negative CV; and P2 ions are transmitted with negative DV and CV. This rule applies only for A and C type ions; for B type ion mobility behavior it is not well defined. The mobility behavior of an ion is affected by many factors, including: size, rigidity, and interacti on with carrier gas (according to Stokes and Einstein). 17 Although use of FAIMS as a separation device decreas es total ion transmission, when the CV is set for a particular ion of interest, the relative signal to noise ratio of that ion increases greatly. 1.4 Ion Optics The instrument used for this experimentat ion was a Finnigan (The rmo-Electron) LCQ, which is a commercial, bencht op quadrupole ion trap mass spect rometer (QITMS). After the 21

PAGE 22

analyte is vaporized, ionized by APCI, and sepa rated via FAIMS, ions pass through the heated capillary into a lower-pressure area, where the tube gate is located. The tube gate carries two potentials of a polarity opposite to the analyte ions one which focuses the ions into the skimmer, which acts as a restriction between the higher pressure area (1 torr) of the tube gate and the lower pressure (10 -3 torr) of the ion optics. The second potenti al is used to deform ions and prevent them from traveling further. The ion optics consis t of a series of an octopole, lens, and a second octopole. An octopole is a set of eight metal rods that carry an RF voltage (400 V) and DC offset voltage (+10 V for negative ions) that induces an electric field that directs and focuses ions into the mass analyzer. The le ns between the octopoles serves both as a focusing medium and as a pressure restriction, for the pr essure once again decreases to 2x10 -5 torr in the second octopole and mass analyzer. As pressure is continually decreased, the mean free path (mfp) is increased, and there are fewer neutral molecu les and other ions present to in terfere with the passage of the focused analyte ion beam. 18,19 1.5 Quadrupole Ion Trap Mass Spectrometry A quadrupole ion trap (Figure 1-5) consists of a hyperbolic ring electrode to which an RF potential is applied, as well as two end caps with a DC potential. 20 A DC offset voltage (-10 V for negative ions) is applied to the end caps so ions will enter the trap. An RF waveform of a particular drive frequency ( ) is applied to the ring electrode which creates the quadrupolar electric field, causing the ions to oscillate on a controlled trajecto ry. He gas at approximately 10 -3 torr removes excess kinetic energy from the i ons so that they can be trapped. The ion trajectory in the center of the trap is not to scale. 22

PAGE 23

Figure 1-5. LCQ quadrupole ion tr ap showing ion trajectory. [Adapted from Thermo-Electron Corp. 2003. Finnigan LCQ Series Hardware Manual Rev. A.] The oscillating ion motion is described by the Mathieu equation: 22 2222 16 2 oo r zzrm eU aa 2222 8 2 oo r zzrm eV qq The variables a z and q z are functions of the dimensions of the trap and the potentials applied, while z and r represent th e axial and radial directions (b etween and perpendicular to the endcaps, respectively); U is the DC amplitude app lied to the ring electrode (if any), V is the RF potential applied to the ring elec trode, e is the charge on an ion, m is the mass of an ion, r o is the inner radius of the ring electrode, z o is the axial distance from the center of the device to the nearest point on one of the endcap electrodes, and is the angular drive frequency. The angular drive frequency is equal to 2 f RF where f RF is the frequency of the main RF voltage in Hertz. 23 A 23

PAGE 24

graphical example of the motion in arbitrary directions is shown in figure 1-6. The actual motion is similar to this example. Figure 1-6. Example of oscillat ing ion motion according to the Mathieu equation (let a=1 and q=1/5). [Adapted from McLachlan, N. W. 1947. Th eory and Application of Mathieu Functions, Dover]. For an LCQ, q z = 0.1963 [RF frequency (number of charges)/ion mass]. 19 Solutions of the Mathieu equation describe a stability diagram (figure 1-7) from which it can be determined whether or not an ion will remain in the ion trap under a given set of conditions. If the q z of an ion with known mass/charge falls wi thin the boundaries of the stability diagram (is stable in both axial and radial directions), th en it can be trapped. If the q z falls outside the boundary, then the ion will collide with the end cap el ectrodes and be lost. The factor is based on the secular frequency of the oscillat ion of an ion. When equals one, the secular frequency equals half the frequency of the RF field, and the magnitude of its oscillation increases such that the ion is lost. Essentially, the lines on the graph are voltage scans; in a quadrupole ion trap the amplitude of the DC and RF voltages are ramped (while keep ing a constant RF/DC ratio), to obtain the mass 24

PAGE 25

spectrum over the required mass range. The sensitivity is a function of the scanned mass range, san speed, and resolution. 24 Figure 1-7. Mathieu stability diagram. [Reprinted from McLachlan, N. W. 1947. Theory and Application of Mathie u Functions, Dover]. The ion trap can only hold a certain number of ions at a time before repulsion forces (space charge) cause ion displacement, resulting in a lo ss of resolving power and band broadening. To counter this phenomenon, the LC Q employs a function known as automatic gain control (AGC) where a prescan is conducted that monito rs the generation of ions in the trap. 25 The AGC scan and the full MS scan form a microscan. In thes e experiments, three microscans were used for each ion injection period. 25

PAGE 26

1.6 Multidimensional Mass Spectrometry Another capability of the LCQ is multidimension al MS or MS/MS. This function can be selected to determine the fragmentation patterns of an ion of a particular m/z in order to identify the ion creating the peak or to examine the chemistr y of that ion. A ion is selected, known as the parent ion, and all other ions are ejected from the trap. An excita tion RF voltage is applied across the end cap electrodes, causing the parent ion as kinetic energy to increase; collisions with the He gas in the trap result in parent ion disso ciation into fragment (daughter) ions. The fragmentation pattern can be changed by altering the amplitude of the excitation RF voltage. 19,26 1.7 Thesis Overview This thesis is organized into four chapters. Chapter 1 is the intr oduction to the research and an overview of the equipment used for the ch aracterization portion. Ch apter 2 presents data relating to the characterization of explosives via FAIMS/MS. Chapter 3 introduces the second instrument and compares it to the instrument used for charac terization. There were several experiments relating to FAIMS/ MS and IMS/FAIMS/MS that we re conducted throughout this research; these are discussed in chapter 3. Chapte r 4 discusses the conclusions of this research and recommendations of other related efforts, as well as possible long-term future work. 26

PAGE 27

CHAPTER 2 PROPERTIES AND CHARACTERIZATION OF EXPLOSIVES 2.1 Explosives that were Characterized A list of explosives was selected for ch aracterization by FAIMS/MS based on threat assessments by various government agencies (i ncluding the Department of Homeland Security (DHS) and Federal Bureau of Investigation (FBI)). This list is contained in the table 2.1. Each of these compounds will be disc ussed in great detail below. Table 2-1. Explosives to be Characterized. Compound Formula Molecular Weight TNT (2,4,6-trinitrotoluene) C 7 H 5 N 3 O 6 227 HMX (cyclotetramethylene tetranitramine) C 4 H 8 N 8 O 8 296 RDX (cyclotrimethylene trinitramine) C 3 H 6 N 6 O 6 222 2,4-DNT (dinitrotoluene) C 7 H 6 N 2 O 4 182 2,6-DNT (dinitrotoluene) C 7 H 6 N 2 O 4 182 NG (nitroglycerin) C 3 H 5 N 3 O 9 227 PETN (pentaerythrito l tetranitrate) C 5 H 8 N 4 O 12 316 ANFO (ammonium nitrate fuel oil) * *dependent on type of fuel used 2.2 Classification of Explosives An explosive material is chemically unstabl e or becomes unstable under certain conditions, and as a result of rapid exothermic decomposition, can be destructive. There are many types of explosives; however, government threat assessments 1,27 have determined that only a certain number of them are significant threats to life and property, primarily due to factors such as supply and cost. As a result of this determinatio n, the compounds of the highest threat levels were examined closely in these experiments, as well as their precu rsors and degradation products, if possible, as the presen ce of these can also be used to detect manufacture or storage of the respective explosive materials. 27

PAGE 28

There are two forms of explosive decomposition. The more powerful of these, detonation, is characterized by an extremely fast supersoni c shock wave released by the energy produced in the favorable chemical reaction. The sudden high pressure of the shock wave is the major cause of destruction resulting from a detonation. The other form of decomposition, deflagration, is the subsonic portion of an explosion, usually propogate d by thermal conductivity of the substance in contact with the outer edge of the plume. Explos ives are classified as either low or high depending on whether or not they detonate. 28 Low explosives include gunpowder and pyrotechnics. High explosives are those used for military applications, mining, or demolitions. All of the explosive compounds included in this study ar e considered high explosives. There are also subcategories of high explosives: primary, s econdary, or tertiary. Primary high explosives are extremely sensitive to shock, friction, and heat, and ar e difficult to store. Secondary high explosives are stable enough to be stored for a period of time, yet can be detonated due to a sudden shock or heat. Most of the compounds in these experiments are secondary high explosives. Tertiary explosives (b lasting agents) will not easily detonate without the aid of a primary or secondary booster. This list includes ammoni um nitrate fuel oil (ANFO), which was a su bject in this study. 2.3 Chemical Properties of Explosives The majority of secondary high explosives ch aracterized in this study are nitroaromatics, nitramines, or nitrate esters. In the nitroa romatics and nitramines, one or more nitro (NO 2 ) groups are attached to an aromatic ring structure. Nitrate este rs also contain nitro groups. These nitro groups are bulky and electronwithdrawing, causing the nitrogen in the nitro groups to be partially positive. Under APCI, primarily negative ions ([M] or [M-H] ) are formed. Nonhalogen groups with atoms that are more elec tronegative than carbon draw substantial electron density from the aromatic system. If enough intern al energy is applied, one or more of the nitro 28

PAGE 29

groups may detach from the ring or central struct ure, producing fragments. In either case, the ions produced are negative, and negative mode must be used in order to separate and detect explosives. Nitroaromatics and nitramines are generally st able, but can be reactive; this property is especially apparent in functional gro ups attached to the ring at a position meta to a nitro group. 29 These compounds are thermally labile, so ionization techniques such as electron ionization (EI) and chemical ionization (CI) will result in signi ficant fragmentation of the molecule, which makes detection difficult in a complicated bac kground matrix. A softer ionization technique, such as atmospheric pressure chemical ioniza tion (APCI) or electrospr ay ionization (ESI) is needed. 13 Due to their low volatility under standard conditions, only a small amount of these explosives will vaporize for detection in air, so a sensitive separation and detection techniques are required. Because these compounds can conden se on dust or other particles present in the immediate environment, instruments designed to detect explosives in air should often sample for explosives present on particles 30,31 as well as the small amounts of vapor. Nitrate esters are much less stable, and require lower instru ment operating temperatures for analysis. Explosive compounds are also highly toxic, 28 creating an environmental hazard for areas around disposal sites. Vapors can be introduced to the lungs, and liquids can be absorbed through the skin, causing a vari ety symptoms depending on the relative concentration, even leading to death. 32 2.4 Instrument Settings Both positive and negative ions can be form ed in the APCI source, and the LCQ can analyze ions of either polarity. As most of the explosives have large electronegative leaving groups, they favor the formation of negative ions. 33 It has also been shown that there is less 29

PAGE 30

background noise in negative mode than positive m ode. All of the mass spectra shown consist of 50 analytical scans, each consisting of three microscans. The explosive solutions were diluted in a solvent containing 64.9% methanol and 35% deionized water, as earlier studies 13 have yielded excellent results us ing this ratio. Solvents that evaporate easily are best used for APCI, so that lower vaporizer temperatures can be used to prevent thermal degradation of the analytes of interest. An extremely small amount, approximately 0.1%, of carbon tetrachloride was adde d to the solutions as some of the explosive compounds favored the formation of a chloride adduct. This behavior will be discussed in turn. All solutions were diluted to a c oncentration of about 10 ppm, as th is is fairly realistic based on calculated explosive detection applications. 34 Solutions containing explosives were injected directly via the syringe pump of the LCQ into the vaporizer at a flow rate of 20 L/min with a maximum ion injection time of 50 ms for AGC. The vaporizer was kept at a temperature of 300C, while the heated capillary was set to 150C. The LCQ software was used to tune the instrument as needed throughout the study in order to maximize signal strength. 2.5 Compensation Voltage (CV) Scans For all of the characterization in this section, CV was scanned from 0-20V four times over 2 min, for a total period of 8 min. The transition at the end of the scan from 20V back to 0V is extremely fast, covering a period of milliseconds. The ion of interest will only pass through the FAIMS at its relative CV, therefore there will be a peak of ion flow at this voltage (ideally 100% transmission); at other voltages the ion collides with the wall of the cell and is not transmitted. (see Figure 2-1) This results in a peak at a particular time (a nd thus CV) in each of the four scans. 30

PAGE 31

Figure 2-1. Compensation voltage (CV) scanning. 2.6 Characterized Compounds 2.6.1 Nitroaromatics 2.6.1.1 (2,4,6 ) Trinitrotoluene (TNT) TNT is one of the most commonly encountered explosives due to its continued use for over a century in both military and commercial applicati ons. It is deemed a significant threat for use by terrorists and other enem ies of the United States. Like othe r nitroaromatics, it is stable until ignition, other than slow decomposition over a period of years. 28 Thus, there is the possibility of low amounts of degradation product s present in samples of TNT. Therefore, the most common 31

PAGE 32

of these should therefore be char acterized, as well as precursors used in the manufacture of TNT. These compounds include the six dinitrotoluene isomers and trinitrobenzene (TNB). For the structures of the nitroaromatics characterized in this study, refer to Figure 2-2. TNT has been used on its own or as a component in mixtures such as with ammoni um nitrate (amatol), aluminum powder, or other explosives. Figure 2-2. Nitroaromatics and molecular weights. [Adapted from Reich, Richard. 2001. (Figure 1-11). PhD dissertatio n. University of Florida.] When TNT detonates it (i deally) decomposes accordi ng to the following equation: 28 2C 7 H 5 N 3 O 6 7CO + 7C + 5H 2 O + 3N 2 although complete decomposition is only reache d after some oxygen from surrounding air is added as an ignition reactant. 32

PAGE 33

[M-H][M][M-NO]-[MOH]Figure 2-3. Negative ion APCI ma ss spectrum of TNT. The firs t spectrum (Figure 2-3) shows negative ion APCI of TNT wit hout FAIMS separation. The [M] ion at 227.0 and [M-H] ion at 226.1 can be seen clearly, a nd a small amount of background noise (ions that correspond to minor fragment ions of TNT or impurities) is visible. 33

PAGE 34

Figure 2-4. TNT total ion count over four 2 min CV scans. Figure 2-4 shows the CV spectrum (total ion count) for TNT. As mentioned ear lier, the ion of interest will only pass through the FAIMS at its corre ct CV; therefore, there ar e ion peaks at these CV values. There were four CV scans from 0-20 V, each with a duration of 2 min, for a total of 8 min. It is from these spectra that the exact CV for each compound can be calculated. From this spectrum, the CV can be calculated. For instance, the peak at 0.64 min: (0.64 min/2.00 min) 20.0V = 6.4V. 34

PAGE 35

[M]-[M-NO][M-OH]Figure 2-5. FAIMS/APCI mass spectrum of TN T. Figure 2-5 shows the FAIMS/MS spectrum for TNT. The background noise has been greatly reduced after the application of the FAIMS. (see figure 2-2) The evident trad eoff for the improved signal/noise ratio is the 50 reduction in signal strength. In addition, two fragments of TNT also pass through the FAIMS, as the fragment ion CV s are very close to the CV of the TNT [M] ion. The fragments show that TNT is a delicate ion that readily loses NO and OH. The fragment peaks are in larger proportion to the molecular ion peak when FAIMS is applied, most likely because the to tal ion count is lower, and there are few to no impurities. Also, the [M-H] no longer appears. 35

PAGE 36

[M]-[M-OH]-[M-NO]-[M-2NO][M-(2NO+CH3)]Figure 2-6. FAIMS/APCI mass spect rum of TNT: measured for 2 min at CV = 6.6V Figure 2-6 shows a 2 min spectrum taken at a CV of 6.6V, which is when the most TNT [M] ion passes through the FAIMS. This spectrum is si milar to that in the previous figure; the only difference is that instead of averaging one of the CV peaks, data points were taken for a stationary CV. As a result, this spectrum is improved further from that in figure 2-4. 36

PAGE 37

[M-NO]-[M-2NO]-[M-(NO+OH)]Figure 2-7. FAIMS/MS/MS of TNT m/z 197.1 p eak at 35% collisi on energy. Figure 2-7 demonstrates that FAIMS/MS/MS of even w eak ions from explosives is feasible. The FAIMS CV was set to 6.2V to allow the [M-NO] fragment of TNT to pass through the cell. This ion was collided with 35% di ssociation energy, causing the loss of a second nitro group. In other word s, the predominant fragment ion peak shown here is the same elemental composition as the [M] ion for mononitrotoluene. 37

PAGE 38

RT: 0.00 8.01 SM: 7B Figure 2-8. CV spectrum of 2,4-DNT. Figure 28 shows the CV spectrum for 2,4-dinitrotoluene (DNT), which is a precursor of TNT, a nd is often an impurity in explosives containing TNT. This CV spectrum is different from that of TNT (figure 2-4) in that the CV is 11.5V, compared to 6.6V. In a ddition, there is a small amount of solvent ions that transmitted at slightly lower CV than the major DNT ion through the FAIMS cell. 0 1 2 3 4 5 6 7 8 Time (min) 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100Relative Abundance NL: 6.59E4 TIC MS 24DNT 5.15 1.13 1.16 3.17 7.18 1.21 3.11 3.22 7.28 7.00 1.00 5.32 7.35 2.96 0.02 6.95 4.97 1.35 6.05 2.91 6.01 2.23 0.39 2.15 4.09 4.31 6.08 6.41 2.83 5.62 1.56 7.53 38

PAGE 39

[M-H]-[M-NO][M-(NO+CH3)][M-2NO]-[M-(NO+2H2O)][M]Figure 2-9. FAIMS/MS mass spectrum of 2,4-DN T. Figure 2-9 shows the mass spectrum of 2,4DNT after FAIMS separation (averaged over the peak at 6.6 min in the CV spectrum in figure 2-8). As with TNT, this ion fr agments readily, and experiments have shown that many of these fragments have CVs that are close to the primary ion. 39

PAGE 40

[M-H]-[M-(NO+CH3)][M-NO][M-2NO]Figure 2-10. 2,4-DNT: 2 min at CV = 11.5V. A mass spectrum of 2,4-DNT at a fixed CV of 11.5V is displayed in figure 2-10. There is little to no noise. 40

PAGE 41

RT: 0.00 8.01 SM: 7B 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 Time (min) 0 20 40 60 80 100 0 20 40 60 80 100Relative Abundance 0 20 40 60 80 100 5.49 5.54 1.54 1.59 7.47 7.44 3.53 7.49 3.50 1.49 5.61 7.55 5.15 1.41 5.12 1.65 7.63 7.15 1.21 3.17 7.18 3.71 3.12 1.13 5.08 7.08 5.71 3.02 3.86 2.54 1.93 5.02 6.59 0.75 5.87 0.56 4.68 7.57 7.52 5.56 3.51 3.36 5.62 7.40 7.89 5.49 3.57 1.75 1.54 1.42 5.47 3.33 3.97 7.37 3.11 5.28 1.29 5.81 7.18 7.06 1.21 3.04 5.49 1.54 5.46 1.59 5.54 7.47 7.44 3.50 1.49 3.53 7.49 3.43 5.62 7.39 1.64 7.58 5.39 1.79 1.41 3.36 7.37 3.76 3.33 5.71 5.33 1.35 3.86 7.26 5.87 1.91 3.01 1.02 5.14 7.04 NL: 3.48E4 TIC MS 26DNT Figure 2-11. CV spectrum of 2,6-DNT. 2,6-dinitrotoluene is an isomer of 2,4-DNT and another precursor of TNT. Both nitro groups are in ortho positions to the toluene methyl group. 2,6-DNT is more likely than 2,4-DNT to lose an NO group, and as such, there are two predominant ions formed in APCI, causing the CV spectrum to display different behavior than 2,4-DNT. The tw o ion peaks have close CVs and are not completely resolved from each other. These can be seen in figure 2-11. Experimentation in improvement of FAIM S resolving power was conducted and will be discussed in chapter 3. NL: 4.89E3 m/z= 181.4-182.4 MS 26DNT NL: 2.33E4 m/z= 151.5-152.5 MS 26DNTm/z 152 m/z 181 m/z 152 m/z 181 TIC [M-H][M-NO]41

PAGE 42

[M]-[M-NO]-[M-2NO][M-H]Figure 2-12. FAIMS/MS mass spectrum of 2,6-DNT. Figure 2-12 is a mass spectrum averaged over the entire 8 min run. The two predom inant ions are visible at m/z 181 ([M-H] ) and m/z 152 ([M-NO] ), as described at figure 2-11. 42

PAGE 43

[M-H][M-NO][M-NO][M-H][M]-[M]CV = 15.3V CV = 11.6V Figure 2-13. Mass spectra of 2,6-DNT: 2 min at CV = 11.6V and 15.3V. The fact that 2,6-DNT more readily fragments under APCI than its isomer does not a ffect the separation when the CV is fixed, as each ion is tran smitted through the FAIMS cell at a separate CV. This behavior can be seen in figure 2-13. Note that the optimal CV values for the [M-H] ions of the two isomers are too close (11.6V vs. 11.7V) for them to be separated. However, only 2,6-DNT produces a second FAIMS peak (at CV = 15.3V) for the [M-NO] fragment ion. 2.6.2 Nitramines 2.6.2.1 Cyclotrimethylene trinitramine (RDX) The nitroamine RDX (research department co mposition X) was developed as an explosive during the 1930s and was used widely during Wo rld War II. This explosive is found in many mixtures (such as Torpex, RDX mixed with TN T and aluminum powder), though it is usually encountered as the base for several types of pl astic explosives, the most common of which is Composition C-4 (RDX with polyisobutylene and di(2-ethylhexyl)sebacat e as the binder and plasticizer). RDX is stable at room temp erature will not detonate without a detonator. 35 Another 43

PAGE 44

type of plastique that contains RDX is Semtex, which is a commercial explosive that has been used by terrorists many times in the past (such as Pan Am Flight 103 in 1988). The other component in the Semtex mixture is PETN (pentaer ythritol tetranitrate), which will be discussed in its own right. RDX decomposes completely according to: C 3 H 6 N 6 O 6 3CO + 3H 2 O + 3N 2 and also reacts with oxygen in the air to complete the reaction. RT: 0.00 15.99 SM: 7B Figure 2-14. CV spectrum of RDX. Figure 2-14 shows the CV spectrum for RDX. The peaks are fairly well reproducible. There are two sm all peaks before the ion of interest in all four scans, which are identified in figure 2-15. 0 2 4 6 8 10 12 14 Time (min) 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100Relative Abundance NL: 9.75E5 TIC MS RDX15 9.52 5.50 13.50 1.56 13.16 8.84 4.86 12.86 8.11 0.17 12.09 4.36 4.08 11.60 15.72 5.87 7.76 13.96 2.22 44

PAGE 45

[M+ 35 Cl][M+ 37 Cl][NO3][M+NO3]Figure 2-15. FAIMS/MS mass spect rum of RDX. The [M+Cl]adduc t is formed with the small amount of carbon tetrachloride in the solven t. (see Figure 2-15) One property of both RDX and HMX is the formation of little to no molecular ion. Interestingly, the adduct formed of RDX with the 37Cl isotope of chloride can also be seen in the mass spectrum. The first additional peak shown in the CV spectrum (figure 2-14) is m/z 62 which is an [NO3]fragment peak; it is seen in many mass spectra of explosives, especially at higher temperatures. The s econd is a small amount of m/z 286, which is another adduct formed with RDX: [M+NO 3 ] 45

PAGE 46

[M+ 37 Cl]-[M+ 35 Cl]Figure 2-16. RDX: 2 min at CV = 5.1V. Figure 2-16 demonstrates that FAIMS/MS is an effective method of separating and detecting ch loride adducts of RDX, as there are no interferences. 2.6.2.2 Cyclotetramethylene tetranitramine (HMX) HMX, derived from High Molecular Wei ght RDX (sometimes known as High Melting Point Explosive), is a nitroamine discovered as a byproduct of RDX that consists of an eightmembered ring structure rather th an a six-membered ring. It is slightly more powerful than RDX because molecules of greater mass generate gr eater energy when they decompose. Another difference between HMX and RDX is that HMX has a greater temperature at which the molecule begins to fragment (hence the name). 35 HMX is used exclusively for military applications, and is 46

PAGE 47

most often included in a mixture, such as in so me types of plastic bonded explosives, or octol, formed with TNT. Figure 2-17. CV spectrum of HMX. HMX behaves similarly to its nitroamine counterpart RDX. The FAIMS CV spectrum shown in figure 2-17 reflects this. There are some differences between the two, though. HMX has a different CV (thus it can be resolved, as will be discussed in chapte r 3), and there is no evidence of the [NO 3 ] peak or formation of the [M+NO 3 ] adduct. 47

PAGE 48

[M+37Cl]-[M+35Cl][M]Figure 2-18. FAIMS/MS mass spectrum of HMX. Similar to RDX, the dominant ion for HMX was [M+Cl] This can be seen in figure 2-18. Once again, adducts of HMX were formed with both dominant isotopes of chloride. There was also evidence of a small amount of [M] 48

PAGE 49

[M+ 35 Cl]-[M+ 37 Cl]-[M]Figure 2-19. HMX: 2 min at CV = 1.2V. Figure 2-19 shows the HMX [M+Cl] peak as the CV is fixed at 1.2V for a period of 2 mi nutes. The mass spectrum contains no interferences, and there is a small amount of the molecular ion peak, as there was during the CV scans (figure 2-18). 49

PAGE 50

2.6.3 Nitrate Esters 2.6.3.1 Nitroglycerin (NG) There were two nitrate esters that underwent characterization via FAIM S/MS. The first of these is nitroglycerin. Nitrat e esters contain nitro groups like the previously mentioned explosives, though they are bonded to oxygen atoms, rather than nitrogen or carbon. There is also no ring structure, unlike the nitroaromatics an d nitramines (Figure 2-20). Because of this structure, nitrate esters are extremely unstable, and fragment easily, requiring only a light shock to explode. Nitroglycerin is one of the ol dest known explosives, discovered in the midnineteenth century. 28 It is mainly used in the manufacture of dynamite, gunpowder, and rocket propellants. Figure 2-20. NG (MW 227) [Adapted from Paul W. 1996. Explosives Engineering, (Figure 3-3 p. 76), Wiley]. 50

PAGE 51

[NO3]NG (MW 227) Figure 2-21. FAIMS/MS mass sp ectrum of NG. Similarly to th e nitroamines, there is no molecular ion for NG. It also did not form adducts. Even though the heated capillary of the APCI source was decr eased to the minimum of 100C, all of the nitrate branches of NG were fragmented. These ni trate ions (m/z 62) were separated and detected with FAIMS/MS, as shown in Figure 2-21. Because the mass range will not extend below 50 amu for this instrument, it is not known how the carbon structure fragments. Ion count was about two orders of magnitude lower than most of the other explosives. 2.6.3.2 Pentaerythritol tetranitrate (PETN) PETN is the second nitrate ester that was characterized. It is related in structure but is less stable than nitroglyceri n and is more sensitive to shock or friction, and is one of the most powerful known explosives. 28 (see Figure 2-22) PETN was used in antitank weapons during WWII, and is today used in military demolition charges, in detonator charges for landmines and artillery, and in some types of small caliber am munition. It is never em ployed alone due to its relative instability, and is always used in a mixture to enhance explosive capability and effectiveness. 51

PAGE 52

Figure 2-22. PETN (MW 316). [Adapted from Paul W. 1996. Explosives Engineering, (Figure 3-3 p. 76), Wiley]. PETN (MW 316) [NO3][M+NO3][M+Cl][M-NO2]Figure 2-23. FAIMS/MS mass sp ectrum of PETN. Like the nitroamines and NG, PETN does not yield a detectable molecular ion. It di d form both chloride and nitrate adducts; however, these were in extremely low intens ity. As a result, PETN did not form any discernable peaks of high intensity during th is experimentation. However, due to its extreme instability, (also like NG) PETN did fragment into nitrate ions which were separated and detected by FAIMS/MS (Figure 2-23). As this figure shows, the total ion count is relatively low. 52

PAGE 53

2.6.3.3 Ammonium Nitrate Fuel Oil (ANFO) ANFO is the most commonly encountered e xplosive today, used mostly for mining operations. It is simple and inexpensive to produce, and the components are not difficult to acquire. It is one of the most used explosives by terrorists, having been used by organizations such as: Fuerzas Armadas Revolucionarias de Colombia (FARC), Provisional Irish Republican Army (IRA), Palestinian extremists, and some Amer ican extremists. Ammonium nitrate (used in fertilizers) is an oxidizer and can explode inefficiently on its own if shocked. A hydrocarbon mixture provides fuel for the reaction. Usually kerosene or diesel fuels serve this purpose industrially; although, terrorists have used othe r fuels (such as nitromethane to make the kinepak mixture used for the Oklahoma City bombing). Since there are so many different production methods and types of fuel s used in different proportions no two batches are identical. This, coupled with the fact that ammonium nitrate and hydrocarbon fuels are present in background air, suggest that ANFO will be extremely difficult to detect. On the other hand, a destructive bombing requires large amounts of it (eg., over 5,000 pounds at Oklahoma City) Ammonium nitrate decom poses according to: 2NH 4 NO 3 4H 2 O + 2N 2 + O 2 the most important product being oxygen. The hydrocarbon fuel added burn s rapidly to include CO 2 as a product. 28 ANFO is extremely stable, but is hygroscopic, which hinders long-term storage. ANFO will not explode without a seconda ry booster (blasting cap); the pow er of the explosion can be controlled by altering the mixture. Five different types of ANFO were examined in this re search. The hydrocarbon fuels used were: n-pentane, n-hexane, tolu ene, nitromethane, and diesel fu el. All were made using about 94% (by mass) ammonium nitrate and the remainder of fuel. The pentane and hexane were used to determine the effects of using a simple stra ight-chain hydrocarbon fuel to simplify (and thus 53

PAGE 54

identify peaks in) the mass spectra. Nitromet hane was used to produce the kinepak mixture that was discussed above; toluene was used to provide a large structure with a methyl group to compare and contrast results to kinepak ANFO. Characterization was done on the FAIMS/MS in both positive mode and negative mode Dilute solutions (about 1015 ppm) were heated slightly to instigate the oxidation reac tion without causing detonation. Characterization of ANFO using this method is difficult. There are no other mass spectra with which to compare the results of this effort. This is likely due to the fact that no two batches of ANFO are alike, and the mass spectra are diffi cult to reproduce. A limitation of the LCQ is that the mass range cannot be set below m/z 50; th erefore, fragment ions from ammonium nitrate and the fuels cannot be seen. Nearly all of the negative ion APCI data sets collected show an ion base peak in common: [NO 3 ] (m/z 62). This may result from ammonium nitrate that has not yet reacted. For those fuels of m/z greater than 50; though, the fuel itsel f was not seen neither in positive mode MS, nor in negative mode MS (though straight chain hyd rocarbons are almost never seen in negative mode MS), the exception to this being ANFO ma de with diesel. In every case for ANFO, a single 16 min scan was conducted comprising an area from CV -40V-40V. In each of the following figures (2-24 and 2-25) a single peak was detected. 54

PAGE 55

[NO3]Figure 2-24. FAIMS/MS mass spectrum of ANFO made with hexane. Figure 2-24 shows a FAIMS/MS mass spectrum of ANFO made with hexane. The base peak is [NO 3 ] which passed through the FAIMS cell at a CV of 9.7V. There is also a small peak at m/z 125. 55

PAGE 56

[NO3]Figure 2-25. FAIMS/MS mass sp ectrum of ANFO made with nitromethane. ANFO made with nitromethane (figure 2-25) as the fuel ha s a much lower ion signal than that made with hexane; this behavior was also seen with ANFO made with tolu ene. It also has a [NO 3 ] base peak, though there are some other ions as well. These ions were in common with some of the other runs as well: m/z 125, m/z 141, and m/z 284. Detection of ANFO with a fiel d instrument would prove to be difficult, as ammonium nitrate exists in the ba ckground (mostly in fertilizers) and the most common fuel, diesel, is used to power engines everywhere. The most likely type of ANFO to be detected would be kinepak, for nitromethane is used for little other than as a fuel additive for race cars. 56

PAGE 57

CHAPTER 3 FAIMS/MS AND IMS/FAIMS/MS OF EXPLOSIVES 3.1 Instrumentation The experiments described in this chapter were conducted using a different instrument configuration than that in chap ter 2. As in the other experime nts, the mass spectrometer was a Finnigan LCQ. 3.1.1 Electrospray Ionization (ESI) Source Electrospray ionization (ESI), employed in th is study, is similar to APCI. Instead of relying on a corona discharge to ionize molecules in the gas phase, ESI uses electrical charge to ionize molecules in solution. A liquid solution cont aining the analyte of in terest in solvent is passed through a capillary held at high potential Droplets leaving the capillary become highly charged. As the solvent in the droplets evaporat es, ionized molecules are ejected into the gas phase. 36 The unique non-commercial ioni zation source used in the studies in this chapter was developed by the Herbert H. Hill laboratory at Wash ington State University (WSU). It is a fairly simple apparatus, consisting of a fused silica ca pillary held in place by a ferrule, and mounted in a nonconductive teflon casing. Sample is introduced via syringe pump into this capillary, to which approximately -13.5 kV of ionization pote ntial is applied to a stainless steel ferrule through which the capillary is threaded. Due to the relatively small (about 0.5 mm) internal diameter of the capillary, the flow rate of the sy ringe pump can be low (as low as 1 mL/min). 57

PAGE 58

Figure 3-1. WSU electrospray io nization source. Figure 3-1 shows the WSU ESI source. The darkly shaded areas are constructed of Te flon, and the capillary is fused silica. The ferrule used is conductive stainless steel. Figure 3-2. WSU electrospray ionization source photograph. Figur e 3-2 is a photograph of the WSU ESI source. Note that the tip of th e capillary extends thr ough the screen of the IMS. This was done so that the ions woul d not be deflected backwards by the electric field. 58

PAGE 59

3.1.2 Ion Mobility Spectrometer (IMS) An ion mobility spectrometer (IMS), as briefly discussed in Chapter 1, is similar to FAIMS, in that it employs an electric field to se parate ions based on mobility. An IMS drift tube was designed and patented 37 by the Hill laboratory at WSU. This tube has a total length of 34 cm, containing a series of flat ring-shaped electr odes composed of conductiv e stainless steel with dimensions of 50 mm outer diameter 48 mm inner diameter 3 mm, separated by ceramic rings. A voltage is applied to each of these ring electrodes that generates the electric field that propels ions through the drift tube For negative ions, the first electrode carries a potential of about -10.5 kV, and this is incrementally decrea sed on successive electrodes such that the last electrode carries about -2 kV. The curtain plate of the FAIMS (s et to -1 kV) acts as the final electrode of the IMS to attract negative ions into the FAIMS cell. This IMS operates at atmospheric pressure and employs heating to help desolvate the sa mple. The heater was adjusted to be as high as possible wit hout fragmenting the sample ions (temperature ranging from about 150C to 250C). Countercurrent gas flow also aids in desolvation. The IMS contains a BradburyNielsen gate ring electrode 38 which divides the tube into a desolvation region and IMS drift re gion. A reference potential is us ed to open the gate, thus allowing ion transmission from the desolvation regi on into the drift region, or a different voltage can be applied to close the gate by creating an orthogonal el ectric field that prevents ion transmission. In these experiments, the IMS wa s first operated in total ion transmission mode, and then it was set to gate ions for separation. 3.1.3 Orthogonal Dome FAIMS Cell This instrument configuration also utilized an Ionalytics Selectra beta prototype; however, the cell geometry was not the same as that used in the first series of experiments (chapter 2). The cell used in the characterization experiments consis ted of a line-of-sight cy lindrical geometry; in 59

PAGE 60

contrast, the cell used in this chapter (figure 3.3) employs an orthogonal dome-shaped geometry. The waveform generator, application, and wavefo rm shape are identical. Two major differences between the two configurations are that the dome-shaped cell permits the distance between the electrodes to be adjusted, and pr ovides slightly higher resolution. 39 Figure 3-3. Drawing of Ionaly tics orthogonal dome FAIMS cell. Figure 3-3 contains a drawing of the orthogonal dome FAIM S cell. The red and black dots describe a simple scenario in which the black dots represent th e analyte of interest and the red dots are noise (interferent ions). Both types of ion enter the FAIMS cell through the curtain plate and are separated as the red ions collide with the wa lls of both electrodes when the proper CV is set. A percentage of the an alyte ion is also lost, but this is the only ion type that enters the mass spectrometer. 60

PAGE 61

Curtain Plate Inner Electrode Outer Electrode Plate Gap Adjuster Figure 3-4. Photograph of Ionaly tics orthogonal dome FAIMS cell. Figure 3-4 is a photograph of the disassembled orthogonal dome cell used for these experiments. The plate gap adjuster rotates to change the distance between the inner and outer electrodes. 61

PAGE 62

ESI Source Drift Tube ESI Voltage FAIMS Cell Mass Spectrometer Figure 3-5. WSU apparatus. A photograph of the entire WSU apparatus including ESI source, IMS drift tube, and FAIMS cell can be seen in Figure 3-5. The waveform generator is not connected to the FAIMS cell so that it is easily visible. 3.2 FAIMS Limitations As discussed briefly in Chapter 1, one of the primary limitations of FAIMS is fairly low resolution compared to chromatographic separa tion methods, hence its use here in conjunction with mass spectrometry as a sepa ration device rather than a sta nd-alone instrument. There are several factors that can affect resolving power of a FAIMS cell. 40 The factors that have been addressed in the following experiments include dispersion voltage, distance between the inner and outer electrodes in the domed FAIMS cell, carrier gas consistency, and addition of a desolvation region prior to the FAIMS cell. In addition, anothe r limitation is band broadening over time as buildup of solvent interferes w ith ion separation with in the FAIMS cell. 62

PAGE 63

3.3 Effects of Addition of Helium to the Carrier Gas For the entire AIMS cell was 100% ned. high vo -resolution FAIMS (HRFAIMS) High-resolution FAI on the WSU apparatus descr on set of experiments was an orthogonal dome e of two here was no gating); althou nt reaching first set of experiments (chapt er 2), the carrier gas in the F pure nitrogen. However, experimentation ha s shown that the addition of a percentage of helium can increase transmission a nd sensitivity. This effect slightly narrows CV peaks and therefore increases resolving pow er. The reason for this behavi or is not clear; although, it is suspected that because helium increases mean free path within the cell, ion dispersion and desolvation are also increased. It is also possible that the high-fi eld is effectively strengthe It is not desirable to use too high a percentage of helium carri er gas into the cell, as the ltage can induce discharge. It should be noted that because the amount of helium in the atmosphere is trace, field instruments devel oped with FAIMS technology would most likely not employ helium in the carrier gas. 3.4 High MS (HRFAIMS) experiment s were conducted ibed in this chapter. Three major differen ces between this instrument and that used for characterization in chapter 2 in terms of FAIMS resolution are: FAIMS cell geometry, additi of an IMS (set to total ion transmission mode ) in front of the FAIMS cell, and use of a percentage of helium (25%) in the nitrogen carrier gas. As discussed above, the FAIMS cell used for this cell, rather than a line-of-sight cylindr ical cell. The dome cell can provide higher resolution, due in part to the longe r path length through the cell, and due in part to the us field areas in series; there is a cy lindrical area and a spherical area. The IMS drift cell was set in full ion transmission mode (ie., t gh, the heater and electric field were still engaged. There is also gas flowing countercurrent through the IMS, which can drama tically reduce the amount of solve 63

PAGE 64

the FAIMS cell. The temperature of the heater was set to be as hi gh as possible without fragmenting the explosive molecules. The IMS had the most significant effect on overall resolution compared to the other factors. Transmission was increased, and CV peaks were narrower, demonstrating increased resolution. This is probably due to the fact that the heate desolvates the sample, and the electric field focu ses ions into a tight beam for more efficient introduction into the FAIMS cell. 3.4.1 Separation of Mixtures of E r xplosives evice may decrease total ion transmission, when the C first mixture that was examined consiste d of about 20 ppm of HMX and about 15 ppm of RD from 0Although use of FAIMS as a separation d V is set for a particular ion of interest, th e signal-to-noise ratio of that ion can increase greatly. The X. The DV was set to -2500V, which was the optimal setting when the IMS was installed. The carrier gas was made up of 75% nitrogen and 25% helium to improve ion count and resolution. Figure 3-6 is the CV spectrum for the HMX/RDX mixture. CV was scanned 20V once over a period of 8 minutes. Each peak is distinct and fully resolved. The RDX peak is significantly smaller because there was not only a slightly smaller concentration in the sample, but also the boiling point is much less than HMX. 64

PAGE 65

HMX CV 2.4V RDX CV 7.2V Solvent CV 5.0V Figure 3-6. CV spectrum of HMX/RDX mixture. 65

PAGE 66

RDX CV 7.2V HMX CV 2.4V Figure 3-7. Mass spectrum of HM X/RDX mixture. Figure 3-7 cont ains the mass spectra of the RDX/HMX mixture, showing the [M+Cl] adducts of both RDX (m/z 257) and HMX (m/z 331) taken from the relative CV peak s (which are seen in Figure 3-6). The HMX peak at m/z 341 is an [M+NO 3 ] adduct 66

PAGE 67

TNT CV 2.1V RDX CV 7.2V Solvent CV 5.0V Figure 3-8. CV spectrum of TNT/RDX mixture. A second mixtur e examined was torpex, which consists of TNT and RDX in approximate ly equal portions. The only difference between the mixture that was examined here and the military explosive is that in this case there is no aluminum powde r. In this case, one CV scan from 0-20V was done over 16 minutes. Once again, both components (see Figure 3-8) are fully resolved from each other and the solvent background in the CV spectrum. 67

PAGE 68

RDX CV 7.2V TNT CV 2.1V Figure 3-9. Mass spectrum of TNT/RDX mixture. Figure 3-9 s hows the mass spectra of the respective CV peaks of TNT and RDX in the mixture (Figure 3-8). Both components (the TNT [M]ion at m/z 2 26 and the RDX [M+Cl]ion at m/z 257) are shown. It should be noted that because the same DV was used for both of these mixtures (-2500V), the CV values of RDX and the solv ent is identical in the CV spectra of both mixtures. (Compare figs. 3-6 and 3-8) 3.4.2 Resolving Isomers The next step in separation is the resolving of two isomers of the same explosive. The explosive chosen was dinitrotoluene (DNT), because it has two isomers that are precursors to TNT: 2,4-DNT and 2,6-DNT. The analysis of th ese isomers can potentially be important to explosives forensics. Few, if any, technologies deployed today can detect and resolve these isomers from each other. Because they are isomers, they have id entical molecular weight. Both have similar mobility behavior, but due to the fact that the shapes of the molecules are slightly different, each behaves uni quely under high field. 68

PAGE 69

2,4-DNT CV 10.8V 2,6-DNT CV 12.2V Figure 3-10. CV spectrum of both isomers of DNT. A mixture was created using about 20 ppm of both isomers in equal proportion. This is a relatively small amount, hence the low ion transmission. The CV scan was conducted slowly from only 10-15V over 16 minutes. Both isomers are clearly resolved from each other, as shown in Figure 3-10. It should be noted that there was a [2,6-DNT-NO] fragment peak that appeared during the characterization (as discussed in Chapter 2, sectio n 2.6). This fragment did not appear while using the high-resoluti on setup in the 10-15V CV range. 3.5 Dispersion Voltage Another effect that the IMS added was th at the optimal DV was no longer the highest possible (-4000V); the ion signal was nearly zero above a DV of -3000V. The FAIMS carrier gas flow rate also had to be reduced to about 0.5 L/min from 4.0 L/min for optimum performance. This is in contrast to the situ ation with FAIMS by itself (as in chapter 2), where optimum sensitivity is observed at maximum DV and high carrier gas flow rate. There are several possible reasons for this unusual behavior. For example, higher flow leaving the curtain plate may cause ion scattering at the IMS/FAIM S interface. This may be overcome by closing 69

PAGE 70

the system with a nonconductive sl eeve that encloses the space between the IMS and FAIMS. This might also make directing the ion beam into the FAIMS easier. Another possible explanation is that the small difference in poten tial between the last electrode of the IMS (-2kV) and the curtain plate (-1kV) required carrier gas fl ow to be decreased to allow the potential to transport ions efficiently into the FAIMS cell. 3.6 Effects of Adjustment of Plate Gap As discussed above, the distance between the inner and outer electrodes of a FAIMS cell affects the electric field generated. Since the electrode gap of the dome cell can be adjusted (figure 3.3), data were collected to determine the optimal electrode gap for explosive detection. The cylindrical cell electrode gap is fixed at 1 mm; whereas, the ga p in the cylindrical portion of the dome cell is fixed, the gap in th e dome portion is adjustable up to 2.4 mm in increments of 0.1 mm. Four data sets were ta ken with a fixed DV at -3000V, each consisting of two CV scans from 0-20V over a period of 8 min each. The sample selected was a solution of about 15 ppm RDX, which was chos en because it has a relatively low fragmentation temperature (about 170C) compared to the other explosives The FAIMS/MS behavior of both the [M+Cl] ion and the fragment ions were analyzed. The expected relationship between resolution and electrode gap is that the lowest gap should provide the highest resolution, as the field is stronger. As expected, the ion count was the lowest, as there is a lower mean free path; mo st ions collide with the electrodes. 70

PAGE 71

RDX+Cl-CV = 16.0V RDX+ClCV = 16.0V m/z 310 CV = 11.0V m/z 310 CV = 11.0V RDX+ NO3 CV = 7.8V RDX+ NO3 -CV = 7.8V Figure 3-11. RDX CV spectrum at 0.5 mm. The first trial was at a relatively small gap of 0.5 mm (Figure 3-11). The CV was scanned from 0-20V over 8 minutes twice. On the figures, ion count is seen after NL. This figure shows three peaks. One of these (m/z 310) is probably a solvent cluster. A second (m/z 283) is the nitrate adduct of RDX. The third (m/z 257) is the RDX chloride adduct. In this case, the RDX analyte of interest is the smallest peak compared to the solvent. Probability of transmission of RDX molecules relative to solvent i ndicates that when th e electrode gap is smallest, most RDX molecules collide w ith the electrodes and are lost. The maximum ion signal is 2.46 4 counts. 71

PAGE 72

Figure 3-12. RDX mass spectra at 0.5 mm. Figure 3-12 shows the mass spectra for each of the three CV peaks at an electrode gap of 0.2 mm. Relative to the previous figure, the top mass spectrum is the right-most CV peak. The RDX mass spectrum has an additional peak at m/z 267, whic h is an adduct ion [RDX+NO 2 ] that co-elutes through the FAIMS cell with the chloride adduct. This second adduct is not unusual, though it is usually found in extremely low intensity compared to the chloride adduct. 72

PAGE 73

RDX+Cl-CV = 18.0V m/z 310 CV = 10.7V RDX+ NO3 CV = 7.3V RDX+ NO3 CV = 7.3V Figure 3-13. RDX CV spectrum at 0.75 mm. Th e second trial done was at a gap of 0.75 mm, which is 1.5 times larger. Again, the CV was scanned twice from 0V to 20V over a total period of 16 minutes. As expected the ion count is slightly higher. With this larger gap, the resolution is lower (CV peaks are wider); furthermore, the relative intensities of the different ions appear different (in partic ular, the RDX [M+Cl] ion is significantly more abundant). Th e maximum ion count is 2.92 4 counts. Due to an error, only the first 12 minutes are shown. 73

PAGE 74

Figure 3-14. RDX mass spectra at 0.5 mm. Figure 3-14 displays the mass spectra corresponding to the three CV peaks in the previous figur e. These are similar to those with the smaller electrode distance, as each ion is well resolved. 74

PAGE 75

Figure 3-15. RDX CV spectrum at 1.0 mm. Next, the electrode gap was increased by a factor of 1.3, to 1.0 mm (Figure 3-15). The ion c ount was higher, yet the transmission ion t RDX+Cl-CV = 18.5V RDX+Cl-CV = 18.5V m/z 310 CV = 8.2V m/z 310 CV = 8.2V behavior is completely different from th e smaller gaps. This CV spectrum appears much like those in previous chapters, with a well-resolved peak for the analyte and only small background. The RDX [M+Cl] ion is much more intense, the peak a m/z 310 is much less intense, and the peak at m/z 283 is hardly visible. The maximum ion intensity has increased to 6.84 4 counts. 75

PAGE 76

Figure 3-16. RDX mass spectra at 1.0 mm. The mass spectra for RDX taken at a FAIMS electrode gap of 1.0 mm can be seen in Figure 3-16. Thus far, both ion count and resolution are superior to those at smaller electrode gaps. At this gap, resolution and ion count are well-balanced. Likely, the elect ric field is at a stre ngth optimal for ion focusing. Note that the peak at m/z 267 is barely visible. 76

PAGE 77

RDX+Cl-CV = 19.6V m/z 310 CV = 10.2V RDX+ NO3 CV = 8.7V RDX+ NO3 -CV = 8.7V RDX+Cl-CV = 19.6V m/z 310 CV = 10.2V Figure 3-17. RDX CV spectrum at 2.0 mm. The last trial was done with an electrode gap of 2.0 mm (Figure 3-17). If the distance is set too high, as in this case then the electrodes can no longer carry a waveform. This ion c ount was the highest of all these trials, presenting three orders of magnitude greater transmission than the smallest setting at 0.5 mm. The RDX peak was robust and wellresolved; however, the peaks at m/z 283 and m/z 310 were again visible. 77

PAGE 78

Figure 3-18. RDX mass spectra at 2.0 mm. The mass spectra in Figure 3-18 show that the ion counts are the highest compared to the spect ra taken at other electrode distances. These four trials confirmed that a distance between the inner and outer FAIMS electrodes in the dome region set at about 1.0 mm yields the optimal transmission and resolution for explosives. It was observed that the CV for the chloride adduct of RDX increased as the electrode gap wa s increased, while the other peak CVs decreased (allowing the peaks to be resolved ), but when the gap was at 2.0mm, the m/z 283 and m/z 310 peak CVs increased. 78

PAGE 79

Table 3-1 shows the intensities of each of the major ions discussed above. As the electrode gap was increased, the intensity of the chloride adduct of RDX in creased by almost two orders of magnitude, while the m/z 310 and m/z 283 ion counts decreased until the electrode gap was set to 2.0 mm, where the trend reverses. Table 3-1. Relative Ion Intensities. Electrode gap (mm) [RDX+ NO 3 ] (counts) m/z 310 (counts) [RDX+Cl] (counts) 0.5 1.50 4 2.50 4 2.50 3 0.75 1.20 3 2.90 3 2.50 4 1.0 5.70 6.00 3 6.80 4 2.0 6.00 4 2.00 4 2.00 5 Table 3-2 lists the CV values for the same ions. The behavior mentioned above occurs for the CV values as well as the ion intensities. Th e chloride adduct of RDX increases slightly as the electrode gap increases, and the other ions d ecrease until the electrode gap is set to 2.0 mm, where the CVs increase. Table 3-2. Relative CV Values. Electrode gap (mm) [RDX+ NO 3 ] (V) m/z 310 (V) [RDX+Cl] (V) 0.5 7.8 11.0 16.0 0.75 7.3 10.7 18.0 1.0 5.9 8.2 18.5 2.0 8.7 10.2 19.6 To examine the ultimate resolution of the FAIMS system, an attempt was made to resolve [M] from [M-H] ions for individual explosive co mpounds. Unfortunately, none of the explosives have both of these i ons in approximately equal propor tion; this makes it more difficult to allow for the separation of the ions. Neve rtheless, for other types of ions, resolution is certainly feasible. To maximize resolution, the CV scan would have to be extremely slow or 79

PAGE 80

over a narrow CV range, and the ga s (and thus ions) entering the FAIMS cell would need to be completely desolvated. Different adducts of the same explosive, however, have been separated, for instance, the [M+Cl] and [M+NO 3 ] ions of RDX (see section 3.5). 3.7 Relationship of DV to CV As discussed earlier, there is no single set of conditions that optimizes separation of every mixture for every FAIMS-based instrument. In order to characterize a FAIMS-based separation instrument, CVs must be found based on field st rength, which is mainly a function of DV and cell geometry. There is, however, the same relationship between DV and CV for all FAIMS instruments, as this relationshi p is based on the change in mobility for an ion between low and high field. 12 Figure 3-19. Relationship between CV and DV. Fi gure 3-19 is a plot of the relationship between CV and DV created using data sets taken using the FAIMS/MS setup described in chapter 2. The curves indicate that the CV of each ion increases at more negative DV (hence greater field strength). TNT and ot her nitroaromatics do not produce ions that are transmitted at a positive DV, although th e chloride adduct ions of RDX and HMX respond to both positive and negative DVs. The Ionalytics instrument will not allow the DV to be set less than 2000V or grea ter than 4000V; this is indicated by the vertical red lines. Because of this DV/CV relationship, most of the explosives appear to exhibit B type behavior when high field is applied. 80

PAGE 81

3.8 Sensitivity Data sets collected on the FAIMS instrument used for characteriza tion (as described in chapter 2) show that the DV for the optimal separation (highest resolution with lowest background) for the nitroaromatic explosives is -4000V, and +4000V for the nitramines. The IMS/FAIMS instrument (as described in this ch apter) did not display the same behavior. The peak DV was at about -2500V, and at any DV se t higher than -3000V th e ion signal decreased rapidly to zero. In addition, the flow rate of th e FAIMS carrier gas for this instrument could not be set higher than 1 L/min, or ions were lost. There are several possibilities for this behavior; the desolvation and focusing attributes of the IMS most likely require that a lower amount of carrier gas and voltage are needed to achie ve the same separation results. One method to measure sensitivity and resolv ing power of a FAIMS instrument is to examine the CV spectrum for both peak width and i on intensity. As discussed earlier, there is a degree of ion loss throu gh a FAIMS (data gathered during this research shows about two orders of magnitude compared to no FAIMS cell in the system). Most FAIMS instruments (including the instrument used for chapter 2) demonstrat e an average CV peak width of about 2-3V; 40 in contrast, the best HRFAIMS system CV peaks were as narrow as about 0.75V (for examples, see figs. 3-6 and 3-8). This has bot h benefits and disadvantages; narro wer peaks require slower scan speed to define the peak, and necessitate precise CV values when fixed, but the narrower peaks are better resolved from each ot her and from background ions. 3.9 IMS/FAIMS of Explosives In another set of experiments, the IMS was ope rated in gating mode in order to attempt to add an additional dimension of separation. This involved manually setting the times at which the gate would pulse open with a delay to allow ions of a fixed drift time (and, thus, ion mobility) to pass through the drift tube. 81

PAGE 82

All other ions would be excluded from transmi ssion. Hence, a percentage of all ions were discarded by the IMS, in addition to those di scarded by the FAIMS portion of the instrument. An attempt was made to separate explosive mixtures by gating the IMS; however, ion counts were not high enough to discern i ndividual components of the mixture. This is largely due to the loss in sensitivity due to the low duty cycle of the IMS/FAIMS system. For instance, if a 500 ms gate width is used, and the ga te is only open for 25 ms, then the duty cycle is only 5%, which results in a significant loss of si gnal. Furthermore, the lack of synchronization between the IMS gate pulses and the fill times of the LCQ quadrupole ion trap may result in even greater loss in sensitivity. Gating the IMS was not critical in these studie s, as data have indi cated that individual explosive peaks as part of a mixture are fully re solved while the IMS is functioning simply as a desolvation and focusing appara tus for the FAIMS. Any futu re experiments should include efforts to increase ion signal when the IMS is gati ng. (For full discussion see Chapter 4) 82

PAGE 83

CHAPTER 4 CONCLUSIONS AND FUTURE WORK 4.1 Conclusions The United States is under cons tant threat from the use of explosives by her enemies. There is a great need for the de velopment of technologies that can eventually be applied in instruments that will be used to detect im provised explosive devices These analytical instruments mainly will consist of laboratory and intermediate-sized instruments or man-portable field instruments, and all will be utili zed both overseas a nd domestically. Instruments designed for the detection of explosive compounds can fall into two categories, depending on function and/or mission. These are either qualitative threat detectors, or quantitative thre at characterizers. 41 Factors that affect instrument use are portability, resolving power, sensitivity, and selectivity. Mass spectrometry is a widely used analytical methodology. When a separation device is coupled to a mass spectrometer, all of the analy tical figures of merit ar e drastically improved. FAIMS is a promising technology that func tions well as a separation device and is compatible with a mass spectrometer. There ar e many advantages to em ploying FAIMS as such. Even though the use of FAIMS ma y lower overall ion transmission, the signal-to-noise ratio of the overall instrument may be much higher than a lone mass spectrometer. This can be extremely important when an instrument is us ed in a field environment, away from the laboratory, when there is a gr eat deal of background matrix with which to contend. The goal of this research was to determine the degree to which a FAIMS/MS device can be employed to detect explosive compounds that are most likely to demonstrate a threat, and to attempt to ascertain methodologies to eventually do so in a field environment. The experiments discussed herein consisted primar ily of three parts. The first characterized explosive compounds 83

PAGE 84

using a FAIMS/MS, while the second included additional FAIMS/MS experiments conducted to analyze and improve the analytical factors of the instrument. Th e third section described a novel high-resolution FAIMS/MS method, employing an IMS drift tube to desolvate and focus ions into the FAIMS cell. The HRFA IMS method is sensitive, and th e resolution is high enough that valuable forensic data can be collected. 4.1.1 Characterization of Explosive Compounds A list of explosive compounds was compiled based on both the frequency of encounter during terrorist events and the presence in common military explosives. The feasibility of detection of these compounds was first evaluate d; once accomplished, the characteristics of each compound were analyzed and noted. All of the explosives in th is study contained nitro leaving groups. This list includes three nitroaromatics: TNT, which is one of the oldest known and commonly encountered explosives, and two is omers of DNT that are precursors to the manufacture of TNT, and, theref ore, are forensically important TNT is also found in many military explosive mixtures. All of the nitroaromatics were easily detectable and were readily characterized using the FAIM S/MS instrument. There were two nitramines evaluated, RDX and HMX, which are encountered along with TNT in military explosives and plastic explosives mixtures. The most common explosives include mixtures of TNT and RDX. Both R DX and HMX are similar molecules that have similar characteristics. The formation of a mol ecular-type ion requires adduct formation to be well detected. The best adduct for this purpose is formed with the addition of a chloride (Cl ) ion (improved by doping the solvent with a small am ount of a source of chloride, such as CCl 4 ), though all will form adducts with nitrate (NO 3 ) as well. There were two nitrate esters evaluated, NG a nd PETN. Both are extremely unstable. The molecular ion cannot be detected by MS; PETN w ill only form adducts in low concentrations. 84

PAGE 85

NG will not form adducts at all. Because of th is instability, though, NG and PETN are rarely, if ever, used without another explosive in a mixture. The last explosive evaluated in this study was ANFO, a mixt ure of ammonium nitrate and a fuel, which is a commonly used commercial explos ive. It is also favor ed by terrorists due to the ease of its manufacture and acquisition of its components. The nature of these components creates difficulties for detecti on, as ammonium nitrate and many of the fuels used (usually diesel) may be found in high levels in the b ackground. Because there is no single structure, ANFO has not been well characterized for detection by MS. 4.1.2 FAIMS Characteristics FAIMS separates ions using a high electric field to affect the change in ion mobility. One of most important FAIMS characteristics that aff ect the strength of this field is the dispersion voltage. Once the optimum DV was determined, th en other analytical parameters could be ascertained for each explosive compound. No two types of FAIMS instruments can be characterized identically, though ther e is a basic relationship that applies in all cases between DV and CV. Other FAIMS factors that were examined were the addition of a small amount of helium to the carrier gas, which improves resolution and ion transmission, and the distance between inner and outer electrodes of a FAIMS cell. The optimal electrode gap for the de tection of explosives on the instrument used corresponds to other FAIM S studies that have been done (a distance of 1mm). 4.1.3 IMS/FAIMS/MS During these experiments, an IMS was added to the instrument to add another dimension of separation. When the IMS was gated, only ions with a specific ion mobility were transmitted on to the FAIMS cell; therefore, only a fraction of the ions were transmitted. The ion counts 85

PAGE 86

during this experiment were not high enough to di stinguish individual com ponents of a mixture. Changes could be made to the instrument to attempt to improve transmission and sensitivity. This is discussed further in the reco mmendations section of this chapter. 4.1.4 High-resolution FAIMS The addition of the IMS drift cell was proven to be beneficial to the instrument, however. When the IMS was in total ion mode, there wa s no ion gating, and all ions passed through the drift tube. Because the tube is heated and carri es an electric field, the analyte was desolvated, and the ion beam was focused into the FAIMS. When this method was used along with the optimal FAIMS parameters, resolving power of the FAIMS, and ion transmission increased drastically. At certain junctures, the CV peaks were so narrow that the CV scan speed had to be slowed to allow the entire peak to pass through before the CV changed. 4.1.5 Mixture Separations with FAIMS/MS Once characterization was completed, then the value of the FAIMS to separate mixtures was analyzed. If an analytical instrument is una ble to resolve mixtures, then its value as a field instrument is limited. This phase of experi ments was greatly enhanced using the HRFAIMS method. First, two mixtures containing different explosives were separate d successfully and the components were well-resolved. Resolution wa s high enough that mixtures containing two isomers of the same compound (DNT) were also su ccessfully separated from each other. This ability can be valuable for forensics analysis of either post-blast or manufacturing of TNT. 4.1.6 Disadvantages of FAIMS FAIMS instruments still have limitations, especial ly pertaining to field use. The resolution and transmission of the FAIMS cell are strongly a ffected by water and othe r solvent, and the cell must be as dry as possible. After continued use, if the FAIMS cell is not periodically dried out, the sensitivity and resolving pow er drop as either the electrode s are hindered from carrying a 86

PAGE 87

high electric field, or the saturation of solven t causes ion-molecule reactions. As such, the carrier gas must not contain impurities; furthermor e, carrier gas is required in large volumes (between 3-5 L/min). This can pr ovide issues in terms of instrume nt period of use, maintenance, and design, for it is undesirable to add large cylinders of gases to an apparatus if the objective is portability. The Ionalytics prototype instruments are extr emely delicate, and are not compatible with field use. For instance, the DV is transporte d from the waveform generator to the FAIMS cell via an insulated cable. Any e nvironmental changes in the area around the cable can significantly affect the waveform and field generated. 4.2 Recommendations Analysis of data sets collect ed during this series of expe riments has determined that FAIMS/MS is a feasible method to use for the detection of explosive compounds. A highresolution methodology for the detection of expl osives with FAIMS/MS has been identified during this research study, and should be further explored. 4.2.1 Further HRFAIMS Improvements The use of the IMS as an ion focusing/desolvation device could be further improved, because its use during this study was not intended to be so. The interf ace between the IMS and FAIMS consisted of an empty space between the tw o components. The IMS drift tube must be lined up with the FAIMS cell manually, and both th e electric fields as well as carrier gas may cause ions to leave the system. A nonconductive sleeve could be us ed to enclose the system, and may decrease ion loss, and may improve tr ansfer into the FAIMS cell. 4.2.2 FAIMS Carrier Gas Experiments Further, experimentation could be conducted into the composition of the FAIMS carrier gas. Future field instruments will probably not include carrier gas cylinders. Other gas mixtures 87

PAGE 88

could be tested, and the effects determined. A desirable approach w ould be, first, to use laboratory-provided pure, dry zero ai r, and if this is effective as a FAIMS carrier gas, then to use ambient air run through a sc rubber/dessicant just prior to introduction to the FAIMS cell. 4.2.3 Practical Separations and Analysis Because the separation of mixtures was done su ccessfully, the next logical step is to separate unknown mixtures by, for example, havi ng another individual prep are the samples. Also, efforts should be made to separate explosive compounds from various background matrices, thus determining the lowest possible concentrations at which explosives can be detected. After this is done, then a study shoul d be done to identify matrices (if any) that strongly interfere with detection, and attempt to ameliorate them. 4.2.4 IMS Gating The IMS/FAIMS/MS method requires more efforts to be made in terms of synchronizing times when the gates open to the FAIMS CV s canning, and the injection times of the quadrupole ion trap. This will further ensure that the lowest number of ions possible is lost. 4.2.5 CV Scanning A practical instrument would require scan speeds that are as fast as possible, so that possible threats can be detected and warnings can be administered. CV scan speed can be adjusted to any length of time, though the higher resolution the in strument, the slower the scan speed that is necessary. For practical use, an optimal length of time should be set such that mixtures are fully resolved, but the instrument can scan as quickly as possible. The CV scan range can also be adjusted so that all of the expl osives of interest are covered, yet it is set to the minimum amount of time needed. If a qualitative th reat warning is all that is required, then the scan time can be minimized to allow just enough of the analyte ions to pass through the system to be detectable above the noise level, thus further reducing the time required for a scan. 88

PAGE 89

4.2.6 Instrument Maintenance Both IMS/FAIMS instruments used in this research introduced samples into the system as liquids. As such, the sensitivity and resolution of the FAIMS degraded over time, and regular cleaning of the cell with a volat ile solvent that did not leave a residue, such as methanol, followed by drying of the cell were needed. Sa mpling explosives in a vapor or particulate sample, rather than in liquid solution, may reduc e this problem. Practical instruments should include a small amount of solvent for regular cleaning; and note th at carrier gas should always be run through the system when in standby. 4.3 Future Work Now that feasibility has been established, a nd the first step toward method development for explosives detection via FAIMS/MS has been taken, the next step includes working toward developing an instrument employing this method. A great deal of effort is required, however, to design and create an apparatus that can be employe d for use in a field environment. The best way to approach this would be to first design and test an intermediate instrument that would be used specifically for explosives and other rela ted compounds. This work would likely include further research in interf aces between each component of the instrument, FAIMS cell geometries, and other ways to allow the instrument to serve its purpose. Also, a method of practical sample introduction to the system is needed, as explosives have low vapor pressures and are not conducive to collec tion via gas collection. 4.3.1 Ionization Sources One possible way to move toward development of a field instrument was actually attempted during this research study: the use of a more portable ionization source. ESI is designed for ionizing liquid solutions, and APCI re quires a heater and cor ona discharge; both of these limit the capabilities of an instrument. 89

PAGE 90

A novel type of ionization source was evaluate d during this research for use with an IMS/MS for the detection of explosives. Developed by Blanchard & Co., the distributed plasma ion source (DPIS) is designed to provide ionization at atmosphe ric pressure using a simple, portable, low-power device. The DPIS is c onstructed with two overl apping electrodes of different sizes that are separate d with a dielectric. A potenti al applied across the electrodes creates a distributed plasma around the edge of the small electrode, which ionizes the surrounding air and sample vapor. This type of source can be useful because sample can be introduced as a vapor. In this case, liquid sample was kept in a bottle, across which a nitrogen carrier gas was blown, and the headspace was direct ed over the DPIS source, into one of WSUs homebuilt IMS tubes. Figure 4-1. Distributed plasma ionization source. Figure 4-1 is a photograph of the DPIS extended through the ion screen of the IMS drift tube. The plasma discharge is a distinct blue color. The sample line is taped to the insulated wire that powers the source. The DPIS circuit has limitations, however. Th ere is no diode to prevent overload of the circuit, and the maximum potential cannot exceed -1kV. Data were collected with several explosives. The DPIS source did ionize the sample s, but the signal-to-noise ratio was extremely 90

PAGE 91

low. There may be many reasons for this. First, the source voltage is low compared to the high voltage of the IMS, which may lim it transmission of ions into th e IMS drift tube. Also, the only sample directed into the instrument is headspace, which contains significantly less analyte than a liquid which is vaporized. This source is not compatible with laboratoryscale IMS/MS instruments, as the ionization voltage is not scaled with the other voltages of the system. Neve rtheless, the DPIS did ionize the explosive samples, and can potentially be used with low-power, portable field instruments. 4.3.2 Development of Field Instruments The first step in this research has been to determine the feasibility of the IMS/FAIMS/MS method for detecting explosives. Now that th is has been accomplished, improvements can be made to the overall method (as discussed above). Once this is done, then work can proceed to reduce size and power consumption of each com ponent, and then durable packaging can be developed for a fieldable instrument. 91

PAGE 92

REFERENCES 1. Country Reports on Terrorism US State Department, 2005. 2. Opportunities to Improve Airport Passe nger Screening with Mass Spectrometry Committee on Assessment of Security T echnologies for Transportation, National Materials Advisory Board, 2004. 3. Shea, Dana A.; Morgan, Daniel, Detection of Explosives on Airline Passengers: Recommendation of the 9/11 Commission and Related Issues Congressional Research Service, Library of Congress, 2006. 4. Eiceman G. A.; Borsdorf H., Ion Mobility Spectrometry: Principles and Applications Applied Spectrometry Reviews, 2006, vol. 41, no. 5, pp. 323-375. 5. Hill, Herbert H.; Simpson, Greg. Ca pabilities and Limitations of Ion Mobility Spectrometry for Field Screening Applications Field Analytical Chemistry, 1997, vol.1, no. 3, pp. 119-134. 6. Lee, Harold G.; Lee, Edgar D.; Lee, Milton L., Atmospheric Pressure Ionization Timeof-flight Mass Spectrometer for Real -time Explosives Vapor Detection Proceedings of the First International Symposium on Explosive Detection Technology, pp. 619-633, 1992. 7. Osorio, Celia; Gomez, Lewis M.; Hern andez, Samuel P.; Castro, Miguel E., Time-offlight Mass Spectroscopy Measurements of TNT and RDX on Soil Surfaces Detection and Remediation Technologies for Mines and Minelike Targets, 2005, v ol. 5794, pp. 803811. 8. Bottrill, A.R., High-energy Collision-induced Dissociation of Macromolecules using Tandem Double-focusing/Timeof-flight Mass Spectrometry PhD thesis, University of Warwick, Coventry, UK., 2000. 9. Gruber, Ludwig. Characteristics of Different Mass Analyzers, Fraunhofer-Institute for Process Engineering and Packaging http://www.ivv.fraunhofer.de/ms/msanalyzers.html, 2000. 10. Purves, Randy W.; Guevremont, Roger; Day, Stephen; Pipich, Charles W.; Matyjaszczyk, Matthew S., Mass Spectrometric Characterization of a High-field Asymmetric Waveform Ion Mobility Spectrometer Review of Scientific Instruments, 1998, v ol. 69, no. 12, pp. 4094-4106. 11. Kolakowski, Beata M.; McCooeye, Ma rgaret A.; Mester, Zoltan; D'Agostino, Paul A., Rapid Separation and Identification of Ch emical and Biological Warfare Agents in Food and Consumer Matrices Using FAIMS-MS Technology National Research Council of Canada, 2006. 92

PAGE 93

12. Guevremont, Roger, High-Field Asymmetric Waveform Ion Mobility Spectrometry (FAIMS), Ionalytics Corp., Ottawa Canada, 2004. 13. Reich, Richard, Fundamentals and Applications of Atmospheric Pressure Chemical Ionization Quadrupole Ion Trap Mass Spectrome try for the Analysis of Explosives, MS thesis, University of Florida, 2001. 14. Finnigan LCQ Series APPI/APCI Co mbination Probe Operators Manual Rev. A, Thermo-Electron Corp., 2004. 15. Buryakov, I.; Krylov, E.; Nazarov, E.; Rasulev, U., A New Method of Separation of Multi-atomic Ions by Mobility at Atmo spheric Pressure Using a High-Frequency Amplitude-asymmetric Strong Electric Field International Journa l of Mass Spectrometry, 1993, vol. 128, pp. 143-148. 16. Kanu, Abu B.; Hill, Herbert H., Ion Mobility Spectrometr y: Recent Developments and Novel Applications, Washington State Universit y, LabPlus International, 2004. 17. Atkins, Peter; dePaula, Julio, Physical Chemistry, 7 th Ed., Chap. 24, Oxford Univ. Press, Oxford, UK, 2001. 18. Welling, M.; Schuessler, H.A.; Thompson, R.I.; Walther, H.; Ion/Molecule Reactions, Mass Spectrometry and Optical Spectroscopy in a Linear Ion Trap, International Journal of Mass Spectrome try and Ion Processes, 1998, vol. 172, no. 1, pp. 95-114. 19. De Hoffman, Edmond and Stroobant, Vincent, Mass Spectrometry Principles and Applications 2 nd Ed., 2005 20. Paul, W., Electromagnetic Traps for Charged and Neutral Particles Modern Physics, 1992, vol. 62, no. 3, pp. 531-540. 21. Finnigan LCQ Series Hardware Manual Rev. A, Thermo-Electron Corp., 2003. 22. Mathieu, E., Mmoire sur Le Mouvement Vibrat oire dune Membrane de forme Elliptique Journal of Pure Mathematics, 1868, vol. 13, pp. 137-203. 23. McLachlan, N. W., Theory and Application of Mathieu Functions Dover, 1947. 24. March, Raymond E.; Todd, John F. J., Quadrupole Ion Trap Mass Spectrometry Wiley, 2005. 25. Murray, K. K.; Boyd, Robert K.; et. al., Standard Definitions of Terms Related to Mass Spectrometry, IUPAC, 2006. 26 Sparkman, O. D.; Klink, F. E., LC/MS: Fundamentals and Applications ACS, 2001. 93

PAGE 94

27. Annual Threat Assessment of the Director of National Intelligen ce for the Senate Armed Services Committee Statement by the Director of National Intelligence, John D. Negroponte to the Senate Armed Services Committee, transcript available on www.globalsecurity.org, Feb. 2006. 28. Davis, Tenney L., The Chemistry of Powder and Explosives Ch. 1, Angriff Press, Hollywood, CA, 1972. 29. Wade, L. G., Organic Chemistry, 6 th Ed., Ch. 15-16, Prentice Hall, 2005. 30. Hannum, David; et. al., Miniaturized Explosives Preconcentrators for Use in Manportable Explosives Detection Systems Sandia National Laboratory, 1999. 31. Committee on the Review of Existing a nd Potential Standoff Explosives Detection Techniques, Existing and Potential Standoff Ex plosives Detection Techniques National Research Council, National Academies Press, 2004. 32. Yinon, Jehuda, Toxicity and Metabolism of Explosives CRC Press, Boca Raton, FL, 1990. 33. Cooks, R. Graham; et. al., Direct, Trace Level Detecti on of Explosives on Ambient Surfaces by Desorption Electrosp ray Ionization Mass Spectrometry, Royal Society of Chemistry, 2005, vol. 15, pp. 1950-1952. 34. Federal Facilities Assessment Branch, Public Health Assessment, US Army Umatilla Depot Activity, Centers for Disease Control and Prevention, 1997. 35. Cooper, Paul W., Explosives Engineering, Ch. 3, Wiley, New York, NY, 1996. 36. Fenn, John B., Electrospray Ionization, Principles and Practices Mass Spectrometry Reviews, 1990, vol. 9, no. 1, pp. 37-70. 37. United States patent 7071465: Ion mob ility spectrometry method and apparatus, inventors: Hill, Herbert H. et. al., Washington State University research foundation, 2006. 38. Bradbury, Norris E., and Nielsen, Russell A., Absolute Values of the Electron Mobility in Hydrogen Physical Review, 1935, vol. 49, no. 5, pp. 388-393. 39. Guevremont, R.; Thekkadath, G.; Hilton C. K., Compensation Voltage (CV) Peak Shapes Using a Domed FAIMS with the Inne r Electrode Translated to Various Longitudinal Positions Journal of the American So ciety of Mass Spectrometry, 2005, vol. 16, no. 6, pp. 948-956. 40. Schvartsburg, A.; et. al., High-Resolution FAIMS Using New Planar Geometry Analyzers, Analytical Chemistry, 2006, vol. 78, no. 11, pp. 3706-3714. 94

PAGE 95

41. Transportation Security: Systematic Planning Needed to Optimize Resources statement of Cathleen A. Berrick to the US Senate Committee on Commerce, Science, and Transportation, USGAO, 2005. 95

PAGE 96

BIOGRAPHICAL SKETCH Jared J. Boock was born in Hazleton, PA in 1980. He grew up in Torrington, CT, and Panama City, FL. He completed his B.S. in chemistry from Florida State University in 2002 and received a commission in the US Air Force. His first duty station was the Molecula r Sciences Division, Materials Technology Directorate, Air Force Technica l Applications Center, Patric k AFB, FL, where he served as the division Test Director. His primary duty was to plan and execute field tests, and evaluate data for the test a nd evaluation of chemical weapons detection equipment. He wrote several classified test and technical reports. In 2005, he was selected for the Air Fo rce Institute of Technology Civilian Institutions program, to study fo r his M.S. degree. He chose to pursue his M.S. degree in analytical chemistry at the Un iversity of Florida. Upon co mpletion of his M.S. degree he will be stationed at the High Explosives Re search and Development Facility at Eglin AFB, FL. 96


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101112_AAAACG INGEST_TIME 2010-11-12T13:18:50Z PACKAGE UFE0019340_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 14533 DFID F20101112_AABJSF ORIGIN DEPOSITOR PATH boock_j_Page_96.QC.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
2a15184cc3647643e58b683fdf9362f3
SHA-1
9b77cb9b14188626c91d948a2be24e48379c4c6d
6254 F20101112_AABJRQ boock_j_Page_83thm.jpg
32e32080666c727aad196a4825fd271c
056afba6d5a8d3a9e31e65b7d8ae0d5e737b462c
6733 F20101112_AABJRR boock_j_Page_84thm.jpg
f6ec4c186f78672ce0b1ef104932b8a7
f0672cf85d0cc0c5db5cdc5bcbf0089803003495
6270 F20101112_AABJRS boock_j_Page_85thm.jpg
992f957386bc7f974266555a93282aa2
5ac5e0c8e46e0a30d666bd4cbdb121b5c5dcf189
29110 F20101112_AABJRT boock_j_Page_86.QC.jpg
2c321e8487d50d17b57e232f91356e92
3cd639b49f2611dc6225f2954ae6a6312d610319
25809 F20101112_AABJRU boock_j_Page_87.QC.jpg
24af13ca00b7088deceff50c8c984198
7d6d788546223456b223ef5d4854211a045061da
6553 F20101112_AABJRV boock_j_Page_87thm.jpg
41610d5fd7b84d79ce8a324236f8014c
179ca40e9b8b1b1bfbf0e55929a3d99eee44718f
26128 F20101112_AABJRW boock_j_Page_89.QC.jpg
c2bd7ee895223d9c4c7e975aab4eb691
1fc4b5524d408c18f04b43ee2f2cb1311dc572fe
6074 F20101112_AABJRX boock_j_Page_89thm.jpg
4af478dd8e6167041e72c788f318e45f
c3cb662ec87fc28422701d3c7d19b572482595c6
26541 F20101112_AABJRY boock_j_Page_90.QC.jpg
389bd5ec618ef6ebeaadf543e81ece5b
9b99eadfe972fbda969823f6f13519dcdc3e7a07
6792 F20101112_AABJRZ boock_j_Page_90thm.jpg
7f6a4f9c80062456931b170801adff13
9e0533367b83179a7166f2000fcbc8409a846613
13939 F20101112_AABISF boock_j_Page_75.QC.jpg
f2d4a50bf9e23acbfa816c015fdbd61f
f7e263307c87d94ea17bd3000ebaf9a50b6f7b44
4257 F20101112_AABISG boock_j_Page_32thm.jpg
b5f4deb7dcd70a9b84a907148766ca9c
dc52d61d53642c94d8b66332c5f762db686afb04
109510 F20101112_AABISH boock_j_Page_21.jp2
fa5e05c4fe38d9d0efdb20241f943510
30f1b1efce787c82deb80e7dd3da7b364628579d
25271604 F20101112_AABISI boock_j_Page_47.tif
f5dd8f879134dd339eba9f8250a6b725
ea9c3bcd46e18c58c6d1aca78604aa3694df575c
5452 F20101112_AABISJ boock_j_Page_25thm.jpg
44004448d8036074c5b7a250b306c1b9
06c115e6ce1b516cf85a4246a6d9f5a1f2b72f8d
F20101112_AABISK boock_j_Page_69.tif
b51c71358ba3285c71babd16d61b28b7
6619e216eef1138b2da60d37bbc83ae7310dc790
2003 F20101112_AABISL boock_j_Page_89.txt
e88b682622bd2e773675e0f725e70a29
510389566f9b6a4174499a9f644bb82061c412b2
1068 F20101112_AABITA boock_j_Page_52.txt
b6d9e7249cc156818b8d2e98080ff838
fa47d1794276683d8741b9ff074e2991925696d2
3140 F20101112_AABISM boock_j_Page_05.txt
1768461d411129ca7c6ffa7372906cc5
b24fd41af4f2e33f6fa64e10f56ad4da858ef961
2070 F20101112_AABITB boock_j_Page_21.txt
10815b71a28af4481dc43394442880d5
fad396aeef52527a1d01e0d8a6c3950efe676ac3
38219 F20101112_AABISN boock_j_Page_54.pro
b0abe0b850f3cd9865d2cea89396dfc7
8eb9794e60b11c1bd91353cee0c5fea6ec74b9a7
475124 F20101112_AABITC boock_j_Page_34.jp2
1c0b068c81fef1adc15fce97a5a2c15d
d919edce3ba7f224b3c72f4e4c1c867dbadffbf4
428553 F20101112_AABISO boock_j_Page_33.jp2
919ad10a8ddd39326a799a5dc2691510
639c22cbd7dd34780b1d1d59be3879c8afe7db44
4039 F20101112_AABITD boock_j_Page_56thm.jpg
791a5d8ca5b0acd8228e6f4a1532e3d0
22a0380ffc7e5e2e90716660e525e2f2fda85f9c
28772 F20101112_AABISP boock_j_Page_59.QC.jpg
f9cf93028db7f7e05cd0b6f46e586b73
82890dfe5d86fa8a10f3f8f11540e968a23d2325
28880 F20101112_AABITE boock_j_Page_42.jpg
5e61f9e6321c74d0a48a2ad0c7ca8e7e
bc283e6b42539c2dc24ddc13d882cd7cd1be3254
88611 F20101112_AABITF boock_j_Page_13.jpg
7e30a6d3e3013fc6b0546cfd63a6f9f7
ab332cd1c1cc63088d97cd8ad073de72c5ec004b
1088 F20101112_AABISQ boock_j_Page_96.txt
f6654ab36bf1c9751e1d48bfa58f3733
31e80cdbae7f7d19f89c844c4a9b49e7a5369ac7
107251 F20101112_AABITG boock_j_Page_89.jp2
c520f2743814347f16d651cadbe4cfa3
2d907d8624d70b1149d54cf52851a3ee55c9f8cf
20588 F20101112_AABISR boock_j_Page_54.QC.jpg
8c85c8b2fc825220fa8f542260c12721
eac86be2eb966b5b4fa4ba45c5ab2ed7882871fc
215 F20101112_AABITH boock_j_Page_95.txt
a8a98ff06f289746ed4324a52368d5f9
994bfce7bb864f004a9fca078a4f0451eecbc52b
3162 F20101112_AABISS boock_j_Page_48thm.jpg
ef86d1588cdf9248e4f908e1ef2416fa
6c08d121429b28dfa98a356a61e94aa6955e2662
1053954 F20101112_AABITI boock_j_Page_89.tif
11f938c80de4b6c64b4b00c20d07b619
d55920cf71cbd5542308d5c48ca99fe105487912
1686 F20101112_AABIST boock_j_Page_79.txt
9b2fe2be3ec6ad60689b6eae05ad4054
6a7154acc22bcd4a4c3ec664713aade031ff9459
7751 F20101112_AABITJ boock_j_Page_07.pro
24023eb57a23491420b5481890ca3115
2f153751cd3a6f8ed5567f1c8a64472f73ad38e0
3556 F20101112_AABISU boock_j_Page_02.jpg
b47490ab3ea7fd489cd6653ea6d13246
d363fbeef111f096a89d41062903edd3ddde95d2
16429 F20101112_AABITK boock_j_Page_46.QC.jpg
aa9407792904d10fb34aa69da6689312
36e253d78a53b10c84f5e8ed4eab7cf67abe984c
74128 F20101112_AABISV boock_j_Page_79.jpg
508261003cfcfbe815b3ee4e06746b80
904d628bbbb8679209fcde5f5edaffc1dec71a2c
1051901 F20101112_AABITL boock_j_Page_90.jp2
c447fdb8483d397de615189d955b1ad9
5730cc652fe65ed6997c85e89b7b74278a32043e
252529 F20101112_AABISW boock_j_Page_07.jp2
30f8293adb8fef4362056c686182fbb4
daa8f17063cb48f8ac3d042e51c27223d9930f1b
F20101112_AABIUA boock_j_Page_42.tif
0193419d34fe7d0486d466ca04d49a27
2ee9efae1affec7f9de79ad3a211ac992c2d75e2
F20101112_AABITM boock_j_Page_01.tif
63a928fd20171f49502885c1e9d7e781
39c1b45f749a535316cd67a358029cfbf0aef9a2
51157 F20101112_AABISX boock_j_Page_21.pro
a1e85554751c96b41aa23999808ba677
4ce8267d7b842b00235d98561699ef1ac1d4fe26
2992 F20101112_AABIUB boock_j_Page_72thm.jpg
b2ff93e613d0183e5c63557e3d4a9a78
2e43a512177f2fb8c08e103c13c505e081be685f
38289 F20101112_AABITN boock_j_Page_43.pro
a24444dd86adf04730c59bfe3fab17ca
357ecc4c025d73eb3f02f6c960f2a7fce8871bfe
6368 F20101112_AABISY boock_j_Page_81thm.jpg
1bcac070ebd3c18260134cd2267cbcc8
987ab5640b536cbe2711f29fa2c1fb159167ac3a
67114 F20101112_AABIUC boock_j_Page_57.jpg
4a6558ecf2ccf1c6ce81da760d72de0b
ada6ca5a6bbde97eb6bfcb9607817f954b631bd8
49101 F20101112_AABITO boock_j_Page_52.jpg
3d12f7cc64af802e3b81aab620369e29
a775b085f6891f1cd693bf4a0142b22c2c1d0fc4
F20101112_AABISZ boock_j_Page_08.tif
3662f026d3c052279d8b1f92de5a19b1
c0d1c760ae4b56d73efcd57a0e53be9eb17d9f7f
1261 F20101112_AABIUD boock_j_Page_60.txt
b3b7f4f156c8ba54d8e89827f3e2e34f
c32d66e30cfdb7f9719d84691fdd290efac09fe1
2333 F20101112_AABITP boock_j_Page_92.txt
828daf0685ebcfd42f7de5541ad4e6ed
32f37110afd628165f16b59155549a1b2ec24cc3
87040 F20101112_AABIUE boock_j_Page_63.jpg
53cd0849e5faaa0fa391fb76144c93af
d72d0c82d1c8b945866e4a146403c329c275ac2b
91187 F20101112_AABITQ boock_j_Page_16.jpg
01b213736e824014248a667bd901e0bd
6d4dea7a5b5d188d7b63b70bd5081f7721edaf99
13465 F20101112_AABIUF boock_j_Page_35.QC.jpg
07bbe0135a37d5c1686f645267c9b9a5
3f436a9c2dda999ee2a1e5f13d08c091a15b8583
F20101112_AABIUG boock_j_Page_26.tif
f172ebe652425d06a8661af82578f9bf
66537d7d6cae1b43d626391fa48c6af84cb671c7
371323 F20101112_AABITR boock_j_Page_48.jp2
c2e07ff89dc39986439429e7fa52dbe3
334f1dc9657f65c4be15561d313fa14886406b4c
71456 F20101112_AABJAA boock_j_Page_70.jpg
df192a59f18bea737f3c9bc5f4406de7
183fe6b1cd7609b8619eb3fbf85e90fef9fcea76
F20101112_AABIUH boock_j_Page_71.tif
0a6e498b9a014ef37c4e51658f2679c8
7c653935e6bddce0d170811ef6480d7dc0f97412
3154 F20101112_AABITS boock_j_Page_39thm.jpg
5f8463c4e4da9f65145749ae477cf1c3
ae102bdeb55016d66894278b7c8ab2c9809ac5f7
57482 F20101112_AABJAB boock_j_Page_04.jp2
c302fd726fa882d534008d6b8aac45f5
2a543bb0e19e0c2d786ebb53343300569b88e39e
134986 F20101112_AABIUI boock_j_Page_92.jp2
5dd7912e2fbe329b5375ee4d4af3694c
3df77e8295f4cf14e7d94bcdfce3972bf093cf75
4021 F20101112_AABITT boock_j_Page_44thm.jpg
d4282ea44f4714dc66361322c7ead33d
c025224e596c36fd514e2d20390e8ae07aa55768
115510 F20101112_AABJAC boock_j_Page_53.jp2
4f6eb889e14ac97884ca34cd94f8fdae
020f115120bad21d947f41f9e07d26ccf00b453b
15605 F20101112_AABIUJ boock_j_Page_74.pro
068f1f4c86a8c7588a48c988f65234d5
ecb7f72e6969d5fd728a66c2160d14833c790827
9553 F20101112_AABITU boock_j_Page_01.pro
88131c501a3cb06189cb3c2fd03c1ead
afe445526bb09fe828c5d7cd23982329f9e27267
49967 F20101112_AABJAD boock_j_Page_30.pro
64dd9af0d20bbb271cb83582cefe7f63
45c2406a1b0129b431c178688e2b4d0ffdbe50d5
105134 F20101112_AABIUK boock_j_Page_22.jp2
42754f1974627df009bd133770428ba4
9a45d872ef5313f4e933f6ecd48df0a5ead788be
F20101112_AABITV boock_j_Page_40.tif
1dae40b821f24950a173766afd6cddf2
d603b76343426f0478d2b7fa63b4dce09601c271
855618 F20101112_AABJAE boock_j_Page_60.jp2
5263160a7b80bb0597a9a60df78003d8
028dd0180baf1caefb10635c1d84a6e41288410b
6931 F20101112_AABIUL boock_j_Page_28thm.jpg
7d64e54411f9c115426ce2fd6e33d06a
72ab04e27c1a3e9eccdf163a15874532c947cd93
689386 F20101112_AABITW boock_j_Page_32.jp2
1a51620366d9705c8c56cf67320bc46c
dbea9e8d2065a2d9b2cca7e9197c0d15bbe8a14d
F20101112_AABJAF boock_j_Page_51.tif
f11b98706407db0a7f1d0f1da93126b2
4c5d3866f0ac09bde724bf65450f257c238af516
2043 F20101112_AABIUM boock_j_Page_29.txt
3d73f7a5492c48c299e344aead52cf85
fd74c7f3bd1ad4d19fe15da32709b4b270f75a37
3783 F20101112_AABITX boock_j_Page_35thm.jpg
52fdb7ffc51687a76f9ef69034e8ed04
cacfc4b51755c98af00bffbd9951090f1516c753
1184183 F20101112_AABJAG boock_j.pdf
63a901fcd7af442929c2c566d3be7f56
3e81f639faae880faf4f0af22eb972f76506754b
38467 F20101112_AABIVA boock_j_Page_15.pro
efe116ec0929240454abb7e7c4f0b7c1
77c5e89d00af7705c27a92146beb6d4d62a17a29
3115 F20101112_AABIUN boock_j_Page_11thm.jpg
0fb2b8b7f04e24338e0cfb2eeb79f637
efcdd5affcc07f27918d350c35372251c7cad3e5
932 F20101112_AABITY boock_j_Page_03.pro
47e52bedbcbd330916c4b1e2c08331ca
da9b6cfa98718a8d3d75be55a04798145a354997
1199 F20101112_AABJAH boock_j_Page_82.txt
bbbb34a0bef488b190b7c84d656e9c54
b777724ac9c554c993c612256e1c9832bb942897
6529 F20101112_AABIVB boock_j_Page_14thm.jpg
040579d1065844921642f801e9d01321
cf7e1384138b76c1f9443f01dc28a816519a1d0e
718858 F20101112_AABIUO boock_j_Page_71.jp2
0075a5eb59ffb753b4efb3beef2f8bbd
fe04d7ab00c30234a88bf0072c03ae9f0e97acde
96135 F20101112_AABITZ boock_j_Page_05.jpg
126a3e350c75978cad5de2971d17069a
73f90fd7fb37631401fb8156e88f69e0ad001d77
F20101112_AABJAI boock_j_Page_34.tif
4d0633518c567e73a1cef5d252bd07e4
dec0e05ca2e8ab53ec7a86e24fa83d5a5fc18572
49136 F20101112_AABIVC boock_j_Page_77.jpg
d9d861241050ea508d91eb5feb366fdd
e5df09d7713ca0e2bd357109f6ee895f524fd237
F20101112_AABIUP boock_j_Page_02.tif
b490d84f7985f8e5a6ce7552072bb0b2
cce3ca4d2492eea747f743b4e5814bdc27381c95
301975 F20101112_AABJAJ boock_j_Page_40.jp2
4b68d4001a3be0c001aa891746993b64
b8d2a2e3ee1a740a10a5a955009c90114d33b706
32287 F20101112_AABIVD boock_j_Page_74.jpg
fc0e7793d260bae4519e2c00405a1851
e2f753dc533024ec602bb43e0b0e4d632a6b2ac3
42527 F20101112_AABIUQ boock_j_Page_27.pro
f68e665f1bd5ab36b1a8108bf5b86d98
5ea78bfdd6be6c4fe96809467db8e321939b5f95
12694 F20101112_AABJAK boock_j_Page_48.pro
7489d93d656743d8b3c5fb7c7116012f
af042978217dcbc5ac3d0fdb68f91dbab886446a
6594 F20101112_AABIVE boock_j_Page_93thm.jpg
72d3de0fd10081d7cf00b7afb043e348
d7a0543a9dc00c1f7452df56c87f925900e2a72d
38081 F20101112_AABIUR boock_j_Page_57.pro
4981c1281641174cd90f6a8c51b40b2d
7f97286b5bf04e76cbad63dbe6ef8ef8d52ab191
11195 F20101112_AABJAL boock_j_Page_33.QC.jpg
dd32e1fb11420b0582dfbc95181906d2
813195d3a6b3a4bdb7c4b5628d8a315fc792c3ec
28764 F20101112_AABIVF boock_j_Page_16.QC.jpg
802a54347a8af6896f83169c0a63e36e
82a386093866fb9ed2d8d9f9fa5aae9cbf2b31c0
4769 F20101112_AABJBA boock_j_Page_26thm.jpg
31bc4b4f01c61bbef19b199b7371553f
10a5733af8b81f0836e837ab1fff382fe41219a7
2839 F20101112_AABJAM boock_j_Page_76thm.jpg
007d8c0b336e435babb399f948b2da45
02e388a21454cad50065b8c06a32336d1eb1407e
5401 F20101112_AABIVG boock_j_Page_69thm.jpg
d7ccb029c4aa1610d4af20b687067a11
880a8d40222f914ca07010b559f060467439624c
F20101112_AABIUS boock_j_Page_52.tif
fa29a75bce21e8408c4b86eb4cf4fd52
7bcfec593fa9d164798427b393d77b7f760e7341
15891 F20101112_AABJBB boock_j_Page_66.pro
150f0e572c924450a80f1fd9dda3765c
c30bccb13e5eacdabc6cbaad916a36567ff0e92f
F20101112_AABJAN boock_j_Page_66.tif
6c4e7457c1c6f43be3ee493e45e252e1
01d5c9fd3a8f21906f069b5463ac0ed8a98aa984
7553 F20101112_AABIVH boock_j_Page_55.QC.jpg
cae10ab9df3df70e6bc3ae84f3edc43f
1fdfb981d0f2082aabe240ef02d7e40de01bf324
F20101112_AABIUT boock_j_Page_15.tif
4b3b609e4e3fda3233fb35962819988f
9b622c5c0fc45bc5bfbc32921921419bbe24a9d1
2098 F20101112_AABJBC boock_j_Page_53.txt
d41cabf06ca7176c41ec4c97d12acc93
0645d46e200e1e2eadb9233af8a9042e4a9b49d2
6179 F20101112_AABJAO boock_j_Page_23thm.jpg
59dca4a8556b0f7795161b19dcdbf3f6
613c5b5ba322c852cdbac4c58bec201c4ea754ce
5189 F20101112_AABIVI boock_j_Page_03.jp2
450c801ba334fa35465ead2929a75411
4c5b897e4b36ee6583f84f3d50d490dbafc7e2cb
893599 F20101112_AABIUU boock_j_Page_43.jp2
856e1e2fce16e33e7466930e31511f5d
947b26d947c543398b3dadeb7490c7ebd69ba467
71369 F20101112_AABJBD boock_j_Page_43.jpg
0501c1c66c0668adf6bdce312e087579
6f4291143bdef8bdcd03bda9cf1c0d719f7007fc
F20101112_AABJAP boock_j_Page_79.tif
9927e9c21afff5ff34db532c16d78263
d834921c922f59649d81b1979af3510d9846932e
28193 F20101112_AABIVJ boock_j_Page_13.QC.jpg
09739a4fbdf1bf632292f20afb537314
8f4168c2c2f33ecb05d934f101078ec40effb4ac
35715 F20101112_AABIUV boock_j_Page_68.pro
4378060cb878ea20111e0ad430dc0ec4
eff407e7ebb5f70fd303405adc906d83cd0bb0ca
F20101112_AABJBE boock_j_Page_33.tif
989a39f62255f5eef192bf7b6a7b4878
0f6e6a78480c818819853aab72fc3618b4b94ca5
9896 F20101112_AABJAQ boock_j_Page_72.QC.jpg
1c6cbc9e065671a9732cf4c2881d7081
89f9a7c0f03666605249abb4f437c8b401ed5daf
12548 F20101112_AABIVK boock_j_Page_95.jpg
8b722df0fb513bc1ac15c679f07cd929
c37dd41c9a40cb5f0b967744634aa2d0791975ea
15761 F20101112_AABJBF boock_j_Page_32.QC.jpg
1cff4cef821ec7dff3f2a85ae1074d91
d91ad58dfa0842a25ceeb9425aa10870b89b0109
12120 F20101112_AABJAR boock_j_Page_49.pro
d3c1d93e0ff1b2a7a38bbd265884cbfd
09f330adac48e4ba201d4179659744c7394a25a5
30307 F20101112_AABIVL boock_j_Page_66.jpg
48308d9d813338b975317c9c3c5c43c8
b5603857100d6f9a14b45b04e9342ba5856e07f4
21492 F20101112_AABIUW boock_j_Page_08.QC.jpg
9d07bf77e522ccd52eea307b4caec9e4
e60c4f638176a3e53298ef4416063eeae18e46c8
615 F20101112_AABJBG boock_j_Page_58.txt
3050064c7dc41284472a8a3db4f7a9ab
ad6f0ea7db21a2bfa3e9c60e54538ccc9d8b2797
113941 F20101112_AABIWA boock_j_Page_14.jp2
b3b905b9ef1af5c52e76776048c9dcec
9c8cf6683aaa5f25174fc919f25fe282d5c468b7
96639 F20101112_AABJAS boock_j_Page_93.jpg
a78a9f00eac617c2d8fc16d797927fec
37414ee167008685fcde30548b8813b93d2e4d14
2566 F20101112_AABIVM boock_j_Page_40thm.jpg
7632f5e082c6101f9ba9688b84c563ca
94b23f1b66241b8538d86868d4d6583d963ecab5
22911 F20101112_AABIUX boock_j_Page_15.QC.jpg
7829b90b3ad39806b17e5d0a69e8c4d1
60bb48157ed85221f72ab6ba78551e8082b3f6d6
22169 F20101112_AABJBH boock_j_Page_50.pro
333d16ad8227b52bec3a1c9c3295bef7
d341dabfdec99c8b47f616b2f652cc1b3b46d8dd
19419 F20101112_AABIWB boock_j_Page_51.QC.jpg
44cf2c1ef2e6b11d9009d45cf6eb9710
b7f06a980087f77cb8bd5371101c14f227594e37
F20101112_AABJAT boock_j_Page_49.tif
28fc57d3c2cb36e625718b791be2de3c
0156660ff7e5d18ee7da0412874eb138adb94c67
436665 F20101112_AABIVN boock_j_Page_39.jp2
e99e8efcb662163e1c167b0a4d27778b
24c7dc6f34d1c259a8aef14db4d2723e7b7c3c54
14315 F20101112_AABIUY boock_j_Page_77.QC.jpg
00eb76fba859952774c0143a304d5b0a
7e067819a533252f8abb0d6f2ca8e8e25ccb4b5c
816 F20101112_AABJBI boock_j_Page_40.txt
80d341947916ffddc756e42c0f14f704
d80f1911c44c9d09929aad8644a7f741198419b7
50425 F20101112_AABIWC boock_j_Page_31.jpg
1cd03631d2aa98c54e8566e77cc7874e
3080109977d383a68edb4ee401b60500b2a576f8
15398 F20101112_AABJAU boock_j_Page_41.QC.jpg
38f1f05802cb9cc7ec4b85f2421730cc
00d839b8edf9fe894a930481b5994f47c7491e64
2276 F20101112_AABIVO boock_j_Page_15.txt
7690b9a12f37014622ebf411db5b3fdc
0e28f770e4b1a94ae0f84e14d4773b2f63b30d3e
1316 F20101112_AABIUZ boock_j_Page_67.txt
9da9de32b38b95d71093f13065ed8820
9747b53ecb90b8a69b11998ba3b179b2203fa468
49122 F20101112_AABJBJ boock_j_Page_22.pro
f4e59acd2388e5fc7971e44250dd5254
5092b1697d634fead7878b55c6d4f8073dfab1a6
27595 F20101112_AABIWD boock_j_Page_88.QC.jpg
ded323e2f712626515c3f75766e597c7
d27a8ac75c36f87ad6ed6670b28dca876bb82b25
90305 F20101112_AABJAV boock_j_Page_59.jpg
e7c648f6b0ad41c7c2b2f34438ad5971
a1d9b2c8d4b873b054ec1370d410630c76878ccb
2358 F20101112_AABIVP boock_j_Page_09.txt
a56bad219fb6a768330c8d4db34cf54c
4a4ed0aa5e58fdf9cb97160d388511701d33f13f
5665 F20101112_AABJBK boock_j_Page_43thm.jpg
ce899f4b9a31a36bed7a4572afa7bdf1
46e1463719304d6747d873de0b3ae04c31d32d23
26534 F20101112_AABIWE boock_j_Page_55.jpg
de469e89ace8d758580e5c37c4930475
8ef50e46fa726903b9f7ff25813f6d19bddd2397
81370 F20101112_AABJAW boock_j_Page_30.jpg
40c2572f7f616294148875af734ab9a4
3a22c3c2a9d94cc4a97516e2bc1825b915a12357
5217 F20101112_AABIVQ boock_j_Page_57thm.jpg
95895ce2bc7b5eb47a19bef076388509
df3522f0c60f1f10e900a82c6997cad19b008393
20140 F20101112_AABJBL boock_j_Page_31.pro
1b9890112705addd8eb9fc32566f2476
17070946ee9bdb44572ff397ae50af8e962c04cd
906156 F20101112_AABIWF boock_j_Page_69.jp2
eba94a62fb895dcbf0e71d4a1287370c
5874e2b96d5ffe901ac1f73ae70dc1c34d93724e
4950 F20101112_AABJAX boock_j_Page_60thm.jpg
ca8ccc275b63ba65e5b75578382536a1
56634b084139f9af192dc33d25f88f64d9103e7f
12360 F20101112_AABIVR boock_j_Page_11.QC.jpg
2909ccf46c3e5eb64823487a00ec3009
bcfcafae55877bd4e461a258b721a0ccb5d8fd56
120831 F20101112_AABJBM boock_j_Page_28.jp2
e6488ee880775a8726bfecbfa3235b33
5ff1d130861b602712840d46a5f83bb734c6470e
80347 F20101112_AABIWG boock_j_Page_22.jpg
87f7c64957669d2b4eb3cade9c7eff42
6d0a05cc74933043e76a7c83f03e3c2ccf8b1107
6291 F20101112_AABJAY boock_j_Page_92thm.jpg
b91ee4113c24c6fba3e7f366ee407862
674ba933baaa7335c265953d6e8cc33010d9951c
1457 F20101112_AABIVS boock_j_Page_75.txt
b9f8e8fd63a9ac5bdd412a007bc04664
7a48dcf77294df342bce6bd7a8a1f6d3f52b86a5
320341 F20101112_AABJCA boock_j_Page_55.jp2
78c23d2b2e8368e6c823fe80f980d224
6b716549055b97d2d0bd798a9dcbc2ab475afef2
40234 F20101112_AABJBN boock_j_Page_17.pro
3a29072bf781fa4265a3a82954319088
692e108b420da62e7d22ec302a3a688454c6d9ac
52528 F20101112_AABIWH boock_j_Page_81.pro
49ac059a96844af7c9716aa99f225396
4a16dd67c35da39e99ebb6975151d6600be023bc
44818 F20101112_AABJAZ boock_j_Page_45.jpg
a5396bcf05ef2e310c32a483204bfcaf
18a870e05573e510e88cd80180eea92afaf22b03
83681 F20101112_AABJCB boock_j_Page_29.jpg
091ce89abc0da31016fa69e6eeed5acf
87f72d9031e657e2f00488909a7504111a4fe37e
25999 F20101112_AABJBO boock_j_Page_01.jpg
cb7a46e00fdb52388ed3f7c5210d6016
8ea87c133fa2a1d2c063613f8e62bf619bfb5ed6
3850 F20101112_AABIWI boock_j_Page_52thm.jpg
906ed50e936104b582876f0019d8a5ad
d08b331ab72ca729768371243896c4b30d593525
29018 F20101112_AABIVT boock_j_Page_23.pro
14aff16f0e42efa4118cf68ff4c988df
1c7810cfcecd5126c5fcaea3ee35e5465653504b
791958 F20101112_AABJCC boock_j_Page_25.jp2
08c8da48e2c3b1bb4804e2d3acde9f11
6a3bb6bb5bdcacd0e15c0f34a58470961e2e5553
18809 F20101112_AABJBP boock_j_Page_06.QC.jpg
a41f2954a2a0bf41e86ef446f134a651
a984ad92221a73192c088e713ab4958a3823a0bb
599086 F20101112_AABIWJ boock_j_Page_75.jp2
cf4abf584e31524bfa5391b4fb4b5502
00e49c4fd142af415388c81a074bf753ee70b998
35425 F20101112_AABIVU boock_j_Page_76.jpg
f1af6fc4b432423fe8e0ef23c56a4fd1
92074eaec25e5417ba0d4e6f3320340865edf1f6
2142 F20101112_AABJCD boock_j_Page_13.txt
3061109c960b9fbee3194668f85c5ea9
3d1803f239740df26c5d180b3d12369ade43533d
98152 F20101112_AABJBQ boock_j_Page_94.jpg
2492929d09c43596f010ec9e142f2ccc
dac745413fb7b1bf340f13dc67d0554ab2a4b48d
F20101112_AABIWK boock_j_Page_11.tif
c06e31975b819cb321819812662abfa2
3d02d8a1102671a26512999ee8cea0493a53fbb8
F20101112_AABIVV boock_j_Page_20.tif
2becfbe2ba0cfff6e3a11c92b23327aa
a600c3fd37630c0c43a29ca64bd3ec5ca5e83265
8123 F20101112_AABJCE boock_j_Page_65.pro
7334e395712cbf0f0445c219a1997ab6
38160b863ddf0ce670db631a3826e0cd924f747b
F20101112_AABJBR boock_j_Page_54.tif
e419a6c2c44c939223e2196018cdd894
a1dfa52ebd0bc746dd0f51fb313e051034ed79f3
693 F20101112_AABIWL boock_j_Page_65.txt
ddf5d8e95d29001549a60385f004c7c3
a9b5df81a793663d6e5a941507674729d2761535
66406 F20101112_AABIVW boock_j_Page_82.jp2
dba40ad2e9b6a1d567bc0214ccad8e77
00d29628c66379becbeb7cbe01e7401c3e00ac11
36119 F20101112_AABJCF boock_j_Page_26.pro
0bc15bd81bd2b35a6387735e6d400a44
43a8a45cb1df63d287634d7182670eec768850ca
53120 F20101112_AABJBS boock_j_Page_14.pro
3739bc1aa63a652bdae6d1160373d9b1
afa87df18efb95fbcc0e9d55635655c8b96312b2
346 F20101112_AABIWM boock_j_Page_07.txt
01df308a462d03bf59f7076a49817a7b
c18e401ddebf1ef7d748790762d7a56b24c00bb3
7793 F20101112_AABIVX boock_j_Page_01.QC.jpg
528196ca45511d696e7d0451b099998d
f255d1d28e1cb80a3f0a5a6ca6b6814a7d087f2b
F20101112_AABJCG boock_j_Page_78.tif
989ac2cd94512a05c6b8d226dd8cf814
3837159aec61a1430dd1baa9877d5c7b64a91b03
4870 F20101112_AABIXA boock_j_Page_68thm.jpg
1cfde42a662d851e4d84df028b66ad98
68e1e45ee9953b789b9be869c8fd9527f4e9d5e0
81279 F20101112_AABJBT boock_j_Page_85.jpg
933fb7657a22f9188715cff4e61a801c
83bd58cbdc37aeb70ae4a8c8b9aeb11a27c5e13f
F20101112_AABIWN boock_j_Page_50.tif
1c4aec39bfb0fb5ed31c5247eaff4023
692e61eefb6c98d403079be11252a2776c660a50
2245 F20101112_AABIVY boock_j_Page_93.txt
603737d66e3e3e5796d1205667d56402
781dbbc22b424d892d649dca396ff93a9f62d769
28970 F20101112_AABJCH boock_j_Page_41.pro
21aa1d876259be9b274c9c2e68c23a51
306a367154ac7011e042d4834fbbd78ceba8a117
5752 F20101112_AABIXB boock_j_Page_17thm.jpg
8729e7bb358655a50bd40c6347f82ee7
9e8ee0b710b5ffa53d0a279ee7b6cdc746c79cd0
929 F20101112_AABIWO boock_j_Page_37.txt
4bbc46d5255f78636187a6c4c4bc875d
0007413b35b467d419b61dfd7fda2d84364b75af
25590 F20101112_AABIVZ boock_j_Page_71.pro
a189f5675de4b1eb5999131031a80e9b
071c02f7e9391aab942c76b9cd24fc8df440421f
F20101112_AABJCI boock_j_Page_83.tif
fc2d235b3829a6f95e725e81221e04f6
f775083d9250a6711298041461e298807e09ccec
22817 F20101112_AABIXC boock_j_Page_35.pro
e5261dc398412251290eaead1a259ad6
a6c8927d0b84f4018c49ef989263d7b757899d16
21130 F20101112_AABJBU boock_j_Page_12.QC.jpg
ddc5dabe2a6c84d94a382af89b195f9b
821ec9954745fe23b824039dc0d5d096a88b8c65
21465 F20101112_AABIWP boock_j_Page_19.QC.jpg
1552c43f0a25e70a4084e24067f9b847
947138f37fd94df6103f3921322e160194c7e249
5172 F20101112_AABJCJ boock_j_Page_05thm.jpg
ad369395fb09fc1f53a442c0cc983c20
a1b946ba0498cb66e2fcf68f1c3c0cd153eaeae1
530 F20101112_AABIXD boock_j_Page_01.txt
a5131972b0a090c929e875821b0e420b
c5c6c87da5bf804e7d30c092e0b7aa7d68f09d9e
46010 F20101112_AABJBV boock_j_Page_44.jpg
8ab86a2a7f5984b2a648df10d77cffd7
15fe628a55220971efa67388440a6522bca73725
12921 F20101112_AABIWQ boock_j_Page_40.pro
c72dd0fe1880831a1267b76f0599c554
4734b8b6fb58b52a9b42bfee8852083794e90eaa
341 F20101112_AABJCK boock_j_Page_61.txt
5a7d45ab3278d273ad6b483ce3e0dbb1
13df07b6c70734c742cdb1c96e72ff8a988be6e0
26858 F20101112_AABIXE boock_j_Page_62.pro
685cd3d5937b2c89af50d346e711c1a0
62752577cbc1234d3a949cb9978d0f7a623a67e6
81375 F20101112_AABJBW boock_j_Page_08.jpg
93c88d749f7752d168fc576ccdef2409
15a1886f0a06859ba1b61812c8dbdcf7d3e0c8bf
6691 F20101112_AABIWR boock_j_Page_88thm.jpg
6a5c05090f31a5ff564267702c38c3da
e53144e33e79b47e635f536e8e25bea1b59d0352
641356 F20101112_AABJCL boock_j_Page_56.jp2
04ac64508ec8e96fe1fe1141963779cc
88e1cdc374ceb0ec6ac217986ebf6338a8d84da2
82368 F20101112_AABIXF boock_j_Page_89.jpg
d4958f51d61a21f8276d2cd474f28769
63f67a1a169ea8873a7ff4d1fd0379a1782e1850
454308 F20101112_AABJBX boock_j_Page_47.jp2
5f67020f034103f407c1a25a5a4fa9f9
0d17938bf031c5a8bc3fdea4d89cf6280df2b46c
1051940 F20101112_AABIWS boock_j_Page_80.jp2
c7a9d549f0aa60db22095cf38fc51d65
e5e384b57a03bf5b9d49bbdc7ca5fbbdc4a4f036
17491 F20101112_AABJDA boock_j_Page_10.jpg
50c8ef16f39fafd2501747e805d5a24c
707f16a062b42bdacf966f0197daf48fe1fd1cc6
2255 F20101112_AABJCM boock_j_Page_28.txt
2c959290e36e8920fd03fad2dc1dec3f
c677c10fe46dbada87a1e42cc585c088b96b836b
60038 F20101112_AABIXG boock_j_Page_94.pro
41c7357ade33ca6d1738a4cdb993cd02
b4fce8ee7c7eeafdf0bc5c2b0a739bbcbe3322c7
3103 F20101112_AABJBY boock_j_Page_49thm.jpg
e7ae4d77300942b857bb5b66f7f0e386
76d1859feea0ce66e4c2108f4663d23cf8c54202
1051958 F20101112_AABIWT boock_j_Page_62.jp2
49be7f8d1525f96615d6a12ce0939d82
4d82b40b60fc7efe2356825700e2e29edb783bb8
35440 F20101112_AABJDB boock_j_Page_11.jpg
3e0b774a38bffb2906d1d765066a4679
a29930c6e6ea6a0e104010d9340436cbb866a7f2
1591 F20101112_AABJCN boock_j_Page_41.txt
b81624f9e34af822e5eccdaa2e70aec5
2044606bffb74b57dab801717f0bb3f3ad5bb752
F20101112_AABIXH boock_j_Page_81.tif
356d35627a95363221fa039ac88ddb97
b750e76066e99c3a295e05cfab3f134b00370b65
21017 F20101112_AABJBZ boock_j_Page_67.pro
69ab4c67541942f5298eeaeb579b6664
a39adbf6622a193f6ffdfbab2696ee4943b4f78a
88039 F20101112_AABJDC boock_j_Page_14.jpg
100772f5339110e0e212e5d773e16a48
0e451c8be720f194df778408b00ac6b999955136
70106 F20101112_AABJCO boock_j_Page_12.jpg
3ccca74c03b120373cf6b1c8f0fdfa7a
81b1e498619df98da282f7504bd6f5b475cd736d
38126 F20101112_AABIXI boock_j_Page_69.pro
7d5a9fd0f70f1061caca53c8ea7a2458
1d8a42ac43725c5fe09d454c4410bc7456a79885
5539 F20101112_AABIWU boock_j_Page_58thm.jpg
c8987715fef40dc6701cee217d516888
817c123e03180a4be2a01cf0f4f9449c1ab35406
82329 F20101112_AABJDD boock_j_Page_18.jpg
33f05ca8e0d5b029b8b379c4a82591a3
b2a95a058d6fb739b3a413ffd7d1a220c1a4c87d
8509 F20101112_AABJCP boock_j_Page_62thm.jpg
c7b3e3cee3a5e8285c9844e97d0167d9
e2c46be802a0c3dfef26b9206ef4c6c549464788
97273 F20101112_AABIXJ boock_j_Page_92.jpg
d1fde2546f55947f7288cac882f9fe6e
1b15129a6c5c8ab3616b1571227da42ca36b173c
20429 F20101112_AABIWV boock_j_Page_43.QC.jpg
fdeed8576411244b36c2501eb40db30d
ae984509a20c4a68bf38de36b259cc764fae2fc3
63989 F20101112_AABJDE boock_j_Page_19.jpg
2acda86320644855903cbeac91a9bb43
47c54f15577cc7fd794a2ab556cc16c3522bd368
57171 F20101112_AABJCQ boock_j_Page_58.jpg
e121644271c4bf8efc0a8e0e1c5b1d25
0be75806009b817c777a00802927a4a931aa9998
9205 F20101112_AABIXK boock_j_Page_40.QC.jpg
4f7b007898341152772d5d91c6e4d804
d5540f152ddb28a8fb0f261cf4f898d4d901381d
49268 F20101112_AABIWW boock_j_Page_87.pro
48a1114bfa71673d5223967b4f477e75
efc3e44d7747a63287b1ced3acf15a3323958feb
63795 F20101112_AABJDF boock_j_Page_20.jpg
d214e6cafc0b4f4d2e7066ef08b18ef0
5471c9a4f242f7fcb973949efebc208e2112bc74
87318 F20101112_AABJCR boock_j_Page_79.jp2
43e2e94ab4635d5c878f11542c3182fe
a4c751f22691c8879120741821c13af65ab49e0a
858361 F20101112_AABIXL boock_j_Page_51.jp2
033011d034ec17070e8b9657b6e5b1ec
1549f739c5485523d7a652167d1a75ae10274273
61536 F20101112_AABIWX boock_j_Page_26.jpg
8529d15187ec4cbccb78c4493fb9fb52
a75eb0a575e80dd23fcb9d8878d772a74ea22bae
86154 F20101112_AABJDG boock_j_Page_21.jpg
06e0432e595827c64262f4f461ce233a
20735a3503d668b018714ab22d244d1c84122629
9992 F20101112_AABIYA boock_j_Page_76.QC.jpg
aca29c50b48eb7ebde2682e333ebdef1
30763113d5aab3d35a2e3317bf2c419284f8f1ce
85865 F20101112_AABJCS boock_j_Page_88.jpg
95455bc47c34634814d976ba649cd91f
86aea0a175817299af2285f871d67c14d546cece
51703 F20101112_AABIXM boock_j_Page_82.jpg
19ae3baffd9c511f14c7414d74aa8e19
93735b793958ff2a70449fd90e4f87a94fd89ad4
4928 F20101112_AABIWY boock_j_Page_02.jp2
e37d10923523ef653f4bd15603368d77
8de3bb41e83227ba9b0c726dda3410443ea5bd98
68420 F20101112_AABJDH boock_j_Page_23.jpg
e39ee25123fbb6ad6a100511e02a6487
5b3919c26965173f3f65d28a289548f9a0c5ece2
87564 F20101112_AABIYB boock_j_Page_12.jp2
698e0d50cd365a6fcaafd4a494ac81e7
a01c3363047cc80e977558fc28ea00b57047a1ea
111067 F20101112_AABJCT UFE0019340_00001.mets FULL
23a705440e9518eb5e0cc58746189f3d
ed4320884127c028d308676350fe984b58e43e29
F20101112_AABIXN boock_j_Page_09.tif
8aaec9ee280c0bef4fd53e4735322820
3cbfa90f50ffa51f8d76428003b0cb719c5c0b30
F20101112_AABIWZ boock_j_Page_64.tif
5f181858bd37d5059afd4d1dfdd4fe37
b3acecaa1595e773dbf9ae1a27f2381a079db6a8
67718 F20101112_AABJDI boock_j_Page_24.jpg
3c48fde149644e8e689df89c8581d621
8c444b0515c5f4424139ba456cb6f88e356a1886
478743 F20101112_AABIYC boock_j_Page_67.jp2
0f2c6edb2d1490f8cb846a0a01c17a19
d9ff0a1e4d04f0e5151f7d5f36071fcf0398e4f7
28666 F20101112_AABIXO boock_j_Page_53.QC.jpg
e7ad9feb8ff5393fbc77b2e444cf2954
0550492f27d68adc3da7f3ca32890947db6d15eb
62598 F20101112_AABJDJ boock_j_Page_25.jpg
916636a123753c31ab3d307c05291e6f
cc08dba9c4108b4fb2ab69bdce610f8c7d9c56ac
78286 F20101112_AABIYD boock_j_Page_64.jp2
e0c650cebb48cc47b4942b2b0c30a3f2
38209febf7c28d2c9ee33bbcd941b9c9eaa57da2
F20101112_AABIXP boock_j_Page_94.tif
3389580dd5eaf8e10debad8f8387a289
1a49a0cabfa15f6f346921679f7871c0c8050b04
73211 F20101112_AABJDK boock_j_Page_27.jpg
bb3a622078c732f04f4613f56a30e2af
796e3ddac9781afffbf5bf54ce67e6b0dab276d0
90005 F20101112_AABIYE boock_j_Page_84.jpg
ee00d556773eafb5af80575efcb3df88
77e8db16c588cd639208b1f63b96dfcebc9e94e0
3658 F20101112_AABJCW boock_j_Page_03.jpg
93e5cbe3324a303507b7eed53a5ecbfe
2caa2fa4d889a94b5eeea7c9608b77da79a03ac7
1038 F20101112_AABIXQ boock_j_Page_02.QC.jpg
41643e4db055d0fa1294e7bbd2ef3a09
d4672584fe886099538c0ef8457a0aca6b5dea0c
92496 F20101112_AABJDL boock_j_Page_28.jpg
4b81a45a7aeb84b5414108aba190fe92
1f56d7c821a2b23d2dcbb532b0fa23113d2a6b52
83682 F20101112_AABIYF boock_j_Page_09.jpg
db4d748de21bbf2f890d715c031ba651
67382fda7d86004a84bda5b55bedefa888f1938a
45213 F20101112_AABJCX boock_j_Page_04.jpg
49481cc0641971d69d039cb21adcb894
9fd7d3bcef8a4e133545347afcfe78bc987818cd
20114 F20101112_AABIXR boock_j_Page_45.pro
f6fa0bfba402e8d4d7bce2735c9294e2
7b8a9ab2ef38c4e68f453df6ce8f95e13db7267f
89904 F20101112_AABJEA boock_j_Page_53.jpg
657570f3e627df1ac7df678c9e07ddbe
447855a51640bc183d5c68ebbd564ad804a5ae10
50446 F20101112_AABJDM boock_j_Page_32.jpg
dd5bc86506026ff2e9812da40214d938
0bd55760d1bc24b41c696485c59bf0a4324534fe
73239 F20101112_AABIYG boock_j_Page_17.jpg
b82d89e70942b3235e095c9b20bb1d8a
fe45dd649d2b17b318d9d78a4b061b6342b907dc
89362 F20101112_AABJCY boock_j_Page_06.jpg
68f5a8296b970a5982ad484bfb2aef64
6e0a2df3e555e5d36e8d73e61a9460b695dd2fc6
55199 F20101112_AABIXS boock_j_Page_16.pro
2711337b7ab6a34c5e576411d81ea4c8
d1662c120297eac837d32a0ff2e9b61b73f88b67
35577 F20101112_AABJDN boock_j_Page_33.jpg
f05ff8af1d5c93fca567de8be900d284
d7507cf74bfaed92e9b5cae99919fb791ad93055
F20101112_AABIYH boock_j_Page_12.tif
0a206f015371e0a692a579e352234a95
47b1f668fe682c103f32316302f3092389611057
14252 F20101112_AABJCZ boock_j_Page_07.jpg
1c8ae70c41fb15c3157c860b45c42135
6a6c238123157734aa75a3ac4a52a1fec04bf968
18242 F20101112_AABIXT boock_j_Page_76.pro
037d89f93e6f51888e481b1f3236ab95
a459779c043c827cb010956356fab6dc3c4b2a8c
65862 F20101112_AABJEB boock_j_Page_54.jpg
395c2c8fe08a1267076af7d88c01d34c
c74e961ccfd957f707ae65720bc2e7e5bdcbc80c
41052 F20101112_AABJDO boock_j_Page_34.jpg
4d58db31dafbf9b76ab00e03e9e032b9
243a9204140dd45201d5d3c12a6f469421f429a9
636975 F20101112_AABIYI boock_j_Page_78.jp2
68fb2d6ef1262fa7781d2d7caeea1bd0
5d63a86008bcb2ef17fbda54cf53f58de7de0a2f
40361 F20101112_AABIXU boock_j_Page_67.jpg
6b09654e4788ad2e33977540f487a9b5
a20a88368d1698f7404898a69fb69c2c77c63acc
50471 F20101112_AABJEC boock_j_Page_56.jpg
2229495c117da6eb90bb7213a2c5e06d
81a7969a65338bdccc63894501a2c17b737529de
49575 F20101112_AABJDP boock_j_Page_35.jpg
34b2732aa8a14fb76204beb5158e8e88
026fc4eefc6bdb16cfc7f813dcbee4ff40252fea
462 F20101112_AABIYJ boock_j_Page_03thm.jpg
f3234fd6f85c6de380b3c5717988e7ee
19654c3765b9b5d7a31a4f3f6449a071321e6802
38597 F20101112_AABJED boock_j_Page_61.jpg
79175d76f9955fd75667603e951385e1
36419dbf795ffb3d2f8dfb30d403e483a09db4bf
41204 F20101112_AABJDQ boock_j_Page_36.jpg
81236f3f5034275a5f37db935f035bf4
a0d341af759124f2b7f970343322f33c6af4355e
34863 F20101112_AABIYK boock_j_Page_64.pro
63f353f617bd5da37245da665e224ffa
2bbe98c3ac29a605ea7396e5b289d929e38f7c4a
29632 F20101112_AABIXV boock_j_Page_28.QC.jpg
2f8becb84fab53d7d71d25a27d33c011
13067e70bcb20e5cc108537c2c1687f26d41a0bf
98816 F20101112_AABJEE boock_j_Page_62.jpg
88a03be4705f501ab5e36ca167b4ba99
be2c26b8ebb6428487720d74585fb66c519206f7
44303 F20101112_AABJDR boock_j_Page_37.jpg
b0d6d0a2b1eae5295487f5a8b3c9ad1a
f1f61b57a7c9f0834447fca63b1ff5842356cbf9
6573 F20101112_AABIYL boock_j_Page_59thm.jpg
c09b298802bf1d2dc8400e87d15ccf71
149d918a97cd67bc997996b7ca7d449b54a38b76
366167 F20101112_AABIXW boock_j_Page_49.jp2
97ca00fa57ac549b3db068c889f264e3
d4545eb9c79e16a698e9bccf059a1d5c7079cf5c
20255 F20101112_AABJEF boock_j_Page_65.jpg
2ba57eab1e1564f40cca7a2f92a2f0f5
bdbe1fec214914df3b66b3991e8b98f9286a1349
52919 F20101112_AABIZA boock_j_Page_88.pro
5d2d859e9e4a156c36aa1b25dd6dee90
34c654920ead36ecb24c324d466f46c9937d104c
38288 F20101112_AABJDS boock_j_Page_38.jpg
f6f0b68731a6cd4f97538c72a7dba65e
ad75ee4fd4e49e148a60a683acc8d05f277d7bd0
1965 F20101112_AABIYM boock_j_Page_22.txt
aed72ab1cbe85ca831fa922ca74f4524
7b7d581f0b1c9a2c5600529e80a7c85738b13abd
4331 F20101112_AABIXX boock_j_Page_41thm.jpg
9ff7911cac3f6ddee69562631417ab72
7189ebf7ff9c60d59eb73ea8e70c78fa55284388
62394 F20101112_AABJEG boock_j_Page_68.jpg
82373d9fce4058373bc51b9bc237eb3d
4ce5a3842140dddb534f9b28eb5be1cf7cef00d4
74184 F20101112_AABIZB boock_j_Page_15.jpg
c8ea6c0969d21ee54a85e134a9e80a70
4c7335a9c8c19abd433db0f3f14210ae9e6f83ef
37451 F20101112_AABJDT boock_j_Page_39.jpg
166c002431846252a749aa76112b70cd
385d00c8be4b3f6b6d5302809d207cd4ef3cd2d1
933 F20101112_AABIYN boock_j_Page_76.txt
a181e88c1e31005498a4f6c7c05a0c33
b343832f1f14a625c7e3f4821c2e1a074d791a38
1948 F20101112_AABIXY boock_j_Page_65thm.jpg
47d6cc3d4799153772c58b83c7d281d3
767a36fb87a778b3b2a57de934087bfd4cf46fac
70974 F20101112_AABJEH boock_j_Page_69.jpg
eeb65115f7a9ea37c4f23523dfab0ed0
1e5d7b398a5d338f1c0c011268fde54a38f1791c
F20101112_AABIZC boock_j_Page_57.tif
34c7db1fa8616c4b5476492acc77909c
046f675816490a4b5cb72e1642d5dc344b47e286
27556 F20101112_AABJDU boock_j_Page_40.jpg
704318458fc3741e8f059e147a87f242
94e9ef10170cb44ecde33f60ffd309b776a3fa43
115860 F20101112_AABIYO boock_j_Page_84.jp2
afabd8ce7545aaff13c3b960a8935999
4bdb3021333191c88b56399fed54b425040ea39c
2264 F20101112_AABIXZ boock_j_Page_55thm.jpg
3ba8be69409113ccff68f165ba18a8c7
83b2e850c6a99a9114f22c4947eef8b503d37f11
58124 F20101112_AABJEI boock_j_Page_71.jpg
4378c667c97c9159e4660e382f7e29db
9636fb47acc581e56fbf512e3461f036c49db8ff
83269 F20101112_AABIZD boock_j_Page_90.jpg
48be4d91ade306a5b8737bd26ff1c469
b2aac2f8ed0796246bcd7d38d2ba5aa431bdc701
53948 F20101112_AABJDV boock_j_Page_46.jpg
3683a1ade02735c46ab660f5dd9e52f3
64bb1be77df63a738d70f9910497fe523196a48d
11570 F20101112_AABIYP boock_j_Page_47.QC.jpg
f70d33c6f97f8fb8985679639f4c821a
fab2e18af7c04fffd12b23c5ecf3a8c41d17f8ba
35106 F20101112_AABJEJ boock_j_Page_72.jpg
3eccc995fc28adc0fdcf5df0f495173d
01b947a1d7f5682da829c22ad7f84e770b328997
F20101112_AABIZE boock_j_Page_67.tif
db71ea3bae88b64bc7e5f47926c5daf9
91b03296ee99f640c7abb0508ab8bdbd39122598
32831 F20101112_AABJDW boock_j_Page_48.jpg
53851f4a9f0e92384b9987930e5f0222
2e94028fdabf7394e09088a33fc5127305011b8b
20795 F20101112_AABIYQ boock_j_Page_69.QC.jpg
2e047e35496a05d2f046cea940ed7734
ccbdd3015224ace123f53c232dc02b38dfcc666d
52003 F20101112_AABJEK boock_j_Page_73.jpg
344b89b1bd29449397c247e7a6a8c02e
f09d6e771753f3c48d6a661a5c09bf6f54ab6508
6173 F20101112_AABIZF boock_j_Page_79thm.jpg
70442a658359b66d3f6bf7266dae94c4
daf620d24ba469adfef3ac853b2a9096a58ef158
32465 F20101112_AABJDX boock_j_Page_49.jpg
00158466962c143c2a4c8aa5061045ea
3e102e1bc3b7323d9901f41ea4f95829bb433b92
F20101112_AABIYR boock_j_Page_61.tif
0060157aea6ea322ef8b73eef8e3bd53
2a55d892148399210d3c5d056743a8a5db9022a5
114402 F20101112_AABJFA boock_j_Page_13.jp2
315ea9755fa81d41b4cc178c11944464
3fc0ad7c376dfdf285de7ea458f39c3d38b40820
48865 F20101112_AABJEL boock_j_Page_75.jpg
e03bd5b2ace1b241981992d015c7ad44
aca4f5a7db22d2533e0b96e0d96e4e8db59496aa
F20101112_AABIZG boock_j_Page_72.tif
6105fa4bba791d11d167650771726fff
42687d33822439647e72ab6ac8166541f6b6f2c5
41121 F20101112_AABJDY boock_j_Page_50.jpg
9ee291c7b1d81a1bd6f15bea1f59294f
ec7f6ed78fbc035063b44ad57b8902e6a8c62d22
F20101112_AABIYS boock_j_Page_80.tif
2a3566e121294d0750cea5c4f65800d8
37986219902bb93e42c8ee36c81a326b0f922f57
1007018 F20101112_AABJFB boock_j_Page_15.jp2
ac56574bb0a1068e51745584288229d3
21270161477f44eb84cd98e75cc238e9f16839fa
45303 F20101112_AABJEM boock_j_Page_78.jpg
2d882f5503b441544ecb252f2b36c3c9
f8ea589f25d8a150f191ba4c87a669ea8e9cfb92
F20101112_AABIZH boock_j_Page_25.tif
2584b0f30be13a3e193cc331a966a4c1
8a45fb1a85cbc9602485e5d797a44133b1b38122
66924 F20101112_AABJDZ boock_j_Page_51.jpg
80c9fa74468644b882bcdac042a5bc38
59c777c0c201cdbe05556a218641b9dbb2380df1
F20101112_AABIYT boock_j_Page_14.tif
be04e22053847d8167fdc37085c2fdb6
92ba7fcdff5486c2b12a922c7c2f8375e902e02d
83159 F20101112_AABJEN boock_j_Page_80.jpg
d53cf8268452d072b7170cefc8daddfc
0dcb40defe292804f458b14560463a4bbb8ac4d8
3851 F20101112_AABIZI boock_j_Page_75thm.jpg
47cafca08748f8ba2044b3d6c3146ec4
b9d550823d75c35817fba9d02a488ecf5ebc96f2
490710 F20101112_AABIYU boock_j_Page_36.jp2
6cd08195b64c293786ddd185ceecca6c
df5225078d730cea02c43b7d8cbe35c730bfdf4a
118378 F20101112_AABJFC boock_j_Page_16.jp2
018da037ce343c4610be19c2f4148b5e
24ba60a50e1c1d8ecb25f33ac72b1a1bda52bdee
87483 F20101112_AABJEO boock_j_Page_81.jpg
184a042b6757d90198864a73ac044347
3a022e131fb2fa308068a866199dad056556c43f
779 F20101112_AABIZJ boock_j_Page_33.txt
9dabad42b2316858986148667423b34d
8bdee1b0e1d73aa0c6052586114a258a5b975f7e
112729 F20101112_AABIYV boock_j_Page_63.jp2
ed913ccb49cc4d12d23253f780e9dfb7
748a75a01e05ae064b228ec37328484eb8192340
978905 F20101112_AABJFD boock_j_Page_17.jp2
a231e8e8d5991c7b27b7a378a9de1c33
13d0781adf415bc9e618ea89cf656536198a572f
83416 F20101112_AABJEP boock_j_Page_83.jpg
94fb6158a55092cce38548a08c1b714a
6c7c58746ff5ab67c0adff463e415667beab98e9
56123 F20101112_AABIZK boock_j_Page_41.jpg
9bcb0ae93fddadeee181ffe3648544d0
836832ac2de735d7dce81424d44288be5635e3ab
106930 F20101112_AABJFE boock_j_Page_18.jp2
ba6d96887b650ff0aca51c65b54e8191
14b66e33c873277e23bd1981e2eb1c0a79ba0110
88573 F20101112_AABJEQ boock_j_Page_86.jpg
9e138f2cd4db92b215a6012494c4ab1d
5057b8193e250549e7899d0e9fcfa7627842da07
F20101112_AABIZL boock_j_Page_07.tif
b99e2c01d6b82ee04665f5af434c3ae7
b1a63ccf30a94e34fc8f12862cc7538260e8a170
2721 F20101112_AABIYW boock_j_Page_66thm.jpg
ba614b779273f92a3ced24cef8f8fd35
68b809a4fb7953abae78c8631ad796e623cfc1e1
1051969 F20101112_AABJFF boock_j_Page_19.jp2
08188b46124d0b69fb791e5c8ada91fc
4cd401ec66df46eadef3d66f872c4c42d9610bf3
84083 F20101112_AABJER boock_j_Page_87.jpg
74a2821b161417442913ad487871bf0b
100df2c4b1521a07e9cd044b3cce3d4c2afc7a89
29939 F20101112_AABIZM boock_j_Page_82.pro
3ef86d12908d79e53985910989fa9cd7
e790a4f0b1d12f4681feb9de1a73e8da1c3c40c2
F20101112_AABIYX boock_j_Page_05.tif
cc423939b7f25eaaee3e090c98301eb1
2c8405da1e5d2bcc0e780f1fd7d054a03946733b
775506 F20101112_AABJFG boock_j_Page_20.jp2
1246f1a3e963a38af00637ce799dd9f5
b8f065774792842dc6fb85fde7705f372e4607c3
46913 F20101112_AABJES boock_j_Page_96.jpg
420938987128eadc50f71cc8d779e189
5dcb05223a006921c2aea37002bbcba5f60142a8
59604 F20101112_AABIZN boock_j_Page_64.jpg
e72584550d35f23a4c8e4158019e1ce8
86fe642ba6ca485522518a542ac430865dbbb545
64970 F20101112_AABIYY boock_j_Page_60.jpg
242348125c3db8a6da73d9b178cbe9e3
98c62da18b4ee5fa83d4475ead73b154f6fbfeb9
918831 F20101112_AABJFH boock_j_Page_23.jp2
c3c3576a0ddd4277c71f0ddf25ba280b
8428a33e01be8e5892d51756dae691fb0ea2b0ca
28351 F20101112_AABJET boock_j_Page_01.jp2
164208164f9a83d213f1cbacaf7f27ba
8605fad09b45377695d242db2820b3c4883779f6
3079 F20101112_AABIZO boock_j_Page_67thm.jpg
13801fb7f05260680fb39bbe8d9a3af9
10d6ee939bdebd60a8f4e55ed695f214fb0ebe3b
1519 F20101112_AABIYZ boock_j_Page_90.txt
716a025a6e69af9e21bfdb35b46455fe
429eecc0fb332e6f924e2123f13322d7ba83eb75
924211 F20101112_AABJFI boock_j_Page_24.jp2
f8bafc6e2c93cc4dd0b63551e1f8cdbf
2f0a9f304c42fcfccbc9939641221187df8497a6
1051982 F20101112_AABJEU boock_j_Page_05.jp2
1a0735138463039e2fc19d264db9370a
42d600df62f514ee5638ed378c3dc2ed70559089
5356 F20101112_AABIZP boock_j_Page_95.pro
9fff9b183f929ffa35e6c73054e3b146
d266c40a02792b0a316230e0f1a83706649f0c8e
80348 F20101112_AABJFJ boock_j_Page_26.jp2
c76e1a03fd51e3be63ea0a1a0d82ecf5
fd2ea049d57bbc78df13f7d831011e6286394e9b
1051981 F20101112_AABJEV boock_j_Page_06.jp2
8f4487c453e3b67d1dadb0dafdeb4163
8f929015f2c172bfe62adb3549a41c0de0552d1c
11632 F20101112_AABIZQ boock_j_Page_67.QC.jpg
b49d7aafb624f63a964604b6ba9e809a
1f12e1ad5c281662ea142285c82d37b8f6dba3fb
90134 F20101112_AABJFK boock_j_Page_27.jp2
60450d423fdc2516309460ecfe4d252e
9ef9c249838bef6c7293277f39fdbf0d92f18db6
1051980 F20101112_AABJEW boock_j_Page_08.jp2
67377da9913dca10e012840d94552798
bd1dacfe860e5ff773ab9f2f67852336c1f0aaee
3576 F20101112_AABIZR boock_j_Page_45thm.jpg
11ab121e4cf6c6daa1c214bf4f5d4a1c
152c19c49283d22ed5f4ddee2598be8a090eba8e
117895 F20101112_AABJGA boock_j_Page_59.jp2
2d6295f2f5a04a04ab789083a9fdd4ef
83cc9c5655a5d5baabfced3b2b49236620e51d75
109412 F20101112_AABJFL boock_j_Page_29.jp2
7493f6a2f60e5d16c0beb1786626a856
19c98a8f0f4eb2b2c6d91dbec1cb8be6ac676c78
1051976 F20101112_AABJEX boock_j_Page_09.jp2
7e22f0e4a1d5fefaca34a8ae25a32d13
463eeb33727fe7a06fa911f811c848d7376684a2
47108 F20101112_AABIZS boock_j_Page_91.jpg
5abaf970175a2e68979b60a3b68023e0
519444e8d996c7d6e81679b77cde564beb1fa9b3
692696 F20101112_AABJGB boock_j_Page_61.jp2
440cda923663a6ca8a3a9dd3d3b0af64
550d0fb89e6215c66a88ab4c257921cfb1316e2b
106953 F20101112_AABJFM boock_j_Page_30.jp2
74db0a6e5606e6658e71846ad6d7f5df
03fdc4ef3787868fea38a9e7c158a9850a33099b
335691 F20101112_AABJEY boock_j_Page_10.jp2
42baac524dbcd4f2dcd9bb1c4c663ed9
24216874c0b8cb27e86389fcf1de15cc0cd1cc34
92 F20101112_AABIZT boock_j_Page_03.txt
586a41b71dea11165012114b2c23e992
920e97c015fc31c54fd044de887dfab484531ec8
212355 F20101112_AABJGC boock_j_Page_65.jp2
78c783b087f8c074b960a7260adf90aa
e8dcaefb5c01f575e3ebd5f3164235fe77223f61
538590 F20101112_AABJFN boock_j_Page_31.jp2
8ccffc1d2a73dbbc48437be5c52ab3f3
c396c73067815d947833cf0af3704b4bed099cec
46228 F20101112_AABJEZ boock_j_Page_11.jp2
9a1c3e088c59ed2c2ce2079385bf9a45
20c6e50f7d0a3fb25c92457d3b115315d117c241
627709 F20101112_AABJFO boock_j_Page_35.jp2
7bc294a93828c0b9e804ddde0ea6a72e
56ffdbcef18b568a9ceccf05a2289936e1863baf
25562 F20101112_AABIZU boock_j_Page_32.pro
02f5c0e7e3435b85aee4e30011d1e3db
098fb234afa97dac9ad8ee031cb92126ffd37100
329652 F20101112_AABJGD boock_j_Page_66.jp2
c3abb36891255ec29db5a2d2000bd89f
d8b577a786e4b65624794da35fc9dde10f1a7dd7
584191 F20101112_AABJFP boock_j_Page_37.jp2
f0a90cb6448d566536d515adad013cd2
62c95e87ace9232f187419eadf1a4e5b3912058e
662547 F20101112_AABIZV boock_j_Page_46.jp2
5b9161adcdb6988a26edd80580260b4e
ccafdebcab6330d98b88fc4ec6a2c1f41ec58854
92656 F20101112_AABJGE boock_j_Page_70.jp2
fe5b72b1b9dc805db8e8661fdb31d084
a52931c90ab884c5c0d6b81af99fc775900227e3
483197 F20101112_AABJFQ boock_j_Page_38.jp2
0b5c705843e689fe4b145a50032ea209
968e348bcf2173f5de24ccaa085ab72265d84519
780753 F20101112_AABIZW boock_j_Page_68.jp2
0ca1aebfc5cd366f67e068a7db2be859
a6879b6e255fea7b036aa345f51fd81cc92c9564
421287 F20101112_AABJGF boock_j_Page_72.jp2
454b7a2c9d278bc995dcd6970125090e
61f9500f1b7eb905c8c6162a95ddae23358053dc
848673 F20101112_AABJFR boock_j_Page_41.jp2
f12929b70e5067af126e976c939c893b
6dc1d300c9e60084fe91c5be0108f68ee6565def
634854 F20101112_AABJGG boock_j_Page_73.jp2
5efcd886d7722e5d4d4ae51128a2a5af
9ef546a0c5462be8b3155c03d6ca8b021c31ef36
340319 F20101112_AABJFS boock_j_Page_42.jp2
8e7d00ad3b413b622583d501b7671906
062b4a8fe1ef3ace895b9403f4e5f25234f337f2
38476 F20101112_AABIZX boock_j_Page_47.jpg
e86dbaef9fc98aceb1382327b36e7e55
adc961046bbb5ed1d2479a9b09e07f6a20380400
426544 F20101112_AABJGH boock_j_Page_74.jp2
a195d11f9df4cf9c26c71c14d87eba6b
83ee1b3d7fa26dccbd3561d84998fe20f12608ec
595204 F20101112_AABJFT boock_j_Page_44.jp2
1d91d8e04aba1e394fb8cf5b42f9ae4e
04bacd84c86dd94dfe4224f4262059e4d2e31b3f
26112 F20101112_AABIZY boock_j_Page_96.pro
eb55269a49cc082ae8c6e3a092b67479
11b2632d654fd6f48a7f56f5ded307aa7f7884c9
452771 F20101112_AABJGI boock_j_Page_76.jp2
740e4666efe7f12bb3e1af4410b0d9b5
d9159bb40857b3abe2a211d12364c5abd85c69ab
559718 F20101112_AABJFU boock_j_Page_45.jp2
ed7a1a647beedde0104d720f8829c72a
d0c97afc843921052a1b4532085c79caadcf8e1d
6466 F20101112_AABIZZ boock_j_Page_18thm.jpg
9c643254a530e3fa71a4ed6330c24a7b
aa740e49b8896a6ac9e44f66fb762c2d554cddc2
594180 F20101112_AABJGJ boock_j_Page_77.jp2
120cbea04e0efe7133dcca7799037d37
b18bcb16dde7aaa69d7e2d2659e5bf3b65916f0a
525871 F20101112_AABJFV boock_j_Page_50.jp2
0ec39a3e918e786277ec9383f695f851
193a656b70e592d5a2a6ebb694ef7feb3a66d873
113934 F20101112_AABJGK boock_j_Page_81.jp2
057a06992ad60c0d4d04d563440959bf
ae58cc446f3a5156d5bbbfec065520570cbf01f0
602406 F20101112_AABJFW boock_j_Page_52.jp2
87eadf2ed354b28a5855a8ce752f36cd
47774bf4cb18f9c98c43f207329ffe5e740a3dc5
108756 F20101112_AABJGL boock_j_Page_83.jp2
cd4e37e51c496994e9026fbeda1e9b60
a926e62c1fcbf39066e603095c2c22fd428a0752
85243 F20101112_AABJFX boock_j_Page_54.jp2
8e02a645bded5993dd82cbb37879beb1
00ae52d46d1110474998704480bd730ff9458c0e
F20101112_AABJHA boock_j_Page_16.tif
5eeae560e3d5681e9e3b01c60dae382e
6e3576819eb76b7ee0b23c1b0685266aa92d61f8
106508 F20101112_AABJGM boock_j_Page_85.jp2
30e80a6c1ee2cb11cea72b62587c5771
e7c268405ada65f1c3e675f3da6376899eae431c
83823 F20101112_AABJFY boock_j_Page_57.jp2
d275a2f6e0af0a3268a5821e1ab7107c
b2e041b7ef99ef4a32de6c858d3bf5da1aa17dd3
F20101112_AABJHB boock_j_Page_17.tif
46bc154047e7d5a3ae1f83fcd3e75923
448268822415e5ce769c47b8dd1f3300a59406df
115425 F20101112_AABJGN boock_j_Page_86.jp2
eabd51acddfbc6b87db33ca73672a63f
99ede16a88501789838bcb3779d30fa93b8b5e75
896411 F20101112_AABJFZ boock_j_Page_58.jp2
2c9452bd9eece8a8366b3bea1a9f03f8
f322f40bb5f26c51ad3a05e4c85087f21f60c15e
F20101112_AABJHC boock_j_Page_18.tif
1b17ac01a2e890657db3e73e610656a0
64a7d1feb9828fd371ff695331e21f4cf1a0e52a
106692 F20101112_AABJGO boock_j_Page_87.jp2
fbc26f974409a7b8ebe2055d6f74f779
23ee813d00b99c6576bdafc30a34bdbfe9957fed
F20101112_AABJHD boock_j_Page_19.tif
c3be511ae9d76d76c84071bb6747c263
28f01eb475d03648d9eaa7ffbfe53df26cfb0337
111872 F20101112_AABJGP boock_j_Page_88.jp2
4525326d86ff43a3f9ed04806c274a40
05e324dea76d5a9a69f4deed711d3006739d27d2
62283 F20101112_AABJGQ boock_j_Page_91.jp2
516fb94947c1c211785e4cf69c26500d
9122b718511878bf0b8281ee3101afae48c219a7
F20101112_AABJHE boock_j_Page_21.tif
9615276e6db40bb7c6b12ac645d9c3d7
6cd7489e2e3677215c59b6156ce56c6a16b603c0
F20101112_AABJHF boock_j_Page_22.tif
da79af7cc722da0e4f55a8102bb46042
eb091f4313d7eb0e18fb0aca1bed63b25a166f26
125650 F20101112_AABJGR boock_j_Page_93.jp2
b74164458727d88ef3759b115b62a754
e86ac9e47fdb5d7d7c3037a8d88c27fd32c6d20d
F20101112_AABJHG boock_j_Page_23.tif
51d8f9405354e7fb37ef5bf218beaa1a
a76df561aa5354a5a62adf2de01374486271e842
132280 F20101112_AABJGS boock_j_Page_94.jp2
9ab9f6595f3136661b84bedb635ad605
0fedcf64ed5728763333e865217b0948893ecc7a
F20101112_AABJHH boock_j_Page_24.tif
75ef7e16ed1181896e5feb37022d717f
c7ad21f434c92b1acc41b353634e189160223e6b
15476 F20101112_AABJGT boock_j_Page_95.jp2
9312f5dfcbf93d34cc44b9e3ccc7fd9a
5eda2b6054e2ecb1087a595e13048c4424675f0c
F20101112_AABJHI boock_j_Page_27.tif
4293330b49d640f6d6825708aa330dd4
efaef0015f64735cae5beecfd78db695d38419e2
59105 F20101112_AABJGU boock_j_Page_96.jp2
e6c829f189c2c9589f478805f874e535
1529829a22b0b2551a50eb72058487130edeac4a
F20101112_AABJHJ boock_j_Page_28.tif
e571effeef005f3b8dafddd78ccd408e
7c0ca5846f38bffb3c8ca61b97712dcca741cbdc
F20101112_AABJGV boock_j_Page_03.tif
19a9ae8e58e618285b46236e99e9d23d
91899956979623ae6a59da2ee50d3d384e5089fd
F20101112_AABJHK boock_j_Page_29.tif
6189dc4c47802ee418158bb41de48916
66ea652bb2b4da0efed4d1cb3f538bc00f404133
F20101112_AABJGW boock_j_Page_04.tif
dff030acb02a05c8854622237c938f7d
2cd3da314b49ea6d61575583572c3d45801a71eb
F20101112_AABJIA boock_j_Page_55.tif
667163f842885927add79aa34246f718
2612974e99b3800805df50a254351e7a01b93571
F20101112_AABJHL boock_j_Page_30.tif
936f8a261b0cf2bbbc710ebb3e45d1f0
c0b3f745c40109e1bd91e0a24bb9f9a6d4baa8a6
F20101112_AABJGX boock_j_Page_06.tif
f822c4ad2ef586d809f9fa3c6812a29b
f14e4163e93af9329088f971e4c8dc7004a6d091
F20101112_AABJIB boock_j_Page_56.tif
066865363bce1ca91ad8c2b5d19124b9
02feb9b504dba50ce3e466b49453c3de51493fb7
F20101112_AABJHM boock_j_Page_31.tif
d246ebef3385f788fa7ab8bad7ef64b0
01bd7552b902c0472284fe17adb7d6bb037f570c
F20101112_AABJGY boock_j_Page_10.tif
093340630c5a1eda9b30d867cdb02ac6
d20be9bb12eed539efd365c9c91eeab31be0a751
F20101112_AABJIC boock_j_Page_58.tif
127bb88b45e5581323e68e1904a57269
9edc78a8d799f1a1d79de361c9361a229152edef
F20101112_AABJHN boock_j_Page_32.tif
3d911bbfc8dfe7065355c6858019bb8f
61259e682c9f8021a51fe528dea1a9cc825e5537
F20101112_AABJGZ boock_j_Page_13.tif
eea1db2785a90de8ba7f7ff7bf43b488
0d754962f3ef7468812c924812eae8d16bfaa572
F20101112_AABJID boock_j_Page_59.tif
2be7850e5fbd67885ae4068b988058d9
eb9f75e7e55b6b70dc7620e0915db82908372e5d
F20101112_AABJHO boock_j_Page_35.tif
d8027be5451ccaf98bc630495d21c9b7
051749961e25a9236176525a2f90b422b1a7b77f
F20101112_AABJIE boock_j_Page_60.tif
7b2a403e9324c8926b8450642ab882fa
c454bdb04dc4d9ce3034467e8d80dfeac62e6485
F20101112_AABJHP boock_j_Page_36.tif
9e3c494db229bd3d1fe5358ad8990e56
44d9732ee6729a3516a60d952030c8df36548105
F20101112_AABJHQ boock_j_Page_37.tif
f860f9ed11719d4e190afd00535be013
2c902de94f94adcd2a35a8e0da5cce26dbbe06b3
F20101112_AABJIF boock_j_Page_62.tif
6f54988748e0f3077e27362b8391ade3
66ff3149d65bfe5a79936be62b6ac2196bbae544
F20101112_AABJHR boock_j_Page_38.tif
2f9e947a2d37fc3a2b75718106440c7b
1a7476ad30d7dff0f182393a6276d832c9dd2886
F20101112_AABJIG boock_j_Page_63.tif
a9b42db9c5a6f8d026a95f3cb056a357
20245f0f4ab1d5fdac889255c44f24f510efb10a
F20101112_AABJHS boock_j_Page_39.tif
0eed744378499de8bf0b215204ce0c0e
ce9655cbaee2ba6346f24f697826bc5f9c340379
F20101112_AABJIH boock_j_Page_65.tif
43c4814638a9c2ea2ce53fa0a439e131
f9a6e749e255e956eae8b9e9c98da730b71cad16
F20101112_AABJHT boock_j_Page_41.tif
9d033fbfde50573c907f525b85b9c8ab
64c0a43a576cb51ce94c87f8e2e51f5593c7c110
F20101112_AABJII boock_j_Page_68.tif
9fd35c24b92e8c882c7c0aae2b276326
1ade37f0a141376f533ca774db2460bb89d55037
F20101112_AABJHU boock_j_Page_43.tif
1791149b9bed280dad0c5886bae99393
f7fc0b6b638026b28381216e2e72f6b5ac93a379
F20101112_AABJIJ boock_j_Page_70.tif
8d4758c457a1c1a6d4d244ca192d8afe
097e3f6d260a85a97f2a0ed7702f28b217c33ab8
F20101112_AABJHV boock_j_Page_44.tif
c0bef28b5efcae36b169e690d2bdd659
8f78bd0c9a339920f25692c09356e32ab2418324
F20101112_AABJIK boock_j_Page_73.tif
850e873461e3aafa700f5605a8954028
f1a8c5a515af3d293f4e9914a903de78176d63e6
F20101112_AABJHW boock_j_Page_45.tif
5438dc0028886747e669fe28a0f4d968
fd745d9a61723c728d587f0ba83a7c2a8a2ccc85
F20101112_AABJIL boock_j_Page_74.tif
4c47873c9bf8d8c593df7e0750ac8f90
7b80457abc2daf72dfaea78f18d49d0c70b0d1be
F20101112_AABJHX boock_j_Page_46.tif
0f50f5ff59fd6d921083db7b62dc497b
3d6fc1393363e4c3eab6bbc0e3c9bf40c0455956
F20101112_AABJJA boock_j_Page_96.tif
8129b2c5224410a48beb6cb081e6b10b
d0cf26e05cfa994d1548fabbb6dd5fbed3dadd7d
F20101112_AABJIM boock_j_Page_75.tif
3ad8ffb20a3db5956deaca3af07fd47c
b68b95e0bc651b7363ae454cf91c6a05e5d34abf
F20101112_AABJHY boock_j_Page_48.tif
f16c0fa7016fa61a1690a4f60a196c0d
7de874f7a4e3ecd76176c02e3cc4d714695ef1c7
828 F20101112_AABJJB boock_j_Page_02.pro
9435a3643edbc1f4acc75a26e24d0390
14934c7510478d8c7c61811e5b8e6f20fdb45203
F20101112_AABJIN boock_j_Page_76.tif
d2fdbe339d1764ce3ecb4671117108b7
a7be9a52d584e15fe5961642029a7dfd15ded0c7
F20101112_AABJHZ boock_j_Page_53.tif
ea74c4fc32ddd06aeb5f0bcfd178dc6b
90c03ef8ca788a659fa51016f6301bed8af054b6
25668 F20101112_AABJJC boock_j_Page_04.pro
ace38b6431073d62e0b444b15508d85c
d44f75a1b775f5b5b801c544a921f9c24c2ce33a
F20101112_AABJIO boock_j_Page_77.tif
eb620b8c26a6b9c7391bf6ddd700b0cb
d2e52c6f573edc9c85c4d57e1cbb51e1efb284af
72577 F20101112_AABJJD boock_j_Page_05.pro
1a66df10e511f345176d67f7af218ee8
6e12047a24754ec7e4a6345cf9d48d4666d546a9
F20101112_AABJIP boock_j_Page_82.tif
b5058f75c1a17590b7d8f55222c58cd2
74595c9fe88ab050cc172de9882f5e5650ca0862
67504 F20101112_AABJJE boock_j_Page_06.pro
ea7443c0628808a62bd02e250f1ad1fb
5e5af8acab78f895251808eccd9f21b667795673
F20101112_AABJIQ boock_j_Page_84.tif
4f74aa41a2270ebe3856b77b25f3c762
0d728bedb2d4f22bae36ea0bd218bfbac90b1292
56126 F20101112_AABJJF boock_j_Page_08.pro
a64c5b682b5e665d6ebd6550401289d5
0cc9c0c2b9bfcd3accf3c1411f44497d598d181a
F20101112_AABJIR boock_j_Page_85.tif
1b2ca86c952119b42f3686e6d69b1f4b
665f64de907db26a3f2c7d8321b1fdbbda9da9c0
F20101112_AABJIS boock_j_Page_86.tif
6b1a972cb4fb84f778ebef9c8fe14014
ba7068960ff171b83d8585cdc9cbed35c3815084
58096 F20101112_AABJJG boock_j_Page_09.pro
e4a9d6498ccf4b950c716cb4a244807f
8e24c5b675ededfa3a0f75e012e9e9d7abaadf05
F20101112_AABJIT boock_j_Page_87.tif
c7d08d76b6af56149617a403a0d9e760
687dacd9378a83534f3dd37aabcd8c606a51da7b
9317 F20101112_AABJJH boock_j_Page_10.pro
91bb35b8a8e3cc693680a3d9696251a8
10f0b4a05ec15199a3481fa13d84c79b233b5a6b
F20101112_AABJIU boock_j_Page_88.tif
dca712c7c82f7300bfb313d8e22b3847
e4bc2de3e80e0754623b895da9af36e895a1b1bd
17616 F20101112_AABJJI boock_j_Page_11.pro
5d1b5e3243a10a3f3d327e41194621c4
d67eb66b3f2894938cae761e136d7cd9f0308659
F20101112_AABJIV boock_j_Page_90.tif
911343965a176b461fffea475496b388
45acaa278e2262fb35db64632684dcc5c4aa295b
38029 F20101112_AABJJJ boock_j_Page_12.pro
e06a56cb2726b3648892589c3432bc30
bdd10f76812ad749c4b7304d2109698350d5ce8d
F20101112_AABJIW boock_j_Page_91.tif
60dbc1854a4a883ab4fffc10129b59f8
ae10270be61dba439460491c3cdf349113e3697c
52045 F20101112_AABJJK boock_j_Page_13.pro
d2f0205b1c79efef4a8840d2f8724c7d
931cab1dcd07a3e4ffbaf72846df4a3de3868e67
F20101112_AABJIX boock_j_Page_92.tif
61f87d233175ba05ac86bd5deec3c804
deed8ebde63f752597dfb0e81f7f694a54b10b74
25194 F20101112_AABJKA boock_j_Page_46.pro
9198b463f5435abf9e72da187a067671
6b93704a563206ff7ff693042a893213793ff658
50440 F20101112_AABJJL boock_j_Page_18.pro
e0b8f647fb601c392d4f660cc3e41bdf
cf692417fc5bce2ee85ac928c2f5604ca3e57f89
F20101112_AABJIY boock_j_Page_93.tif
1ea9960520410b0e37b4a3e9cd3c6b79
976219bb7d23ac4074e18689c46b6fd974df8275
19575 F20101112_AABJKB boock_j_Page_47.pro
b963397f7400fd3f5d20d16211f87560
e5593e5770428cd9dfcbb7b52a46aca57efb3574
20567 F20101112_AABJJM boock_j_Page_19.pro
bef49ed7c9b528c3c9b798bd41136113
60587af916e284182ab41d1d8153ca44e2eb6a85
F20101112_AABJIZ boock_j_Page_95.tif
b2816d23cf2ce4c87c4f300f634641d5
06f7256216c123e42adafa6806d8dbd414d9ddda
22703 F20101112_AABJKC boock_j_Page_51.pro
2144d594a15187328439bb5d848f48b7
851d60f31eec4ebb6ac0a6a21d1053f844f68754
26782 F20101112_AABJJN boock_j_Page_20.pro
d9ca3318d133224c0d6baa733797dd8d
ac71b05b24f0a30a873b46160f296938a5a9b021
23359 F20101112_AABJKD boock_j_Page_52.pro
197a3fd5d1827c939edd1c6e99a80ecc
8aff31c463b0d26e5122d001906d2b8814250d32
35034 F20101112_AABJJO boock_j_Page_24.pro
cb774d81695d65696fd8db419bf9c5b6
539b368ba424d334c3b81347eec4a234c52c5767
53177 F20101112_AABJKE boock_j_Page_53.pro
9a573067060b56e7991219bb77d461cb
fae76cf5faab9d632d8df57d46497e0dd6367c7e
21398 F20101112_AABJJP boock_j_Page_25.pro
93151c02ebfa8d58fbb836c8eb26ebb3
a821babd40d8eb8fc9db46653f4602516e00230a
9136 F20101112_AABJKF boock_j_Page_55.pro
34ccddfbfc73bc5f66db92240b669258
77a946e3ceef882c71d5f73906d6c755b8ad7832
56728 F20101112_AABJJQ boock_j_Page_28.pro
67d32435a07225b7b97c2b8762ff1aeb
c91acac36fa74271f814c64e972fbd089627b1c7
25907 F20101112_AABJKG boock_j_Page_56.pro
876bed9f26c324da7f0c196e833a45a6
ecc1b37d3a6484c6314c6fd4f178ebfa05e25c34
50780 F20101112_AABJJR boock_j_Page_29.pro
06ec8fe6381861e79384bf4c2861d372
6491953569c0f8c2ed458efdba22cec3f51f0609
14962 F20101112_AABJJS boock_j_Page_33.pro
fc8e3233aeb2ad9cde87b21c8a667529
06b8ed9ca874b0d99ffb4f31e760f237182436c8
13662 F20101112_AABJKH boock_j_Page_58.pro
d32275ba6538ff6615f8dd4d4f945eb0
8d871db1b2cd10bdeec68985b3d3543299e8bddc
19207 F20101112_AABJJT boock_j_Page_34.pro
d3131aaa3e9ed10e33f2e1b7be131501
de0119ab250ae1bb49ec374d478f4fdd771abf67
55431 F20101112_AABJKI boock_j_Page_59.pro
9fa42874a1d62e3383727ae6d2bc14be
c5f82841adfa3902dbd1e7320ea27353cf00cd07
18544 F20101112_AABJJU boock_j_Page_36.pro
477bea9cbc07d5e025c22f00bd75c9d5
bd33df4d93c2c65850b225e0eeeeab7dc6081422
28542 F20101112_AABJKJ boock_j_Page_60.pro
ed4b010b35a85b9b5e2e308b6b89934a
94c3f15f58469eaf07cb3ca1b002ae8cd6ffe136
18618 F20101112_AABJJV boock_j_Page_37.pro
b7102eb148fd35e33263c4a6fc329166
ff8ef0a8f75afd655eb5475da0f971b2644d9915
6915 F20101112_AABJKK boock_j_Page_61.pro
4b4cd1f0bbded306abd5766a89e65f37
50f7fd096b2933c0a12e474acde2700dbe976e4d
17642 F20101112_AABJJW boock_j_Page_38.pro
665cfe42e667e9e7500510b8ac5105d9
db67c8d7496ee368367f8acca3f552bf20a808ef
27345 F20101112_AABJLA boock_j_Page_91.pro
30fe8066f8ced89e2fdc02824cabd98f
946c9ab0132498a9a044d1431004e5c90f80b605
53222 F20101112_AABJKL boock_j_Page_63.pro
41e796f60cbc2f412e9b4668a5facd03
4d5726630d30787709039aa96ccd60b6e856be63
16423 F20101112_AABJJX boock_j_Page_39.pro
467fa46043889210b59325217fc655df
3e3f5f98c1ca575a83db7c4c5c720274e55cd0aa
60205 F20101112_AABJLB boock_j_Page_92.pro
0dfb2742cb2bb2c44b7911a0d76abd60
1acdab7426a4242859c3d3047e6fc49b7814237b
42092 F20101112_AABJKM boock_j_Page_70.pro
e56e16b9534ba82aabe871e66feda41d
77f73e6727c88cb7b70a4ca00e9e34914ddb5c5f
9188 F20101112_AABJJY boock_j_Page_42.pro
316f9ce5517fff2b187b7e9a36c1e2aa
496223f9f046ceddbea03a1691e3b30b35fa97fe
57544 F20101112_AABJLC boock_j_Page_93.pro
3bbd052b59a4c8455ca3b85cb0c6f39a
b5bad1a7dfdda1f310b3e922679e2f75a3cdbdda
19923 F20101112_AABJKN boock_j_Page_72.pro
dce36d3541ef094010dfb860bd16dea7
1ff746f989b8f81fa2457f82e224186aece39f47
22668 F20101112_AABJJZ boock_j_Page_44.pro
a914b2f541aaa58a3b2e07a07cb2a006
32c1016fbb9526472146da826ffd2a43338bf263
94 F20101112_AABJLD boock_j_Page_02.txt
e45f4f2a6bb80c19b1eb26a9df73bcc0
c9ff5f15c04d12fa929c0d4c6983b3fec9391176
27000 F20101112_AABJKO boock_j_Page_73.pro
75aefd48e417350a8283e3c298606044
0adf71d29cf3f86becc374d0cc31663c737a6eb5
1062 F20101112_AABJLE boock_j_Page_04.txt
b0a62f86b7683953052bf532b4ce7f94
d23918fb75e09caf7448282571c6092b386c567b
25227 F20101112_AABJKP boock_j_Page_75.pro
ba043ceb6ff6157a8555535ef8bd09f8
be217f076ad2fc2043c0b74e6c7a3ae7c81f020a
2910 F20101112_AABJLF boock_j_Page_06.txt
169f10240b6513eb1cb2f98d17330fd6
23141dd28140a80d34be13f4743bda79b332b777
14343 F20101112_AABJKQ boock_j_Page_77.pro
a9eb513a17d48be5ae47c861bb0492fd
0501155ead8219bbfef81843d68b8d41d95296d9
2291 F20101112_AABJLG boock_j_Page_08.txt
dbdecb81a9b704f1365a84c7a6f4e991
1d99d30106844d7a4ab6e5e4837764f5864279c3
19801 F20101112_AABJKR boock_j_Page_78.pro
bd6fb7364c091c95fa97029fcc8bd244
f0677b8cd6ad8c4285d828d58bbd37b8e49630d4
408 F20101112_AABJLH boock_j_Page_10.txt
ee26e798c89d6b8289b816a7ec417b14
0243872aeba0943e295d2935031194dea8c6cce6
39574 F20101112_AABJKS boock_j_Page_79.pro
b6bbbccd818f95fb5e45939c15eb0f59
f7519eb3802a1a99135c378b60fffdc03d8023a5
38177 F20101112_AABJKT boock_j_Page_80.pro
f0b2a3772866599eb16f80f35903f0e1
e0686f2770e2e4957ec0a77666b13b7c87962fff
781 F20101112_AABJLI boock_j_Page_11.txt
11ffde2f0b6e5966a7e97b093877be66
97f6cdf3f52ba3b28a12317d37621ad2069cee48
50347 F20101112_AABJKU boock_j_Page_83.pro
015e3853c62e6a18e1177c22ffc09baf
a76057e91a28bbd2adbe952586cb2d6c92400dcd
1717 F20101112_AABJLJ boock_j_Page_12.txt
38d79686fd774580f99edb612dfb3d86
67102ef218fd24f97530bb58ae4d11685ca32238
53378 F20101112_AABJKV boock_j_Page_84.pro
2b9edb531b790a5b70289e64220a84a9
3b62184cece55683d92857a87bd3c93e1dc46830
2089 F20101112_AABJLK boock_j_Page_14.txt
63ba01b0fe24871afe2528bb856125cb
924305d37dded0f7b3ec632d9bc7d87309451742
48874 F20101112_AABJKW boock_j_Page_85.pro
e90fe5e599aa915b60659a6a9be8e8e9
a1c0b75272c0e5ba377cbdd5b7aaca56bb99984c
2183 F20101112_AABJLL boock_j_Page_16.txt
82be146765c0e06c2269500107d48db5
ca18d6602a2da52ea995c16190124aa541db898b
53522 F20101112_AABJKX boock_j_Page_86.pro
554113ff94fbc77672fdd194a94d8803
e6ea5f10dccf9de3f3b4895d61548e5fd2ab4eea
894 F20101112_AABJMA boock_j_Page_36.txt
8de44cfc960614ed697fb4a7be2e5a3d
d4acce06a743101a341de317a5b8fd321fc9f51e
1966 F20101112_AABJLM boock_j_Page_17.txt
e74bf5da066153c5b053b74646b136a1
986da59065e1a1159131ec39106499a562e49d41
49716 F20101112_AABJKY boock_j_Page_89.pro
fd275ae0ad2be406d16d1d842b1e23af
6e754f31c421516feb522484d45bd7300815571a
944 F20101112_AABJMB boock_j_Page_38.txt
cddbebca63fb930b646052fae19336ee
f6d0aa583be25a03e6cf859301d879335a8e5376
2033 F20101112_AABJLN boock_j_Page_18.txt
65097c34c13f9ed1d90f1d86a3768bb4
804da51b770d3a0d21738ad8a9db4e3008d56956
37151 F20101112_AABJKZ boock_j_Page_90.pro
aa63f256d8bc0292469d85b6a592707d
cfdb3f9f5cfff75a939fabe53f11927b4721b2b0
878 F20101112_AABJMC boock_j_Page_39.txt
b7d42d5f4d6e7bc5d695479c6cca8ff1
8d6c2395801c404bc603800c21e4b7a247b6d255
414 F20101112_AABJMD boock_j_Page_42.txt
278b492c4ca80ae515bd8b29884bda4f
1ccc11ff092a2aad7f59c0f88d3da0f8f8a5b1f6
1139 F20101112_AABJLO boock_j_Page_19.txt
5623d3c8683195f4c807269af9f51854
b847a7ef261f2641f31d6ba291aaefe5c2e5ba30
2059 F20101112_AABJME boock_j_Page_43.txt
9e43d2e6afb152e6a1953e8a614f8fab
a67c8331e0482eaeec457828b3c9a81ba4229244
1298 F20101112_AABJLP boock_j_Page_20.txt
acf8248d10e39cd03e1a69c2be92415e
b5f549ac2458bc612bb32c8d05002d54d453ce6b
1076 F20101112_AABJMF boock_j_Page_44.txt
2721b11a0eabd9604b40684d2589aee8
a079f1a4689ff45eae50c6a851e31714249b8838
1579 F20101112_AABJLQ boock_j_Page_23.txt
214aeac235184aea5c3472cde80cd23a
0c50813629a7d80abf99ddf89daec9d2553688ce
889 F20101112_AABJMG boock_j_Page_45.txt
08f6dd9f198ea6e23ca9c308996327b4
1f1a98a8c70331596bd8cb6422638acfe7e33804
1454 F20101112_AABJLR boock_j_Page_24.txt
3f5a5e988a24a30dc58455cfe5271ab2
d9fd3925d29c5baa7b71fc227b3dcd6a30e2d6ae
1148 F20101112_AABJMH boock_j_Page_46.txt
dab01d2f1264d3e8778f1bf8a33ac55e
1a9b261e2c619cc97eb40651067e0293be062ff2
897 F20101112_AABJLS boock_j_Page_25.txt
63572d51ac3c7b35c5b1dba23d0f7926
9953917acc89ed053d05d8e5e91748db4403d310
1014 F20101112_AABJMI boock_j_Page_47.txt
3a98ec263b4dc2842ea54af86dd1cbce
a519b0f015e35c059e72711a6bd15792e47f9a2c
1504 F20101112_AABJLT boock_j_Page_26.txt
f79c0917a01f3b182c4f478414a54c31
7c865f9316003cc67e11d102e5900f96a352d67f
1972 F20101112_AABJLU boock_j_Page_27.txt
88e6215f7047104384c624d42ce24126
83008483bb1647105de39cfee0b7857cb0e10ef9
650 F20101112_AABJMJ boock_j_Page_48.txt
c2810581be7f2e86f5cc2a15808f946c
531f7e314acb11d13bdaa2a6379d3dae41e1eee8
2008 F20101112_AABJLV boock_j_Page_30.txt
546451567a513ad3d0b651f19ffa4202
344f0abf67846561ab41a914dfd0d26a902458b1
729 F20101112_AABJMK boock_j_Page_49.txt
2f671277499f58119db898ee69a0e783
79224baa21926c28737cf2d816159bcc5b550c66
1009 F20101112_AABJLW boock_j_Page_31.txt
6f7bcc519341a95c332246b0191d35a0
7b6da6c9e1dd7bf496ceef37f92b83643a6760be
983 F20101112_AABJNA boock_j_Page_72.txt
a25e6668ba1872a1f110dadcf0e89d4e
e3795f5a46cd24952140842270437572864db21f
1044 F20101112_AABJML boock_j_Page_50.txt
0a1ba17b79f4e1521fdac9bdeb373726
c8358d4392fe8eeb6b165d7f94cf14908a744395
1119 F20101112_AABJLX boock_j_Page_32.txt
41abbcd6d7271d463fe79b5ea8eed395
0ae863d27f8f1e270db5d746de4c61fba0fee6e0
1288 F20101112_AABJNB boock_j_Page_73.txt
168438ef256ab04f48be94465e632c87
1cd4c5c8223826d2357ff7580fb7d682d8f7ab47
F20101112_AABJMM boock_j_Page_51.txt
8d3432877c1cf21618e84771e9de1555
daa4504c512e5ed873e93d218bb200f26533a967
1169 F20101112_AABJLY boock_j_Page_34.txt
ffc3f995bf5d9e82e8b848074b4d0f31
11d5b82ce58bcebb06f3df1549cc84c750434c4d
746 F20101112_AABJNC boock_j_Page_74.txt
29416606c0c65521eaf710b50f2cdd6e
486a6e30a4e6e0d49c3a5638d02e80092daa9fb2
1527 F20101112_AABJMN boock_j_Page_54.txt
02b4001f78a89b915553e3b156b7243d
a7a1e4935c746641ff9f3e3a756c2c993291bf6c
1073 F20101112_AABJLZ boock_j_Page_35.txt
df3ee0d6417554bd96e7cd64b8e85fb5
78511922864cf5b4555b25b7aa021c72a2300b1f
623 F20101112_AABJND boock_j_Page_77.txt
f53c721dd8058762965c88aea8499e50
7916dbd8b67a7458b2dae8d53669347234e2ae5b
406 F20101112_AABJMO boock_j_Page_55.txt
78ea7bed49ff7e9931af99d64ace8ac3
629798a4fd276c58fe133f3a7f17c40f6a8ead85
936 F20101112_AABJNE boock_j_Page_78.txt
659695b7e4e73247b8a6810c8d451001
19b18c08db100295d9a2b990a5a46f234c508972
1223 F20101112_AABJMP boock_j_Page_56.txt
50b9019f2327d95356236daa2cf0c3df
57207b20aa848a92b45c0731018d176454f60c1b
1537 F20101112_AABJNF boock_j_Page_80.txt
0ac5c3a2787a1aceec2d3e3f562c27a1
57e19bacf393675e291baba47b4cc3ce965b2984
1627 F20101112_AABJMQ boock_j_Page_57.txt
c04827851b3ffae73edc8444548cf16f
046f00afe23f089e73e4e9711799fd5e16760a90
2140 F20101112_AABJNG boock_j_Page_81.txt
d87e29162c0c409fd4ed69fc42a39153
7fc6bd48beaa6bc807051823fbc4fe05e2c70d33
2185 F20101112_AABJMR boock_j_Page_59.txt
ea9100e8e360794740a44454cad8bec4
ceb52539649e9adf614d5b579db65dfd9b4ffc3a
2096 F20101112_AABJNH boock_j_Page_83.txt
e7ad7233d14c00c217f9308d2934f2d6
ec2affbcaf9c79c631a9127017af5d0286e77068
1152 F20101112_AABJMS boock_j_Page_62.txt
1122f415e581a47fa5ef0b8096d8d6d7
01dc34e501b8b3b3ed31c3397f80c7a1217d6257
2103 F20101112_AABJNI boock_j_Page_84.txt
6d4c78b908c410fe62cc3515611aa96f
65514b90192c20a872e809e3f4450503a2d7278c
2152 F20101112_AABJMT boock_j_Page_63.txt
fb683598f53478aa47eae3c07420f796
cc31dbf1575c733c8b41b61584155e3de17d7fdf
1944 F20101112_AABJNJ boock_j_Page_85.txt
ff22995c34da353c4123c91f469fd941
bb3d7c8200b4464fa3af923ccd68ff63e89c1440
1398 F20101112_AABJMU boock_j_Page_64.txt
23edaae56ffc92c3ea329446367be4ed
41dbb38b4e7541e6a4dfee122c42ae1de233777f
843 F20101112_AABJMV boock_j_Page_66.txt
48c98ce0e46f6e35ac4cf1f9bc626e56
c206e38b924923cf0debcf4e403db7c579dde883
2107 F20101112_AABJNK boock_j_Page_86.txt
f7fb586654e45d217dfa146acabcb982
d98f4726aa894b23df3f23cae098d020b5fcf550
1856 F20101112_AABJMW boock_j_Page_68.txt
ce6352f2026231a23e717dfa0556644e
f95f72a1c43ff59eb47bc0ef2513ab54132314b8
1989 F20101112_AABJNL boock_j_Page_87.txt
9b27f6759f66f07dd65c4785cc7afc6d
cbb8f1afd67712bd52efba38d8470285953d76da
F20101112_AABJMX boock_j_Page_69.txt
a6929c54f14181d6296d9f6b1bd44abc
41b1b60b1a0daa039b8839a721674d3da6726ab0
4947 F20101112_AABJOA boock_j_Page_12thm.jpg
8a24acc57a343f25b6683f4770f57e56
b2746a2bfd5051fa10c69a1ba912717a9ca4aa69
2082 F20101112_AABJNM boock_j_Page_88.txt
52bbb545037b7ff5c936e2a9ee3f6652
d0a68e64f129eab1e66c6d80808fe07ff814bb49
1713 F20101112_AABJMY boock_j_Page_70.txt
071515dce9008d58437b46b6b4d4a046
9b6f6d46e3a10243dcc717625fa5c4b718be72c9
1374 F20101112_AABJOB boock_j_Page_10thm.jpg
cf54a717116d025554f317168b79b5c3
607e925b842b3bb688998cf2dfc9f5fb8d82b3df
1099 F20101112_AABJNN boock_j_Page_91.txt
4d66a01e6a1861f494fbbe6a556c081a
93a933735f207ebabfd72c84ec260fc5c22af74c
1201 F20101112_AABJMZ boock_j_Page_71.txt
b9d3bddc1c0dd885c6a3f2d35f8a2842
da48abc4d6760f2a6d9e08b87afd3496ac9d9e9b
5469 F20101112_AABJOC boock_j_Page_70thm.jpg
99e74b4ce27fbfe141e9d9d4837e1d44
0fc37bc7a0891eba4f97682ac750a0c5c0c2fedf
2328 F20101112_AABJNO boock_j_Page_94.txt
78523e298a9dd73ced0d2c00a7fd3daa
e9d2fa5ded1374c28a8482989e54073484525582
11514 F20101112_AABJOD boock_j_Page_34.QC.jpg
590126cd6f5d4a120e57c53a67e903f6
4450eb675d8b095b6b13888fa3422ab57aaa4aac
26104 F20101112_AABJNP boock_j_Page_85.QC.jpg
5131a81132f4e30597b17aafb1256a8e
1dca75db192f8288a4d019f16f6bd440648a1043
6487 F20101112_AABJOE boock_j_Page_65.QC.jpg
de97e2565ce1e327d77494007b709d88
0bdfcd324e74daf9646930a5a59ed709e7c54524
14374 F20101112_AABJNQ boock_j_Page_44.QC.jpg
6808a532d165bdb34b6095cf5a4ed1ff
6106709f3236f0c5cb81f452427d435db51b4f99
21513 F20101112_AABJOF boock_j_Page_24.QC.jpg
17919ae1baef3a770c293c4c90f1a1ad
4cdbad231b80207cd25492355cf11d9563b6223b
4341 F20101112_AABJNR boock_j_Page_71thm.jpg
341661dfaec97d709cbd085cb831ba7a
60e45a2652e0993d4a8e68fde8a3c895f00db7eb
3730 F20101112_AABJOG boock_j_Page_96thm.jpg
bfe5b3a5779f118bd45f3fde26638adf
2f6f868d3e4b308c1f7fe9adf808e34e427e724c
16001 F20101112_AABJNS boock_j_Page_71.QC.jpg
18ad0b4ed90357b4d0eedac79b74abd2
eab77fc25cd4c191916eead43ca847675caec625
6174 F20101112_AABJOH boock_j_Page_19thm.jpg
e542e778b789e970e17e7dca1d9e5274
bf1560f039d65f7967445d259685e6abc99fcab2
16537 F20101112_AABJNT boock_j_Page_82.QC.jpg
a3e0c476978b98a0b4211018eb73b614
67518ea42faa4e74f0fbf62d66fa607a2db053d9
27115 F20101112_AABJOI boock_j_Page_14.QC.jpg
542649c0a848a1c44ddbddce4511e24b
6e57499f96f1a82b6b41c92dc597ea1c9ccc8f18
15984 F20101112_AABJNU boock_j_Page_91.QC.jpg
547a5eecb420f13270c09bc2081afed4
f6a08c7745be97f0c0e100aa0b345637da654106
1056 F20101112_AABJOJ boock_j_Page_03.QC.jpg
02e8a3b42a590b5e310213fa67d40ba8
6c07423276ff7c3f9c93a211dbcb213564d1da55
28191 F20101112_AABJNV boock_j_Page_21.QC.jpg
c62fb521ef7d6de9b0ff77027f61102e
4f995dd61e26310a3c9676af6cce0e795278b664
1829 F20101112_AABJOK boock_j_Page_01thm.jpg
7e2f97dc1657954418a640b381353921
ab4d9487e3d109f089e534148d3fb450e9595c87
12722 F20101112_AABJNW boock_j_Page_61.QC.jpg
a0f06d3141f40fa9fc10eddedc5350c5
0ece9eb597424fe734963aca9eecb36f41d7b616
3087 F20101112_AABJNX boock_j_Page_74thm.jpg
46ea1cf61f79a0fc5bc1a3bd560fcd53
2dbcd3fc3e956763e66efe557fb4b60791351cef
28594 F20101112_AABJPA boock_j_Page_84.QC.jpg
19bd48e726a6a839c3b09ed8b0e2327d
8acdfa2f1bab9e09c27c141d16d1f2c4cc28905e
3412 F20101112_AABJOL boock_j_Page_36thm.jpg
1c7899ac83734575e11d1a32911c0222
9b54ac6495e0b3df58e8670461adfd2ed4fcd831
6545 F20101112_AABJNY boock_j_Page_53thm.jpg
80002803d1dcab1cb4ce0aa5d61da236
a097779f5562cc2b4680dfb10bea569ddfd2ce64
27402 F20101112_AABJPB boock_j_Page_81.QC.jpg
4b7b5c16734007e38113cfe09959104a
523beea9ec5c8877121f3abd7b97d6205267ca06
18389 F20101112_AABJOM boock_j_Page_58.QC.jpg
7d945d4c207f5d2b941f3493797179db
59318a70ab79e399f9f41063e8391484e5852eee
3571 F20101112_AABJNZ boock_j_Page_95.QC.jpg
96545db1b977762698304fc60c5f61ed
fd32e91743bb102186122f9f1f65e5289143f0d6
3667 F20101112_AABJPC boock_j_Page_37thm.jpg
7cc61dea4a2bdf909c9d7bd4b78adb40
d22efdcb605e51c91f740cf50e9611431317b234
12521 F20101112_AABJON boock_j_Page_45.QC.jpg
965198ce08d6bf6e11ce05a7dcb3d0d9
7fcf669b482fae67e443f768ff6986eb6c276999
3062 F20101112_AABJPD boock_j_Page_34thm.jpg
1253afd075a411760030953c84f05683
2534515522a628782c86b781d30932fb0c2becea
908 F20101112_AABJOO boock_j_Page_95thm.jpg
ca551bdeb43fcbc22ce23725fc052d46
676be66fac7e5192fb57b79126105a4c1e9e7535
24711 F20101112_AABJPE boock_j_Page_79.QC.jpg
dcbde190ece1fd6823bd4bf75c553178
22b111da1b672b088aeac77b2c1c75651ef57c65
4771 F20101112_AABJOP boock_j_Page_46thm.jpg
593ba64d2547702bf5803fc927fd7235
53dd2fb01b6c18ea3fe9c9d290c7fdbbcc4b9cfb
4906 F20101112_AABJPF boock_j_Page_51thm.jpg
8e755ea231790255b955d605559615a6
45593fa8c73d2e46b38e3d1e8ea7b26b5d332bd1
5723 F20101112_AABJOQ boock_j_Page_27thm.jpg
039472a1db830eb16a42b85eb9180bd2
05dbe9cdd4f8d13b2aec25da3e197fa0cc524a96
6347 F20101112_AABJPG boock_j_Page_30thm.jpg
3faade9f3216e2fbab35c0ad390211ed
7828acab7b0028216506ffd439d8aaa733113324
3693 F20101112_AABJOR boock_j_Page_61thm.jpg
21e8d4c017ae9f5b1e933e9f5130ff38
e5f83e2f2ee493fc173661d9fb3f4e3ad15e5237
25792 F20101112_AABJPH boock_j_Page_29.QC.jpg
b58ec1ad7b97ea6388de4c5372913ee7
ce443093e60d7a1b4b00a2e283289ff8201315b8
8872 F20101112_AABJOS boock_j_Page_66.QC.jpg
abbea456e8c825536f659ae674be5205
67c0223b6bda9851428b1bb31230843b4acbdcdc
4259 F20101112_AABJPI boock_j_Page_07.QC.jpg
acd7c1b765f6bd5b7954a2b17e2dd776
d7d48dac66b776bb08972de6102d8bd3b476be30
26564 F20101112_AABJOT boock_j_Page_83.QC.jpg
6dabc2f41a55642432ff18d033bd4a38
3a661b7efdfe1eb316a0f058a273465f21bd93e2
25180 F20101112_AABJPJ boock_j_Page_30.QC.jpg
1ea9401b4f92c5f08db38c446fd548bb
f27a00fb4cbf5fec3e601249a72063a268355765
5291 F20101112_AABJOU boock_j_Page_20thm.jpg
abc928dfd9bd7378fc8744af75a120f0
ea03847978551a5d9e2500f2d55597885016df44
6266 F20101112_AABJPK boock_j_Page_29thm.jpg
46554d3ee08cdb7cccd99c7945d31d39
ffa151b9bf497485ead7924a3e79c17eec1c85c5
449 F20101112_AABJOV boock_j_Page_02thm.jpg
1bea5e30387204ba99a18277ca1e6ae1
88a47b8ff0f054635c1bd78665038a6c5b6d9649
143718 F20101112_AABJPL UFE0019340_00001.xml
b7565c4c1d6df019329dadc162637177
677f7238c77b8bff874ad7ac727e75d4602a2f48
8609 F20101112_AABJOW boock_j_Page_42.QC.jpg
86fb483e6dc3e0acb535104981eba44c
8e95bc5e33673444836934c7dabd3cb70bb22524
20184 F20101112_AABJQA boock_j_Page_20.QC.jpg
0b053b4ab3ae310fd351ea06a1790785
7c55efbda0c882644372114022647ee641ebe100
5370 F20101112_AABJOX boock_j_Page_24thm.jpg
9a90fd49f0a2d8bbd2cecd2c784bd810
501245c6ea3f5dfec2f35ce5d108284f0585cb22
6590 F20101112_AABJQB boock_j_Page_21thm.jpg
7a0ab8757c032676cfd4a888f0a914ff
ce3f8dc9098535d0951080bf373297de8ad172c0
14259 F20101112_AABJPM boock_j_Page_04.QC.jpg
fedb923e785ef22d88397208dfb6e604
ebf2ffddc10e4d71afb042107162f4c426612b9d
5899 F20101112_AABJOY boock_j_Page_22thm.jpg
9e568d8abeecb876c5a6b7129a3b79a5
3390abc76eba48b301edf7f0aebcd389bf02a2f3
25728 F20101112_AABJQC boock_j_Page_22.QC.jpg
2eb446e0271a700d492b099deeab1730
2d2663f22b75edfb229fc5c6124b6e66d42a2f5a
3420 F20101112_AABJPN boock_j_Page_04thm.jpg
14e56b43313ff5cc1ce93282d28d3efc
4d6bf24483e52417ee0346479d5f73d6c7442299
6704 F20101112_AABJOZ boock_j_Page_86thm.jpg
d489bda1fc4f5b4f3e4c95562bc9aa04
d5e49aff5102da44dec1126b28326a1be671ca98
22597 F20101112_AABJQD boock_j_Page_23.QC.jpg
83b8527f6d3a5de1e712dd59d7de4c62
d3e1f27e4b4af19fa54c3c0b29f060a10fcd8542
22081 F20101112_AABJPO boock_j_Page_05.QC.jpg
76f3f2cb3e6b5d03d097979c46a49d8f
3a51435535ea2a2b21b8eeecc1b9bc3cff666fd6
21212 F20101112_AABJQE boock_j_Page_25.QC.jpg
0dc792b1f4aa3ce31de4e49fe97d6a99
424f95bfba028faa92ceafa6e46e327047158541
4459 F20101112_AABJPP boock_j_Page_06thm.jpg
9e494a248af4983f3a7dda835c817c83
3f1bad8ec35e9a889032ae6155c348cb05966393
19123 F20101112_AABJQF boock_j_Page_26.QC.jpg
46d6c32d06dcb436f640735e035b8c94
261e014e795ef8707de9d96587db524f77f5f449
1270 F20101112_AABJPQ boock_j_Page_07thm.jpg
7c88a521e4d764e334fd65ff4ea23cec
b86e4c81b0047b157f38f64549b2ed71283c834d
23158 F20101112_AABJQG boock_j_Page_27.QC.jpg
5fadd73523ea802fc259f41c5f409d6d
c863dd1b262a2f0a9abc671c8d54e935ece0cdce
5581 F20101112_AABJPR boock_j_Page_08thm.jpg
1a3d2aa658a2aeafbd2d9b5a0e2415d3
44a622cc19615032e713b08af6b8ee21e5e4d4c9
17240 F20101112_AABJQH boock_j_Page_31.QC.jpg
6ecf96c7c6951c43a62a34c534341306
538ed0bfe1c2b5f1ad41d24b13f42d70079e1a28
21683 F20101112_AABJPS boock_j_Page_09.QC.jpg
90a66c8a4357e0d62be20ac9754d02be
95eb61216c4850f66104f37e10488722236bc147
4766 F20101112_AABJQI boock_j_Page_31thm.jpg
f7a788252ea020eac6114b1aaa1253c4
420be09b0bd3c7a79e49a5adb4d00d4f2c5c0541
5330 F20101112_AABJPT boock_j_Page_09thm.jpg
8150048fb14cc6e3061fab39ee827498
f9847dc7ee3edfce5175451c78968ec92a387c29
3528 F20101112_AABJQJ boock_j_Page_33thm.jpg
b776c891f302b1e9a262f0decfcdb1f6
737a2c65cff6e9f95e774ab80150e34ddbc5ef54
4867 F20101112_AABJPU boock_j_Page_10.QC.jpg
9b6c5ce905ee7ee387bee3b628735368
1c25b8e8174732b2f1a9a38c53bbdd2ff11f6f16
12234 F20101112_AABJQK boock_j_Page_36.QC.jpg
31711be30e6942d0e10df6463de9c265
3227d6ab15257c643248299bee18908887e9e553
6378 F20101112_AABJPV boock_j_Page_13thm.jpg
24def54d7b6b9a16a998f628b8f4da01
aaf00bf545c16cd2dc6a8e26bd221b5ab51753cb
12634 F20101112_AABJQL boock_j_Page_37.QC.jpg
a12741cc2e6e5dc611968a8de1c7a5a7
a9cc7221cb093149862df214be7062d8868ca218
5740 F20101112_AABJPW boock_j_Page_15thm.jpg
539bd760694faf649f7ea07ce57e51f2
fb8a0f44a3c4e2d4c7a162d48338dbfb99efc72a
6775 F20101112_AABJPX boock_j_Page_16thm.jpg
e807081348a6fc9fdd83449ee31ca8a8
dc5408c0b45de1f03485e9ba0b591946efd1197c
31613 F20101112_AABJRA boock_j_Page_62.QC.jpg
6d83bbd6b03e1d1f6b871174f4dd044b
8968c13d03e1d74b4bd5a57bbb73c70372e34788
11015 F20101112_AABJQM boock_j_Page_38.QC.jpg
3239ecf5010bd8db9a89d2b5da0ec339
88c808da16f6fea77d4d8887294899a693c2bf29
22181 F20101112_AABJPY boock_j_Page_17.QC.jpg
5cb22832b46ab439324b2f86dad1423b
3ed8758f316f33fc5f96833ad86116a4b52481f8
28184 F20101112_AABJRB boock_j_Page_63.QC.jpg
dee448cf49752d179c11ef52d10bda78
9dc96a57ab241edf99c12dbd562131d7ea8f702e
26634 F20101112_AABJPZ boock_j_Page_18.QC.jpg
c68d902ec584e264b4eb57eaabe7e53e
4289da9cf1dd856c6a493960104d619640342eed
6616 F20101112_AABJRC boock_j_Page_63thm.jpg
92119d325d5b416f761993bd12fac764
6550556532d6910f9d75c65e70a311fd1292cd8e
3175 F20101112_AABJQN boock_j_Page_38thm.jpg
012eb673ffaf2efd796a35598a3c2874
860d1f6836a9eeda54937a233fb7b8f00eb9d72a
18822 F20101112_AABJRD boock_j_Page_64.QC.jpg
85b0758412ae0b72ed6ffa00698ddb13
b98cc9c4684dd53b1272f71128a7cf25f1953450
11393 F20101112_AABJQO boock_j_Page_39.QC.jpg
203bff7eefffd097fe5c6f4e698032e4
275deca100c9491d03b4e8599673a91698ecefe4
4808 F20101112_AABJRE boock_j_Page_64thm.jpg
4b275033a101bba7106eff0ec58ff079
5a5e3dc53e6ac30cdd71c85bd5ac9dfc3d03c640
2625 F20101112_AABJQP boock_j_Page_42thm.jpg
d320c53d62f6d855e04cfa0962611f75
9437556087bf3f9ef0494ce525baabf676a844de
18525 F20101112_AABJRF boock_j_Page_68.QC.jpg
3df3564d2520fee7a43a915004286f13
850c77cfd0478908eb6643e6d9c5fc2958cd935b
3277 F20101112_AABJQQ boock_j_Page_47thm.jpg
712a7127076554ba2901fba232194a07
7957f6e308467fad7bd9b4b96b0828ed8eec049c
22601 F20101112_AABJRG boock_j_Page_70.QC.jpg
c0fe26c6c8480147310431f087de17af
baebd92117327bfbc45fc1ed52ab6870c102e726
9965 F20101112_AABJQR boock_j_Page_48.QC.jpg
ffd13edadd5eeb4ded72e09847fb5e86
5ab34cbe43fce5c359142da9cba202deead38853
14871 F20101112_AABJRH boock_j_Page_73.QC.jpg
834d2b17f1b49329f7994df11438a44a
917998e677bc69a20f5c6a9c752503af962dc683
9768 F20101112_AABJQS boock_j_Page_49.QC.jpg
3b0c1de013a097dceac7ec79ea92ee2f
1e9b2a37388a04d66679963d485157a818b57352
4223 F20101112_AABJRI boock_j_Page_73thm.jpg
0f4027bb84ed1ebdf2e019bdace9c737
3bb1ec7537a821624e05edc86bdb14b67df4ed58
13211 F20101112_AABJQT boock_j_Page_50.QC.jpg
119f20ad83b0757ae4167e34fe793ac5
012c06e87cf33453dd6b44e9c30bbd1278541afa
10305 F20101112_AABJRJ boock_j_Page_74.QC.jpg
fc1dac9b23363160f821fd3476396476
6f682feb2ff4b70609df8b835232b947b0b0a160
3500 F20101112_AABJQU boock_j_Page_50thm.jpg
c167cd3352bb5ca76a462fcb28b19877
51ca2f941239251eb53537592673970d1789947c
3770 F20101112_AABJRK boock_j_Page_77thm.jpg
f0437a8c774acecb89247115870d33e9
c086a24ec784c372c27f93181b0b17ac453e6d9e
13872 F20101112_AABJQV boock_j_Page_52.QC.jpg
04692576d15c306603e1c08ed6d28a91
1ae461c32be7d4ae55b3891e3167b3f398bde90d
12252 F20101112_AABJRL boock_j_Page_78.QC.jpg
eac1a180ed0f062ce8065f23a63b7320
d29acea1a23503461bebcbc47ae59e445ea9a4b7
5196 F20101112_AABJQW boock_j_Page_54thm.jpg
1fe1041315f1e2862967b3dd8a28741c
69faa2529bca4b94a711eaf1d64b2d144dd8be4b
3781 F20101112_AABJSA boock_j_Page_91thm.jpg
a25a403ae32e10221187dea380baf1ff
30d02b8b0a175df3a722be904470c20aca43697d
3696 F20101112_AABJRM boock_j_Page_78thm.jpg
3eb08a94384190aea64408a061feed4f
c85ce8ac2f0834fe0a7980ae3d48249534df0c8e
14701 F20101112_AABJQX boock_j_Page_56.QC.jpg
788faf71c3ac4b3d3395c2f0db8ff4de
eb28adffad65cb2ee983ba9cf64efd2b48609bb8
26336 F20101112_AABJSB boock_j_Page_92.QC.jpg
6adaa8b7e8942d08db06a5494fdc5fae
68a0bd58174f9152984fbc5835fca61d855c69c4
24923 F20101112_AABJRN boock_j_Page_80.QC.jpg
0ac4cbc1afa41296eb038fc4dd8a2b10
5ae7c941c55fb864bf1cc61b634caf779aef2f1b
20933 F20101112_AABJQY boock_j_Page_57.QC.jpg
80ead092603112a110a1d44265797af3
90c84e8c77598ad8f2340f3f76827460b5f010ed
26723 F20101112_AABJSC boock_j_Page_93.QC.jpg
1c18408f79a748c6f3cb67a0fb75d325
b1383ea5ae00c053a249abbce1be316d045eb030
18495 F20101112_AABJQZ boock_j_Page_60.QC.jpg
af6faaa0ae6f95e346494304819e4d25
e588ee8d64e44c5128cad1a9ceb8e915984d5087
27279 F20101112_AABJSD boock_j_Page_94.QC.jpg
0dd70accae4065206c0f15318dcf3dec
f7e9ee0ed42011ab7cba7bac5d9b6a6431e02550
6151 F20101112_AABJRO boock_j_Page_80thm.jpg
4240b9fc21b59ed25805ba9fb9a56c64
c22ab92bd8c2063d2fda9d193a298bf84f334a8e
6574 F20101112_AABJSE boock_j_Page_94thm.jpg
56c6e4968017aa505e605bb006dae043
9a939908d7bf34be97a112473ed8d5f9fe4a4375
4011 F20101112_AABJRP boock_j_Page_82thm.jpg
04f737f53dff0325af9d610415b9d47d
c3f428c9fecd211ecb0f039d458d27c32c18f1e1


Permanent Link: http://ufdc.ufl.edu/UFE0019340/00001

Material Information

Title: Characterization of Commonly Encountered Explosives Using High-Field Asymmetric Waveform Ion Mobility Spectrometry Coupled with Mass Spectrometry
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0019340:00001

Permanent Link: http://ufdc.ufl.edu/UFE0019340/00001

Material Information

Title: Characterization of Commonly Encountered Explosives Using High-Field Asymmetric Waveform Ion Mobility Spectrometry Coupled with Mass Spectrometry
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0019340:00001


This item has the following downloads:


Full Text





CHARACTERIZATION OF COMMONLY ENCOUNTERED EXPLOSIVES USING
HIGH-FIELD ASYMMETRIC WAVEFORM ION MOBILITY SPECTROMETRY
COUPLED WITH MASS SPECTROMETRY
















BY

JARED J. BOOCK


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2007





































O 2007 Jared J. Boock


































To my family and friends.









ACKNOWLEDGMENTS

I wish to thank my research director, Dr. Richard A. Yost, for teaching me a great deal

about the fundamentals of mass spectrometry and providing guidance throughout my research. I

would also like to thank Dr. Matthew Pollard at the Dr. Herbert H. Hill laboratory at Washington

State University for providing two opportunities for me to conduct research on their instruments,

and for providing invaluable assistance and conferring with me on the science. I thank lon

Metrics, Inc., of San Diego, CA, for funding these trips. I would like to recognize the members

of my committee: Dr. David H. Powell and Dr. John R. Eyler. I would like to thank the US Air

Force Institute of Technology Civilian Institutions program for giving me this opportunity.

Thanks also go to my colleagues in the Yost group for their support.

Lastly, I would like to thank my parents and my friends, without whom I would have gone

insane several times during this year and a half.












TABLE OF CONTENTS


page

ACKNOWLEDGMENT S ............ _...... ._ ...............4....


LI ST OF T ABLE S ............_...... ._ ...............7....


LIST OF FIGURES .............. ...............8.....


LIST OF ABBREVIATIONS ............_ ..... ..__ ...............11...


AB S TRAC T ............._. .......... ..............._ 12...


CHAPTER


1 INTRODUCTION AND INSTRUMENTATION ................ ...............13................


1.1 Introducti on ............... ... ... ...... ........... ............1
1.2 Atmospheric Pressure Chemical lonization................ .. ...............1
1.3 High-field Asymmetric-Waveform lon Mobility Spectrometry............... ..............1
1.4 Ion Optics............... ....... .. .............2
1.5 Quadrupole lon Trap Mass Spectrometry............... ..............2
1.6 Multidimensional Mass Spectrometry ................. ...............26........... ...
1.7 Thesis Overview .............. ...............26....


2 PROPERTIES AND CHARACTERIZATION OF EXPLO SIVES .............. ...............27


2. 1 Explosives that were Characterized ............ ......__ ...............27.
2.2 Classification of Explosives .............. ...............27....
2.3 Chemical Properties of Explosives ................. ...............28........... ...
2.4 Instrument Settings .............. .. ...............29...
2.5 Compensation Voltage (CV) Scans .............. ...............30....
2.6 Characterized Compounds ................. ...............3.. 1......... ...
2.6. 1 Nitroaromatics ................... ............... .................3 1..
2.6.1.1 (2,4,6 -) Trinitrotoluene (TNT)............... ...............31.
2.6.2 Nitramines ................ ........... ...............4
2.6.2.1 Cyclotrimethylene trinitramine (RDX) ............. ...............43.....
2.6.2.2 Cyclotetramethylene tetranitramine (HMX) .............. ....................4
2.6.3 Nitrate Esters ................... ...............50..
2.6.3.1 Nitroglycerin (NG) ....................... ...............50
2.6.3.2 Pentaerythritol tetranitrate (PETN) .............. ...............51....
2.6.3.3 Ammonium Nitrate Fuel Oil (ANFO)............... ...............53.

3 FAIMS/MS AND IMS/FAIMS/MS OF EXPLOSIVES .............. ...............57....


3.1 Instrum entation ............... .. ........_ .. ...............57...
3.1.1 Electrospray Ionization (ESI) Source ................. ...............57........... .












3.1.2 lon Mobility Spectrometer (IMS).. ......___ ......_._......_. ..............59
3.1.3 Orthogonal Dome FAIMS Cell .............. ...............59....
3.2 FAIM S Limitations............... ......... ...............6
3.3 Effects of Addition of Helium to the Carrier Gas. ................ .............. ........ .....63
3.4 High-resolution FAIMS (HRFAIMS) .............. ...............63....
3.4.1 Separation of Mixtures of Explosives ................. ...............64...............
3.4.2 Resolving Isomers .............. ...............68....
3.5 Dispersion Voltage ................. ... ...............69.
3.6 Effects of Adjustment of Plate Gap ............ ....._ ...............70
3.7 Relationship of DV to CV .............. ...............80....
3.8 Sensitivity ............... ...............8 1..
3.9 IMS/FAIMS of Explosives ............... ...............8 1...


4 CONCLUSIONS AND FUTURE WORK ....._ .....___ .........__ ...........8


4.1 Conclusions............... .. ................8
4.1.1 Characterization of Explosive Compounds ............ ..... ................84
4.1.2 FAIMS Characteristics .............. ...............85....
4. 1.3 IMS/FAIMS/M S ............... ...............85....
4. 1.4 High-resolution FAIMS...................... ..............8
4.1.5 Mixture Separations with FAIMS/MS .............. ...............86....
4.1.6 Disadvantages of FAIMS .............. ...............86....
4.2 Recommendations............... .. ...........8
4.2.1 Further HRFAIMS Improvements .............. ...............87....
4.2.2 FAIMS Carrier Gas Experiments .............. ...............87....
4.2.3 Practical Separations and Analysis............... ...............88
4.2.4 IM S Gating .............. ...............88....
4.2.5 CV Scanning............... ...............88
4.2.6 Instrument Maintenance .............. ...............89....
4.3 Future Work............... ...............89..
4.3.1 Ionization Sources .............. ...............89..
4.3.2 Development of Field Instruments ................ ...............91..............


REFERENCE S .............. ...............92....


BIOGRAPHICAL SKETCH .............. ...............96....











LIST OF TABLES

Table page

2-1 Explosives to be Characterized. .............. ...............27....

3-1 Relative lon Intensities. ............. ...............79.....

3-2 Relative CV Values.............. ..............79..











LIST OF FIGURES


Figure page

1-1 Finnigan LCQ APCI source ................. ...............15........... ...

1-2 FAIMS waveform and ion mobility behavior ................. ...............17........... ..

1-3 APCI source and FAIMS cell mounted on a Finnigan LCQ. ................... ...............1

1-4 Ionalytics line-of-sight FAIMS cell geometry. .............. ...............20....

1-5 LCQ quadrupole ion trap showing ion traj ectory. ................ ....._. ........._._....23

1-6 Example of oscillating ion motion according to the Mathieu equation (let a=1 and
q=1/5)................. ...............2

1-7 Mathieu stability diagram. ............. ...............25.....

2-1 Compensation voltage (CV) scanning. ............. ...............31.....

2-3 Negative ion APCI mass spectrum of TNT. ............. ...............33.....

2-4 TNT total ion count over four 2 min CV scans. ............. ...............34.....

2-5 FAIMS/APCI mass spectrum of TNT. ............. ...............35.....

2-6 FAIMS/APCI mass spectrum of TNT: measured for 2 min at CV = 6.6V ......................36

2-7 FAIMS/MS/MS of TNT m/z 197.1 peak at 35% collision energy ................. ...............37

2-8 CV spectrum of 2,4-DNT. ............. ...............38.....

2-9 FAIMS/MS mass spectrum of 2,4-DNT. .............. ...............39....

2-10 2,4-DNT: 2 min at CV = 11.5V. .............. ...............40....

2-11 CV spectrum of 2,6-DNT. ............. ...............41.....

2-12 FAIMS/MS mass spectrum of 2,6-DNT. .............. ...............42....

2-13 Mass spectra of 2,6-DNT: 2 min at CV = 1 1.6V and 15.3V. ...........__.. ........__........43

2-14 CV spectrum of RDX ................. ...............44........... ...

2-15 FAIMS/MS mass spectrum of RDX. ................ ...................... ..................45

2-16 RDX: 2 min at CV = 5.1V. ............. ...............46.....











2-17 CV spectrum of HMX ................. ...............47........... ...

2-18 FAIMS/MS mass spectrum of HMX. .............. ...............48....

2-19 HMX: 2 min at CV = 1.2V. ............. ...............49.....


2-20 NG (MW 227) ................. ...............50................

2-21 FAIMS/MS mass spectrum of NG. ........._._ ...... .__ ...............51.

2-22 PETN (MW 3 16)............... ...............52..

2-23 FAIMS/MS mass spectrum of PETN. ............. ...............52.....

2-24 FAIMS/MS mass spectrum of ANFO made with hexane. ............. .....................5

2-25 FAIMS/MS mass spectrum of ANFO made with nitromethane. .............. .............. .56

3-1 WSU electrospray ionization source ......................... ...............58.....

3-2 WSU electrospray ionization source photograph. ............. ...............58.....

3-3 Drawing of lonalytics orthogonal dome FAIMS cell. ......____ ........_ ................60

3-4 Photograph of lonalytics orthogonal dome FAIMS cell. ......____ ........_ ...............61

3-5 W SU apparatus. ............. ...............62.....

3-6 CV spectrum of HMX/RDX mixture ................. ...............65...............

3-7 Mass spectrum of HMX/RDX mixture ................. ...............66...............

3-8 CV spectrum of TNT/RDX mixture.. ............ ...............67.....

3-9 Mass spectrum of TNT/RDX mixture. ............. ...............68.....

3-10 CV spectrum of both isomers of DNT. ................ ....__ ...............69 ...

3-11 RDX CV spectrum at 0.5 mm............... ...............71...

3-12 RDX mass spectra at 0.5 mm............... ...............72...

3-13 RDX CV spectrum at 0.75 mm ................. ...............73........... ..

3-14 RDX mass spectra at 0.5 mm............... ...............74...

3-15 RDX CV spectrum at 1.0 mm ................. ...............75..............

3-16 RDX mass spectra at 1.0 mm............... ...............76...











3-17 RDX CV spectrum at 2.0 mm ................. ...............77........... ..

3-18 RDX mass spectra at 2.0 mm............... ...............78...

3-19 Relationship between CV and DV. .............. ...............80....

4-1 Distributed plasma ionization source ................. ...............90........... ...









LIST OF ABBREVIATIONS

Air Force Research Laboratory

automatic gain control

atmospheric pressure chemical ionization

Defense Advanced Research Products Research Agency

compensation voltage

direct current

dispersion voltage

high-field asymmetric-waveform ion mobility spectrometry

improvised explosive device

ion mobility spectrometry

limit of detection

mass spectrometry

quadrupole ion trap mass spectrometer

Research, Development, and Engineering Command

radio-frequency

Sandia National Laboratory

time-of-flight

unexploded ordnance

volatile organic compounds


AFRL

AGC

APCI

DARPA

CV

DC

DV

FAIMS

IED

IMS

LOD

MS

QITMS

RDECOM

RF

SNL

TOF

UXO

VOC









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

CHARACTERIZATION OF COMMONLY ENCOUNTERED EXPLOSIVES USING HIGH-
FIELD ASYMMETRIC WAVEFORM ION MOBILITY SPECTROMETRY COUPLED WITH
MASS SPECTROMETRY

By

Jared J. Boock

May 2007

Chair: Richard A. Yost
Major: Chemistry

The goal of this research was to characterize explosive compounds using high-Hield

asymmetric waveform ion mobility spectrometry (FAIMS) as an ion separation device interface

to a mass spectrometer (MS). FAIMS is a relatively recently developed technology that is

promising, adding a dimension of separation. This method was employed in such a way as to be

conducive to possible future development of field instruments (e.g., use of atmospheric pressure

chemical ionization (APCI)). In addition, several experiments were conducted with the

application of this method, mainly dealing with optimization of the instrument parameters.

These experiments led to the development of a new method: high-resolution FAIMS. An ion

mobility spectrometer was used as a desolvation and ion-focusing device to drastically improve

FAIMS resolution. This allowed for the separation of mixtures that otherwise could not have

been separated. The resolution was high enough that isomers of the same explosive were

successfully resolved. This can be beneficial to forensics studies.









CHAPTER 1
INTRODUCTION AND INSTRUMENTATION

1.1 Introduction

A maj or threat to the national security of the United States is the continued use of

explosives abroad and possible use by terrorists on the domestic front.' As a result, there is a

great need for improved separation and detection technologies for trace amounts of explosive

compounds. There are many possible applications for such technologies, such as prevention of

terrorist events, military weapons/cache detection, interception of explosives trafficking,

improvised explosive device (IED)/landmine detection and identification, post-event forensics

analysis, and unexploded ordnance (UXO) detection. A significant challenge to the detection of

minute amounts of explosives is that oftentimes these analytes will be overwhelmed by

complicated background matrices.2 Another is that benchtop-size instruments are inadequate to

provide timely analyses to meet the needs of the user. For instance, a military unit searching for

landmines or IEDs in a combat zone cannot easily deploy a large and cumbersome instrument.

To meet these needs, an instrument is required that can both provide separation of analyte from

atmospheric background containing large amounts of volatile organic compounds (VOCs). The

required limit of detection (LOD) for such an instrument has been determined by many field

trials conducted by government and commercial agencies (US Air Force Research Laboratory

(AFRL), US Army Research, Development, and Engineering Command (RDECOM) Sandia

National Laboratories (SNL), and Defense Advanced Research Products Agency (DARPA) are a

few of these) to be in the parts-per-billion (ppb) range.3

One of the primary technologies in use for explosives detection today (as of 2006) is the

ion mobility spectrometer (IMS).4 An IMS consists of an ionization source, an atmospheric

pressure drift tube, and a detector. A sample is ionized and the ions are introduced to the drift









tube and are passed through a series of ring electrodes held at increasing voltages of polarity

opposite to that of the ions, which creates a strong electric field. Separation is based on mobility

of each ion in the field, which depends on size, shape, and mass of the ion. The IMS has

limitations, however. First and foremost is that it is easily susceptible to contamination from

background matrix. The IMS as an explosives detector is primarily used in indoor facilities such

as airports or railroad stations, and samples are usually introduced as a swipe rather than

sampling vapor from air.'

Mass spectrometry could address many of the limitations of ion mobility. Research is

currently being conducted by many different organizations to design and construct a field-

portable time-of-flight (TOF) mass spectrometer (MS) to meet these needs.6,7 IOns are

accelerated through a vacuum drift tube by an electric field. The mass-to-charge (m/z) ratio is

determined by the amount of time each ion takes to pass through the drift tube. In most cases, a

TOF/MS is coupled with a gas chromatograph, which provides initial separation.8 Current

commercially available field-portable TOF/MS designs do not have a high enough resolving

power to meet the above needs, and they are not sensitive enough to detect levels of explosive

vapor that are practical to operational needs.9

A relatively recent (theorized in the late 1980s and developed in the 1990s and 2000s) and

promising technology is high-field asymmetric-waveform ion mobility spectrometry (FAIMS),1o

which will be discussed in detail. FAIMS is a novel, continuous chemical separation method

that has already proven successful in clinical, proteomics, and environmental analytical

applications.ll FAIMS does not have a high enough resolving power to function as a stand-alone

device;12 however, it is ideally suited to function as an ion separation device (in lieu of a

chromatography instrument) that can be coupled to a mass spectrometer.











1.2 Atmospheric Pressure Chemical Ionization

Atmospheric pressure chemical ionization (APCI) is a gas-phase ionization technique that


uses either a radioactive P-emitting source or a corona discharge at a pressure of I atm to


produce reagent ions and ultimately analyte ions. This method was chosen for this series of


experimentation because it most closely resembles an ionization source that is amenable to be


used for field instruments. Most other ionization techniques either require vacuum or are large,


complex systems containing lasers or high-voltage; these techniques are less suitable for a man-


portable field instrument. The APCI method is not identical to the ionization method to be


employed projected field instruments, as a difference arises from sample introduction. The


APCI ionization source used for this series of experiments (see figure 1-1) is designed for


AP Cl
FLANGE
VAPORIZER
POWER
CONNECTOR

AP Cl AP ORI ZE R
NOZZLE RHE R

VAPORIZER
AU XILIARY'
GASI
INLET FI VAPORIZER
SAMPLE ,,TB
IN LET I SAMPLE
TUBE
SHEATH ,
INLET CRN
APCI PROBE CORONADIHAE
D ISCH ARGE C ''NEEDLE
P OWE R
CONNECTOR o


APCI PROBE
RETAINER BOLT



Figure 1-1. Finnigan LCQ APCI source. [Reprinted from Thermo-Electron Corp. 2004.
Finnigan LCQ Series APPI APCI Combination Probe Operator 's Manual, Rev. A,
Thermo-Electron Corp.]









interfacing to liquid chromatography, and, therefore, introduces analyte samples as liquid. The

liquid sample is nebulized into droplets, which are passed pneumatically via nitrogen sheath gas

into a heated region where they are vaporized. The nitrogen and vaporized solvent then serve as

reagent gas as the vapor is passed over a corona discharge which produces reagent ions that

ionize the analyte via ion-molecular reactions.13 There are different mechanisms depending on

the type of analyte and whether or not ionization is positive or negative. Explosive compounds

tend to favor negative ionization (this behavior and ionization mechanisms will be discussed

further in a later section).

1.3 High-field Asymmetric-Waveform lon Mobility Spectrometry

As stated in the introduction, high-field asymmetric waveform ion mobility spectrometry

(FAIMS) is a relatively recently developed technology that in this series of experiments serves as

an ion separation device. FAIMS cells have many different configurations, though all basically

consist of two electrodes separated by a gap; the simplest configuration is two parallel metal

plates, which, consequently, was the first iteration of this design (Figure 1-2).15 An asymmetric

waveform is applied between these plates, which can either be the sum of two sine waves,12 Or a

square wave; in this case, a sum of sines was used with a high negative potential of -4000 V for a

short time, and a low positive potential for a slightly longer time. This peak voltage is known as

the dispersion voltage (DV), and creates a strong electric field. The rapid change in polarity of

the applied alternating potentials causes the ion to oscillate due to attraction/repulsion forces.

For this reason, a high negative potential is typically used for negative ions.

The basis for separation for a FAIMS device is a change in the mobility of an ion as this

field varies, unlike IMS which measures the mobility itself.16 IOns exhibit a variety of mobility

behaviors under high field, thus making FAIMS an ideal separation device. The changes in













lon Source k v Exit to Detector



Increasing,
Constant, or Asyrmetic Wavefrm;
decreei, : at ___---- High Field
High Electric Fields
Dispersion
Voltage (DV)

Time t Hil? Fela, t CTime at Low Flr-Lu idld



Figure 1-2. FAIMS waveform and ion mobility behavior. [Adapted from Guevremont, Roger.
2004. High-Field Asymmetric Waveform lon Mobility Spectrometry (FAIMS). (p.
2, fig. 1-1). Ionalytics Corp., Ottawa, Canada.]

mobility are classified into three different categories: "A-type" behavior, or exponential

increase in mobility proportional to change in field strength, "B-type" behavior, or exponential

increase in mobility followed by exponential decay as field strength increases, or "C-type,"

which is exponential decay in mobility as field strength increases. The change in mobility

affects the direction the ion travels toward or away from the plates (Figure 1-2).

If these conditions are maintained, then all ions will drift toward the plates according to

each separate mobility. However, the drift of a particular ion can be halted by the application of

a small DC voltage to one of the plates. This compensation voltage (CV) allows a single type of

ion to pass through the plates. As mobility changes are greater, an increased compensation

voltage is needed to balance the drift and allow transmission of the ion.

The mobility of an ion influenced by an electric field is described by

Kh(E) = K[1 + f(E)]

where f(E) is the dependence of ion mobility (Kh) as a function of electric field.12









Field strength is described by units known as Townsend (Td), for which 1 Td =

1017 x (E/N) where N is the number density of the carrier gas (expressed in atoms/m3) and E is

the field strength (in V/cm). Change in mobility is described by

Kh/K = 1 + a(E/N)2 + P(E/N)4

where Kh is the ion mobility at high field and K is the mobility at low field.

The two constants a and P are dependent on the dimensions of the FAIMS cell. The

estimated field strength for this series of experiments was calculated to be about 80 Td

(20,000 V/cm) at 1 atm.

To analyze a mixture of analytes, with differing CVs, it is necessary to scan CVs over a

range. Therefore, each analyte in the mixture is transmitted through the cell separately and a CV

spectrum consisting of ion signal vs. CV can be acquired.

A simple FAIMS cell geometry of two parallel plates is not necessarily ideal. It is

understood that when FAIMS is interfaced between an ionization source and mass spectrometer,

there will be a degree of ion loss, though it should be noted that there is no loss of sensitivity.

This may not make much of a difference; however, for even though signal is reduced, noise may

be reduced even further. It has been proven that different types of cells such as the cylindrical

cell used in this study have an additional ion focusing capability.12

An lonalytics (Thermo) Selectra beta II prototype was used for the FAIMS work described

in this thesis, though there were two different types of cell geometries used. The line-of-sight

FAIMS cell was used for the characterization, and it consists of two concentric cylinders that are

separated by a 1 mm gap. In order to assemble the cell, a solid stainless steel inner electrode fits

into a hollow stainless steel outer electrode, and a PEEK endcap maintains uniform distance

between these electrodes. The ends of the inner and outer cylinders are machined to have









hemispherical dimensions toward the exit aperture. Unlike a flat-plate design, the electric field

in a cylindrical cell is not uniform; this configuration, along with the hemispherical exit area,

allows for the focusing of ions toward the exit aperture. A PEEK sleeve prevents arcing and

other types of electrical interference between the FAIMS cell and the environment. Figure 1-3

shows a photograph of the APCI source and FAIMS cell.









Waveform lead


FAIMS Cell



--Curtain Plate


Curtain Plate lead




APCI Source



Figure 1-3. APCI source and FAIMS cell mounted on a Finnigan LCQ.



Prior to entering the FAIMS cell, ions pass through a hole in a curtain plate, which has a

potential of the same polarity as the ion mode of the source, in order to ensure that ions are










repelled into the FAIMS cell. For positive ion mode, the curtain plate voltage is kept constant at

+1000V, while negative ion mode curtain plate voltage is set to -1000V.




._00V1 To Waveform Generator













Heated capillary
to mass spectromneter






I Sour ce 1m




Figure 1-4. Ionalytics line-of-sight FAIMS cell geometry.



A clean, dry curtain gas is added to the FAIMS cell via a side port at a rate of 3 L/min.

This curtain gas serves two purposes. First, a portion exits through the orifice in the curtain

plate, which assists with desolvation of the ions and droplets from the ionization source and

minimizes the accumulation of solvent neutrals from the APCI source into the FAIMS cell. The

remaining gas propels the ions through the FAIMS cell and into the heated capillary of the mass

spectrometer. This gas can consist of a pure gas or a gas mixture, typically nitrogen, helium or

carbon dioxide; however, nitrogen was chosen for these experiments due to its predominance in









the atmosphere. Any moisture or trace hydrocarbons introduced to the carrier gas will affect the

performance of the FAIMS device by reacting with the ions, so charcoal and/or molecular sieves

are used, to purify the gas.

A waveform generator serves as the power source for the asymmetric waveform. A cable

connects the generator directly to the FAIMS cell. There is also a connector on the waveform

generator that allows the operator to observe the waveform on an oscilloscope and ensure that

the waveform DV and shape are constant. The Selectra prototype FAIMS unit is extremely

sensitive, and any slight interference with the waveform cable can affect the waveform.

There are four different FAIMS "modes" depending on the polarity of the CV and DV

polarity required to transmit a particular ion.12 The determination as to which mode an ion falls

under at this juncture is made experimentally, as a library has not yet been fully constructed,

which justifies a portion of this thesis. An ion in the "Nl" quadrant (including many explosives)

passes through the FAIMS cell with a negative DV and positive CV; an ion in "N2" requires

positive DV and positive CV; an ion in "Pl" requires positive DV and negative CV; and "P2"

ions are transmitted with negative DV and CV. This rule applies only for "A" and "C" type ions;

for "B" type ion mobility behavior it is not well defined. The mobility behavior of an ion is

affected by many factors, including: size, rigidity, and interaction with carrier gas (according to

Stokes and Einstein).l7

Although use of FAIMS as a separation device decreases total ion transmission, when the

CV is set for a particular ion of interest, the relative signal to noise ratio of that ion increases

greatly .

1.4 Ion Optics

The instrument used for this experimentation was a Finnigan (Thermo-Electron) LCQ,

which is a commercial, benchtop quadrupole ion trap mass spectrometer (QITMS). After the









analyte is vaporized, ionized by APCI, and separated via FAIMS, ions pass through the heated

capillary into a lower-pressure area, where the tube gate is located. The tube gate carries two

potentials of a polarity opposite to the analyte ions, one which focuses the ions into the skimmer,

which acts as a restriction between the higher pressure area (1 torr) of the tube gate and the lower

pressure (10-3 torr) of the ion optics. The second potential is used to deform ions and prevent

them from traveling further. The ion optics consist of a series of an octopole, lens, and a second

octopole. An octopole is a set of eight metal rods that carry an RF voltage (400 V) and DC

offset voltage (+10 V for negative ions) that induces an electric field that directs and focuses ions

into the mass analyzer. The lens between the octopoles serves both as a focusing medium and as

a pressure restriction, for the pressure once again decreases to 2x10-5 torr in the second octopole

and mass analyzer. As pressure is continually decreased, the mean free path (mfp) is increased,

and there are fewer neutral molecules and other ions present to interfere with the passage of the

focused analyte ion beam.18,19

1.5 Quadrupole lon Trap Mass Spectrometry

A quadrupole ion trap (Figure 1-5) consists of a hyperbolic ring electrode to which an RF

potential is applied, as well as two end caps with a DC potential.20 A DC offset voltage

(-10 V for negative ions) is applied to the end caps so ions will enter the trap. An RF waveform

of a particular drive frequency (02) is applied to the ring electrode which creates the quadrupolar

electric field, causing the ions to oscillate on a controlled trajectory. He gas at approximately

10-3 torr removes excess kinetic energy from the ions so that they can be trapped. The ion

traj ectory in the center of the trap is not to scale.
















EKIT LENB



TRAPPQED


IPN






SECOND EKIT
ocTAPOLELENS




SPACER NROD
BRINGS


ENrTRANCE ENlDCAP SPRING
ELECTRODE POST WSASHER

Figure 1-5. LCQ quadrupole ion trap showing ion trajectory. [Adapted from Thermo-Electron
Corp. 2003. Finnigan2 LCQ Series Hardw~are Manual, Rev. A.]

The oscillating ion motion is described by the Mathieu equation:22

-16eU 8eV
az = -2ar= q, = -2q =



The variables az and qz are functions of the dimensions of the trap and the potentials


applied, while z and r represent the axial and radial directions (between and perpendicular to the

endcaps, respectively); U is the DC amplitude applied to the ring electrode (if any), V is the RF


potential applied to the ring electrode, e is the charge on an ion, m is the mass of an ion, ro is the

inner radius of the ring electrode, zo is the axial distance from the center of the device to the


nearest point on one of the endcap electrodes, and 0Z is the angular drive frequency. The angular


drive frequency is equal to 2x1fRF where fRF is the frequency of the main RF voltage in Hertz.23 A


MOaU NT










graphical example of the motion in arbitrary directions is shown in figure 1-6. The actual motion

is similar to this example.




















-6 -4 -2 0 2 4 6


Figure 1-6. Example of oscillating ion motion according to the Mathieu equation (let a=1 and
q=1/5). [Adapted from McLachlan, N. W. 1947. Theory and Application of
Mathieu Functions, Dover].

For an LCQ, qz = 0. 1963 [RF frequency (number of charges)/ion mass].19 Solutions of

the Mathieu equation describe a stability diagram (figure 1-7) from which it can be determined

whether or not an ion will remain in the ion trap under a given set of conditions. If the qz of an

ion with known mass/charge falls within the boundaries of the stability diagram (is stable in both

axial and radial directions), then it can be trapped. If the qz falls outside the boundary, then the

ion will collide with the end cap electrodes and be lost. The factor P is based on the secular

frequency of the oscillation of an ion. When P equals one, the secular frequency equals half the

frequency of the RF field, and the magnitude of its oscillation increases such that the ion is lost.

Essentially, the P lines on the graph are voltage scans; in a quadrupole ion trap the amplitude of

the DC and RF voltages are ramped (while keeping a constant RF/DC ratio), to obtain the mass









spectrum over the required mass range. The sensitivity is a function of the scanned mass range,

san speed, and resolution.24





wS ~instability scan
a stabllty


Figure 1-7. Mathieu stability diagram. [Reprinted from McLachlan, N. W. 1947. Theory and
Application of Mathieu Functions, Dover].

The ion trap can only hold a certain number of ions at a time before repulsion forces (space

charge) cause ion displacement, resulting in a loss of resolving power and band broadening. To

counter this phenomenon, the LCQ employs a function known as automatic gain control (AGC)

where a prescan is conducted that monitors the generation of ions in the trap.25 The AGC scan

and the full MS scan form a microscan. In these experiments, three microscans were used for

each ion inj section period.









1.6 Multidimensional Mass Spectrometry

Another capability of the LCQ is multidimensional MS or MS/MS. This function can be

selected to determine the fragmentation patterns of an ion of a particular m/z in order to identify

the ion creating the peak or to examine the chemistry of that ion. A ion is selected, known as the

parent ion, and all other ions are ej ected from the trap. An excitation RF voltage is applied across

the end cap electrodes, causing the parent ion as kinetic energy to increase; collisions with the

He gas in the trap result in parent ion dissociation into fragment (daughter) ions. The

fragmentation pattern can be changed by altering the amplitude of the excitation RF voltage.19,26

1.7 Thesis Overview

This thesis is organized into four chapters. Chapter 1 is the introduction to the research

and an overview of the equipment used for the characterization portion. Chapter 2 presents data

relating to the characterization of explosives via FAIMS/MS. Chapter 3 introduces the second

instrument and compares it to the instrument used for characterization. There were several

experiments relating to FAIMS/MS and IMS/FAIMS/MS that were conducted throughout this

research; these are discussed in chapter 3. Chapter 4 discusses the conclusions of this research

and recommendations of other related efforts, as well as possible long-term future work.










CHAPTER 2
PROPERTIES AND CHARACTERIZATION OF EXPLOSIVES

2.1 Explosives that were Characterized

A list of explosives was selected for characterization by FAIMS/MS based on threat

assessments by various government agencies (including the Department of Homeland Security

(DHS) and Federal Bureau of Investigation (FBI)). This list is contained in the table 2. 1. Each

of these compounds will be discussed in great detail below.

Table 2-1. Explosives to be Characterized.
Compound Formula llolecular Weight

TNT (2,4,6-trinitrotoluene) C7HSN306 227
HNIX (cyclotetramethylene tetranitramine) C4HsNsOs 296
RDX (cyclotrimethylene trinitramine) C3H6N606 222
2,4-DNT (dinitrotoluene) C7H6NZO4 182
2,6-DNT (dinitrotoluene) C7H6NZO4 182
NG (nitroglycerin) C3HSN30, 227
PETN I.~ ... lll. 1... tetranitrate) CSH8N4012 316
ANFO (ammonium nitrate fuel oil) *

*dependent on type of fuel used

2.2 Classification of Explosives

An explosive material is chemically unstable or becomes unstable under certain conditions,

and as a result of rapid exothermic decomposition, can be destructive. There are many types of

explosives; however, government threat assessments1,27 have determined that only a certain

number of them are significant threats to life and property, primarily due to factors such as

supply and cost. As a result of this determination, the compounds of the highest threat levels

were examined closely in these experiments, as well as their precursors and degradation

products, if possible, as the presence of these can also be used to detect manufacture or storage

of the respective explosive materials.









There are two forms of explosive decomposition. The more powerful of these, detonation,

is characterized by an extremely fast supersonic shock wave released by the energy produced in

the favorable chemical reaction. The sudden high pressure of the shock wave is the major cause

of destruction resulting from a detonation. The other form of decomposition, deflagration, is the

subsonic portion of an explosion, usually propogated by thermal conductivity of the substance in

contact with the outer edge of the plume. Explosives are classified as either "low" or "high"

depending on whether or not they detonate.28 Low explosives include gunpowder and

pyrotechnics. High explosives are those used for military applications, mining, or demolitions.

All of the explosive compounds included in this study are considered high explosives.

There are also subcategories of high explosives: primary, secondary, or tertiary. Primary

high explosives are extremely sensitive to shock, friction, and heat, and are difficult to store.

Secondary high explosives are stable enough to be stored for a period of time, yet can be

detonated due to a sudden shock or heat. Most of the compounds in these experiments are

secondary high explosives. Tertiary explosives (blasting agents) will not easily detonate without

the aid of a primary or secondary "booster." This list includes ammonium nitrate fuel oil

(ANFO), which was a subject in this study.

2.3 Chemical Properties of Explosives

The maj ority of secondary high explosives characterized in this study are nitroaromatics,

nitramines, or nitrate esters. In the nitroaromatics and nitramines, one or more nitro (NO2)

groups are attached to an aromatic ring structure. Nitrate esters also contain nitro groups. These

nitro groups are bulky and electron-withdrawing, causing the nitrogen in the nitro groups to be

partially positive. Under APCI, primarily negative ions ([M]- or [M-H]-) are formed. Non-

halogen groups with atoms that are more electronegative than carbon draw substantial electron

density from the aromatic system. If enough internal energy is applied, one or more of the nitro









groups may detach from the ring or central structure, producing fragments. In either case, the

ions produced are negative, and negative mode must be used in order to separate and detect

explosives.

Nitroaromatics and nitramines are generally stable, but can be reactive; this property is

especially apparent in functional groups attached to the ring at a position meta to a nitro group.29

These compounds are thermally labile, so ionization techniques such as electron ionization (EI)

and chemical ionization (CI) will result in significant fragmentation of the molecule, which

makes detection difficult in a complicated background matrix. A "softer" ionization technique,

such as atmospheric pressure chemical ionization (APCI) or electrospray ionization (ESI) is

needed.13 Due to their low volatility under standard conditions, only a small amount of these

explosives will vaporize for detection in air, so a sensitive separation and detection techniques

are required. Because these compounds can condense on dust or other particles present in the

immediate environment, instruments designed to detect explosives in air should often sample for

explosives present on particles30,31 as well as the small amounts of vapor. Nitrate esters are

much less stable, and require lower instrument operating temperatures for analysis.

Explosive compounds are also highly toxic,28 creating an environmental hazard for areas

around disposal sites. Vapors can be introduced to the lungs, and liquids can be absorbed

through the skin, causing a variety symptoms depending on the relative concentration, even

leading to death.32

2.4 Instrument Settings

Both positive and negative ions can be formed in the APCI source, and the LCQ can

analyze ions of either polarity. As most of the explosives have large electronegative leaving

groups, they favor the formation of negative ions.33 It has also been shown that there is less










background noise in negative mode than positive mode. All of the mass spectra shown consist of

50 analytical scans, each consisting of three microscans.

The explosive solutions were diluted in a solvent containing 64.9% methanol and 35%

deionized water, as earlier studieS13 have yielded excellent results using this ratio. Solvents that

evaporate easily are best used for APCI, so that lower vaporizer temperatures can be used to

prevent thermal degradation of the analytes of interest. An extremely small amount,

approximately 0. 1%, of carbon tetrachloride was added to the solutions as some of the explosive

compounds favored the formation of a chloride adduct. This behavior will be discussed in turn.

All solutions were diluted to a concentration of about 10 ppm, as this is fairly realistic based on

calculated explosive detection applications.34

Solutions containing explosives were inj ected directly via the syringe pump of the LCQ

into the vaporizer at a flow rate of 20 CLL/min with a maximum ion inj section time of 50 ms for

AGC. The vaporizer was kept at a temperature of 3000C, while the heated capillary was set to

1500C. The LCQ software was used to tune the instrument as needed throughout the study in

order to maximize signal strength.

2.5 Compensation Voltage (CV) Scans

For all of the characterization in this section, CV was scanned from 0-20V four times over

2 min, for a total period of 8 min. The transition at the end of the scan from 20V back to OV is

extremely fast, covering a period of milliseconds. The ion of interest will only pass through the

FAIMS at its relative CV, therefore there will be a peak of ion flow at this voltage (ideally 100%

transmission); at other voltages the ion collides with the wall of the cell and is not transmitted.

(see Figure 2-1) This results in a peak at a particular time (and thus CV) in each of the four

scans.














CV CV CV CV

100% Transmission






0% Transmission
20V/ /









OV

2 min 4 min 6 min 8 min


CV CV CV CV


Figure 2-1. Compensation voltage (CV) scanning.

2.6 Characterized Compounds

2.6.1 Nitroaromatics

2.6.1.1 (2,4,6 -) Trinitrotoluene (TNT)

TNT is one of the most commonly encountered explosives due to its continued use for over

a century in both military and commercial applications. It is deemed a significant threat for use

by terrorists and other enemies of the United States. Like other nitroaromatics, it is stable until

ignition, other than slow decomposition over a period of years.28 Thus, there is the possibility of

low amounts of degradation products present in samples of TNT. Therefore, the most common









of these should therefore be characterized, as well as precursors used in the manufacture of TNT.

These compounds include the six dinitrotoluene isomers and trinitrobenzene (TNB). For the

structures of the nitroaromatics characterized in this study, refer to Figure 2-2. TNT has been

used on its own or as a component in mixtures, such as with ammonium nitrate (amatol),

aluminum powder, or other explosives.


O, N


NO,

2 ,4-DNT
1VW = 182


N0 O

TNT
VI~W = 227


ON


CH


O,N


N O,


2,6-DNT
VI~W = 1 82


TNE
MIW = 213


Figure 2-2. Nitroaromatics and molecular weights. [Adapted from Reich, Richard.
2001. (Figure 1-11). PhD dissertation. University of Florida.]


When TNT detonates it (ideally) decomposes according to the following equation:28

2C7H5N306 4 7CO + 7C + 5H20 + 3N2

although complete decomposition is only reached after some oxygen from surrounding air is

added as an ignition reactant.


C H3



~ NO2
NO2

3,4-DNT
MIW = 182
































II


979 113.212so 1~


TNT #1-105 RT:ODD-1 80 AV 105 NL: 1.34E5
T: p Full rns [50 DD-3 00 DO]i


60 80 100 120 140


160 180 200 220 240 260


280 300


Figure 2-3. Negative ion APCI mass spectrum of TNT. The first spectrum (Figure 2-3) shows
negative ion APCI of TNT without FAIMS separation. The [M]- ion at 227.0 and
[M-H]- ion at 226. 1 can be seen clearly, and a small amount of background noise
(ions that correspond to minor fragment ions of TNT or impurities) is visible.


[M-H]~












~~ [M]~


C H

ON NO,



N O,

TNT
MIW = 227


[M-NO] [M-


18817 OHs1











RT: 0.17-7.67 SM: 7B
6.66 NL
100. 7 B2E5
TIC MS
90-
0.64

80- 2.66
4.65
70-

(1.)








30-


20-


1 O-i 74

OBi 02 231 362 _450 1 89 65 9

1 2 3 4 5 6 7
Time (mln)




Figure 2-4. TNT total ion count over four 2 min CV scans. Figure 2-4 shows the CV spectrum
(total ion count) for TNT. As mentioned earlier, the ion of interest will only pass
through the FAIMS at its correct CV; therefore, there are ion peaks at these CV
values. There were four CV scans from 0-20 V, each with a duration of 2 min, for a
total of 8 min. It is from these spectra that the exact CV for each compound can be
calculated. From this spectrum, the CV can be calculated. For instance, the peak at
0.64 min: (0.64 min/2.00 min) x 20.0V = 6.4V.









TNT #34-660 RT: 0.38-7.58 AV: 627 SM: 7B NL: 2.81E3[M
T: p Full ms [ 50.0 0-40 0.0 0] M
226.8


CH,

ON iNO,



NO,

TNT
MIW= 227


Figure 2-5. FAIMS/APCI mass spectrum of TNT. Figure 2-5 shows the FAIMS/MS spectrum
for TNT. The background noise has been greatly reduced after the application of the
FAIMS. (see figure 2-2) The evident tradeoff for the improved signal/noise ratio is
the 50 x reduction in signal strength. In addition, two fragments of TNT also pass
through the FAIMS, as the fragment ion CVs are very close to the CV of the TNT
[M]- ion. The fragments show that TNT is a delicate ion that readily loses NO and
OH. The fragment peaks are in larger proportion to the molecular ion peak when
FAIMS is applied, most likely because the total ion count is lower, and there are few
to no impurities. Also, the [M-H]- no longer appears.











ThiT #1-173 RT: 0 00-1 98 AV: 173 NL' 1.71E4
T: p Full ms [50.00-400.00]
100-



85
80-
T5
70-
65.


~55~ [M-NO]



S40- [M-(2NO+CH3 1
35-[-N]
30-
25-
20-
15-
10

-63.3 78.3 151.9 166 8
100 150


226.8 [M]


ON


NO,


-- ]


228.9
200 250 300 350 400


Figure 2-6. FAIMS/APCI mass spectrum of TNT: measured for 2 min at CV = 6.6V Figure 2-6
shows a 2 min spectrum taken at a CV of 6.6V, which is when the most TNT [M]- ion
passes through the FAIMS. This spectrum is similar to that in the previous figure; the
only difference is that instead of averaging one of the CV peaks, data points were
taken for a stationary CV. As a result, this spectrum is improved further from that in
figure 2-4.








































IIIIIIIIII


TNT #816-344 RT: 0.23-4.90 AV. 329 NL. 2.55E2
T. p Full ms2 197.00@35.00 [50.00-400.00]

so- 9


85
8-
75







35


15-
1016


[M-NO]


ON


NO,


[M-NO]~




197.1


MW = 227


II I I.I


I i.


Figure 2-7. FAIMS/MS/MS of TNT m/z 197. 1 peak at 35% collision energy. Figure 2-7
demonstrates that FAIMS/MS/MS of even weak ions from explosives is feasible.
The FAIMS CV was set to 6.2V to allow the [M-NO]- fragment of TNT to pass
through the cell. This ion was collided with 35% dissociation energy, causing the
loss of a second nitro group. In other words, the predominant fragment ion peak
shown here is the same elemental composition as the [M]- ion for mononitrotoluene.


[M-(NO+OH)]~












RT: 0.00 -8.01 SM: 7B

100-
95 1.
90

85

80

75

65


S55



50

45
40-

30-0


20-0
1.00
-0.02
1 n


0 1


NL:
6.59E4
TIC MS
24DNT


13

1.16


1.21


3.17


3.11




3.22


7.18


7.28










7.35




7.53


5.32


4.97


7.00


2.96



2 3


4
Time (min)


Figure 2-8. CV spectrum of 2,4-DNT. Figure 2-8 shows the CV spectrum for 2,4-dinitrotoluene
(DNT), which is a precursor of TNT, and is often an impurity in explosives
containing TNT. This CV spectrum is different from that of TNT (figure 2-4) in that
the CV is 11.5V, compared to 6.6V. In addition, there is a small amount of solvent
ions that transmitted at slightly lower CV than the maj or DNT ion through the
FAIMS cell.












24DNT #399-419 RT: 500-5 25 AV 21 NL: 553E3 [M-H]~
T: pms[50 00-400 00]
181 1
100-
96-

90- CH,



TNO





Ss 2,4-D3NT
5 [M-(NO+CH3)I
2 so- 1VW = 182
46
[M-(NO+2H20)] [M-NO]


35 1[M-2NO]- [M]~
30-
26-
20-
136 1
15-
10-
11B-6 1519
5~ 5.6 877

100 150 200 250 300 350 400
m/z


Figure 2-9. FAIMS/MS mass spectrum of 2,4-DNT. Figure 2-9 shows the mass spectrum of 2,4-
DNT after FAIMS separation (averaged over the peak at 6.6 min in the CV spectrum
in figure 2-8). As with TNT, this ion fragments readily, and experiments have shown
that many of these fragments have CV' s that are close to the primary ion.











24DNT3 #2-159 RT 0 02-1 99 AV 158 NL 5 35E3
T: p ms [50.00-400.00]
181
100,


85-1 H,







60

as~ 2,4-DNT
so M0 IW = 182
S45-
m ~[M-(NO+CH3)I

35-
30 [M2O~\ [M-NO]~
25 [M2O
20- 135.1
15-
10
5-120.9
152.1 _1183.4

100 150 200 250 300 350 400
m/z



Figure 2-10. 2,4-DNT: 2 min at CV = 11.5V. A mass spectrum of 2,4-DNT at a fixed CV of
11.5V is displayed in figure 2-10. There is little to no noise.













RT 0 00 801 SM 7B



80 TIC154
1 59




05 071513141 9 1 6513


52


m/z 181 m/z 1



3 50 3








7




343



3 36 35

3 3 3 76 6


25 30 3 0


Tim ( in


549

554

i56


348E4
TIC MS
26DNT


7 44 7 4 4




n6 59 65

7 57


485 5 15
571587




556




5 49

546 54
562





5 39

5 33 1 5 71
514 57

45 50 55 60


NL
4 89E3
miz=
M8S 4-182 4


m/z 181

[M-H] 14


7 52


7407


70


NL
2 33E4
m/z=
15S 5152 5


80_ m/z 152 1419
1 49
60- [M-NO]

40 1 41 1 6 179
20~1 35
C 11 .1 1.1 102191
00 05 10 15 20


704

65 70 75 80


Figure 2-11i. CV spectrum of 2,6-DNT. 2,6-dinitrotoluene is an isomer of 2,4-DNT and another
precursor of TNT. Both nitro groups are in ortho positions to the toluene methyl

group. 2,6-DNT is more likely than 2,4-DNT to lose an NO group, and as such, there
are two predominant ions formed in APCI, causing the CV spectrum to display

different behavior than 2,4-DNT. The two ion peaks have close CVs and are not

completely resolved from each other. These can be seen in figure 2-11.

Experimentation in improvement of FAIMS resolving power was conducted and will
be discussed in chapter 3.









26DNT#B12-636 RT 0 14-7 98 AV 625 NL 3 62E2
T p ms [50 00-400 00]


O,N 0

2,6-DNT
MTW= 182


N O,


[M-H]



[M]~


Figure 2-12. FAIMS/MS mass spectrum of 2,6-DNT. Figure 2-12 is a mass spectrum averaged
over the entire 8 min run. The two predominant ions are visible at m/z 181 ([M-H]~)
and m/z 152 ([M-NO]-), as described at figure 2-11.











[M-H]~
181 1 26DNT#234-262
NL 4 43E2
[M]~
80-,


[M-NO] C 16 2 O

I CV =11.6V2,6-DNT
30 1 1 0
20)c~ M= 182


92 152 O 26DNT#12-636
625 NL: 3 62E3 T
0-[M-N O]- -pms
50 00-400 00]


so- CV = 15.3V
40-- ~[M-H]- M



53 8 eg 01 1 1 8 1

100 150 200 250 300 350 400
mtz


Figure 2-13. Mass spectra of 2,6-DNT: 2 min at CV = 1 1.6V and 15.3V. The fact that 2,6-DNT
more readily fragments under APCI than its isomer does not affect the separation
when the CV is fixed, as each ion is transmitted through the FAIMS cell at a separate
CV. This behavior can be seen in figure 2-13. Note that the optimal CV values for
the [M-H]- ions of the two isomers are too close (1 1.6V vs. 1 1.7V) for them to be
separated. However, only 2,6-DNT produces a second FAIMS peak (at CV = 15.3V)
for the [M-NO]- fragment ion.

2.6.2 Nitramines

2.6.2.1 Cyclotrimethylene trinitramine (RDX)

The nitroamine RDX (research department composition X) was developed as an explosive

during the 1930s and was used widely during World War II. This explosive is found in many

mixtures (such as Torpex, RDX mixed with TNT and aluminum powder), though it is usually

encountered as the base for several types of plastic explosives, the most common of which is


Composition C-4 (RDX with polyisobutylene and di(2-ethylhexyl)sebacate as the binder and

plasticizer). RDX is stable at room temperature will not detonate without a detonator.35 Another











type of plastique that contains RDX is Semtex, which is a commercial explosive that has been

used by terrorists many times in the past (such as Pan Am Flight 103 in 1988). The other


component in the Semtex mixture is PETN (pentaerythritol tetranitrate), which will be discussed

in its own right.


RDX decomposes completely according to:


C3H6N606 4 3CO + 3H20 + 3N2

and also reacts with oxygen in the air to complete the reaction.

RT: 0.00 -15.99 SM: 7B
5.50 9.52 NL:
100 9.5E


8 60
S55
50
S45
a 40


1.56


13.16
12.86 1
,,,, -~
11.60 ~i~
12


6 8
Time (min)


2 4


0.17


14


Figure 2-14. CV spectrum of RDX. Figure 2-14 shows the CV spectrum for RDX. The peaks
are fairly well reproducible. There are two small peaks before the ion of interest in
all four scans, which are identified in figure 2-15.










RDX #193-205 RT:012-15.97AV 13 SM 7~B NL 4.20E3
T~ p ms [ 50 00-400 00]


2568


150


Figure 2-15. FAIMS/MS mass spectrum of RDX. The [M+Cl]- adduct is formed with the small
amount of carbon tetrachloride in the solvent. (see Figure 2-15) One property of both
RDX and HMX is the formation of little to no molecular ion. Interestingly, the
adduct formed of RDX with the 37Cl isotope of chloride can also be seen in the mass
spectrum. The first additional peak shown in the CV spectrum (figure 2-14) is m/z 62
which is an [NO3]- fragment peak; it is seen in many mass spectra of explosives,
especially at higher temperatures. The second is a small amount of m/z 286, which is
another adduct formed with RDX: [M+NO3 -.


O Cr

RDX (MIW 222)




[M+37 -]


~[M+NO31











O~ O










O Cr

RDX (MIW 222)





[M+37 I-


RDX20 #~2-167~ RT 0.01-1.99 AV: 166 NL 2 26E4
T: p Full ms [50.00-400.00]
10 -
[M+35 I-






70-

65
Sc --

45
40-
35








1-


Figure 2-16. RDX: 2 min at CV = 5.1V. Figure 2-16 demonstrates that FAIMS/MS is an
effective method of separating and detecting chloride adducts of RDX, as there are no
interference.


2.6.2.2 Cyclotetramethylene tetranitramine (HMX)


HMX, derived from High Molecular Weight RDX (sometimes known as High Melting


Point Explosive), is a nitroamine discovered as a byproduct of RDX that consists of an eight-


membered ring structure rather than a six-membered ring. It is slightly more powerful than RDX


because molecules of greater mass generate greater energy when they decompose. Another


difference between HMX and RDX is that HMX has a greater temperature at which the molecule


begins to fragment (hence the name).35 HMX is used exclusively for military applications, and is













most often included in a mixture, such as in some types of plastic bonded explosives, or octol,


formed with TNT.


RT: 0.00 16.OO SM: 7B

100- .6





75-

70-




850


40-





25-

2 0-

15-

1 0-



35-
30-l


NL:
9 41ES
TIC MS
HMX


4 27


12.27






























12.29 j13 54 1431


4 2 _


2_81


)_310.73


II 2 4


iIl


5.43 6 20 7.95 8.28
l I l ll
6 8
Time (rr~


l ilIl
10 12


14 16


Figure 2-17. CV spectrum of HMX. HMX behaves similarly to its nitroamine counterpart RDX.
The FAIMS CV spectrum shown in figure 2-17 reflects this. There are some

differences between the two, though. HMX has a different CV (thus it can be

resolved, as will be discussed in chapter 3), and there is no evidence of the [NO3]

peak or formation of the [M+NO3]- adduct.


































[M]96


HMX #r22-796 RT. 0.43-16.00 A 775 NL. 5.05E3
T: p Full ms [50.00-400.00]
100-




s, O

70-
65- -g I\
TO N



see ON .O
S45-/
S40. -0
35~ HMX (MIW 296)


25-

15-
10-


[M +35C]


O


[M +37 I-


Figure 2-18. FAIMS/MS mass spectrum of HMX. Similar to RDX, the dominant ion for HMX
was [M+Cl]-. This can be seen in figure 2-18. Once again, adducts of HMX were
formed with both dominant isotopes of chloride. There was also evidence of a small
amount of [M] .











HMX #1-109 RT O 01-1 98 W*/ 109 NL 7 73E4
T p Full ms [50 00-400.00]







70-

80



754-


~~-0


a: 40-
35] HMX (MIW 296)

25-
20-


15-
10-


[M+35 ~


331 1















[M+37 I-


296.2


Figure 2-19. HMX: 2 min at CV = 1.2V. Figure 2-19 shows the HMX [M+Cl]- peak as the CV
is fixed at 1.2V for a period of 2 minutes. The mass spectrum contains no
interference, and there is a small amount of the molecular ion peak, as there was
during the CV scans (figure 2-18).




























49









2.6.3 Nitrate Esters

2.6.3.1 Nitroglycerin (NG)

There were two nitrate esters that underwent characterization via FAIMS/MS. The first of

these is nitroglycerin. Nitrate esters contain nitro groups like the previously mentioned

explosives, though they are bonded to oxygen atoms, rather than nitrogen or carbon. There is

also no ring structure, unlike the nitroaromatics and nitramines (Figure 2-20). Because of this

structure, nitrate esters are extremely unstable, and fragment easily, requiring only a light shock

to explode. Nitroglycerin is one of the oldest known explosives, discovered in the mid-

nineteenth century.28 It is mainly used in the manufacture of dynamite, gunpowder, and rocket

propellants.




O_ O"N'.O-

O 'O




N'


Figure 2-20. NG (MW 227) [Adapted from Paul W. 1996. Explosives Engineering, (Figure 3-3
p. 76), Wiley].










NG #1-538 RT 0.01-6.75S AV 538 NL 9.19E2
T -pms[50 00-40000]
10- 1.9[N 3O_ O N O


75-3 O-4,0




eJ s o* '"o-

ss-~ NG (MW 227)






500 .5 0 5 0 5 0


45ur -2 FISM a setu fG iial t ntomns hr o
40-ua o o G I loddntfr adcs vntog tehae ailr
35-eAC orewa erae ote iiu f10C alo h irt
30-hsoN eefamntd hs irt os(mz6)wr eaae n
25-tdwt AM/Ma hw nFgre22.Bcuete asrnewl o
20-dblw5 m o hs ntuet ti o nwnhwtecro tutr











Figerure 2-21. FAIMoS/MeS. maseeFgr -2 Es spcrmofN.Smlalote i aniitramnes thereis drno

WWIIandof thea APIsource wiiasy decresedton thres miniu deoaof 10C alle fof thenditraes

artiler branchn oe yes of NGl weefagmbentd Thesenitra. te ios (ee m/z 62) were separated an

relatve detecilted with FlAIMS/MSd as hw m intrt Figure -21. Because mpasst rangewl o


effciexplosivs



























Figure 2-22. PETN (MW 316). [Adapted from Paul W. 1996. Explosives Engineering, (Figure
3-3 p. 76), Wiley].


PETN #81-158 RT 0 00-1 98 AV 158 NL 7 93E2
T p ms [50 00-400 00]
61 9
100

[NO3]










45







15


o o


PETN (MW 316)


100 150 200 250 300 350 400


Figure 2-23. FAIMS/MS mass spectrum of PETN. Like the nitroamines and NG, PETN does not
yield a detectable molecular ion. It did form both chloride and nitrate adducts;
however, these were in extremely low intensity. As a result, PETN did not form any
discernable peaks of high intensity during this experimentation. However, due to its
extreme instability, (also like NG) PETN did fragment into nitrate ions which were
separated and detected by FAIMS/MS (Figure 2-23). As this figure shows, the total
ion count is relatively low.


[M-NO2]- [M+Cl]- +NO3]





261 3498A,57 1 r377 ,









2.6.3.3 Ammonium Nitrate Fuel Oil (ANFO)

ANFO is the most commonly encountered explosive today, used mostly for mining

operations. It is simple and inexpensive to produce, and the components are not difficult to

acquire. It is one of the most used explosives by terrorists, having been used by organizations

such as: Fuerzas Armadas Revolucionarias de Colombia (FARC), Provisional Irish Republican

Army (IRA), Palestinian extremists, and some American extremists. Ammonium nitrate (used in

fertilizers) is an oxidizer and can explode inefficiently on its own if shocked. A hydrocarbon

mixture provides fuel for the reaction. Usually kerosene or diesel fuels serve this purpose

industrially; although, terrorists have used other fuels (such as nitromethane to make the

"kinepak" mixture used for the Oklahoma City bombing). Since there are so many different

production methods and types of fuels used in different proportions, no two batches are identical.

This, coupled with the fact that ammonium nitrate and hydrocarbon fuels are present in

background air, suggest that ANFO will be extremely difficult to detect. On the other hand, a

destructive bombing requires large amounts of it (eg., over 5,000 pounds at Oklahoma City)

Ammonium nitrate decomposes according to: 2NH4NO3 & 4H20 + 2N2 + 02

the most important product being oxygen. The hydrocarbon fuel added burns rapidly to include

CO2 aS a product.28

ANFO is extremely stable, but is hygroscopic, which hinders long-term storage. ANFO

will not explode without a secondary booster (blasting cap); the power of the explosion can be

controlled by altering the mixture.

Five different types of ANFO were examined in this research. The hydrocarbon fuels used

were: n-pentane, n-hexane, toluene, nitromethane, and diesel fuel. All were made using about

94% (by mass) ammonium nitrate and the remainder of fuel. The pentane and hexane were used

to determine the effects of using a simple straight-chain hydrocarbon fuel to simplify (and thus









identify peaks in) the mass spectra. Nitromethane was used to produce the "kinepak" mixture

that was discussed above; toluene was used to provide a large structure with a methyl group to

compare and contrast results to kinepak ANFO. Characterization was done on the FAIMS/MS in

both positive mode and negative mode. Dilute solutions (about 10-15 ppm) were heated slightly

to instigate the oxidation reaction without causing detonation.

Characterization of ANFO using this method is difficult. There are no other mass spectra

with which to compare the results of this effort. This is likely due to the fact that no two batches

of ANFO are alike, and the mass spectra are difficult to reproduce. A limitation of the LCQ is

that the mass range cannot be set below m/z 50; therefore, fragment ions from ammonium nitrate

and the fuels cannot be seen.

Nearly all of the negative ion APCI data sets collected show an ion base peak in common:

[NO3]- (m/z 62). This may result from ammonium nitrate that has not yet reacted. For those

fuels of m/z greater than 50; though, the fuel itself was not seen neither in positive mode MS, nor

in negative mode MS (though straight chain hydrocarbons are almost never seen in negative

mode MS), the exception to this being ANFO made with diesel. In every case for ANFO, a

single 16 min scan was conducted comprising an area from CV -40V-40V. In each of the

following figures (2-24 and 2-25), a single peak was detected.











ANFOHex#82-639 RT~ 0 02-8 02 AV~ 638 NL~ 2 04E5
T p Full ms [5O 00-400 00]
61.9


[NO31


Figure 2-24. FAIMS/MS mass spectrum of ANFO made with hexane. Figure 2-24 shows a
FAIMS/MS mass spectrum of ANFO made with hexane. The base peak is [NO3 ~,
which passed through the FAIMS cell at a CV of 9.7V. There is also a small peak at
m/z 125.










































I _I I In II I


ANFOk~inepakt_0701 04150533#~81-1294 RT~ 0 01-15 55 AV~ 1294 NL~ 9 48E2
T p Full ms [50 00-400 00]
61 7

as-[NO31







9r 0- 10






70;6 124.7
65-


283 8


3 2


Figure 2-25. FAIMS/MS mass spectrum of ANFO made with nitromethane. ANFO made with
nitromethane (figure 2-25) as the fuel has a much lower ion signal than that made
with hexane; this behavior was also seen with ANFO made with toluene. It also has a
[NO3]- base peak, though there are some other ions as well. These ions were in
common with some of the other runs as well: m/z 125, m/z 141, and m/z 284.





Detection of ANFO with a field instrument would prove to be difficult, as ammonium


nitrate exists in the background (mostly in fertilizers) and the most common fuel, diesel, is used


to power engines everywhere. The most likely type of ANFO to be detected would be


"kinepak," for nitromethane is used for little other than as a fuel additive for race cars.


331 9
304 2350.7 38O392 7









CHAPTER 3
FAIMS/MS AND IMS/FAIMS/MS OF EXPLOSIVES

3.1 Instrumentation

The experiments described in this chapter were conducted using a different instrument

configuration than that in chapter 2. As in the other experiments, the mass spectrometer was a

Finnigan LCQ.

3.1.1 Electrospray Ionization (ESI) Source

Electrospray ionization (ESI), employed in this study, is similar to APCI. Instead of

relying on a corona discharge to ionize molecules in the gas phase, ESI uses electrical charge to

ionize molecules in solution. A liquid solution containing the analyte of interest in solvent is

passed through a capillary held at high potential. Droplets leaving the capillary become highly

charged. As the solvent in the droplets evaporates, ionized molecules are ejected into the gas

p ase. 36

The unique non-commercial ionization source used in the studies in this chapter was

developed by the Herbert H. Hill laboratory at Washington State University (WSU). It is a fairly

simple apparatus, consisting of a fused silica capillary held in place by a ferrule, and mounted in

a nonconductive teflon casing. Sample is introduced via syringe pump into this capillary, to

which approximately -13.5 kV of ionization potential is applied to a stainless steel ferrule

through which the capillary is threaded. Due to the relatively small (about 0.5 mm) internal

diameter of the capillary, the flow rate of the syringe pump can be low (as low as 1 mL/min).

































Figure 3-1. WSU electrospray ionization source. Figure 3-1 shows the WSU ESI source. The
darkly shaded areas are constructed of Teflon, and the capillary is fused silica. The
ferrule used is conductive stainless steel.


Figure 3-2. WSU electrospray ionization source photograph. Figure 3-2 is a photograph of the
WSU ESI source. Note that the tip of the capillary extends through the screen of the
IMS. This was done so that the ions would not be deflected backwards by the electric
field.









3.1.2 lon Mobility Spectrometer (IMS)

An ion mobility spectrometer (IMS), as briefly discussed in Chapter 1, is similar to

FAIMS, in that it employs an electric Hield to separate ions based on mobility. An IMS drift tube

was designed and patented37 by the Hill laboratory at WSU. This tube has a total length of 34

cm, containing a series of flat ring-shaped electrodes composed of conductive stainless steel with

dimensions of 50 mm outer diameter x 48 mm inner diameter x 3 mm, separated by ceramic

rings. A voltage is applied to each of these ring electrodes that generates the electric field that

propels ions through the drift tube. For negative ions, the first electrode carries a potential of

about -10.5 kV, and this is incrementally decreased on successive electrodes such that the last

electrode carries about -2 kV. The curtain plate of the FAIMS (set to -1 kV) acts as the Einal

electrode of the IMS to attract negative ions into the FAIMS cell. This IMS operates at

atmospheric pressure and employs heating to help desolvate the sample. The heater was adjusted

to be as high as possible without fragmenting the sample ions (temperature ranging from about

1500C to 2500C). Countercurrent gas flow also aids in desolvation.

The IMS contains a Bradbury-Nielsen gate ring electrode38 which divides the tube into a

desolvation region and IMS drift region. A reference potential is used to "open" the gate, thus

allowing ion transmission from the desolvation region into the drift region, or a different voltage

can be applied to "close" the gate by creating an orthogonal electric Hield that prevents ion

transmission. In these experiments, the IMS was first operated in total ion transmission mode,

and then it was set to gate ions for separation.

3.1.3 Orthogonal Dome FAIMS Cell

This instrument configuration also utilized an lonalytics Selectra beta prototype; however,

the cell geometry was not the same as that used in the first series of experiments (chapter 2). The

cell used in the characterization experiments consisted of a line-of-sight cylindrical geometry; in










contrast, the cell used in this chapter (figure 3.3) employs an orthogonal dome-shaped geometry.

The waveform generator, application, and waveform shape are identical. Two major differences

between the two configurations are that the dome-shaped cell permits the distance between the

electrodes to be adjusted, and provides slightly higher resolution.39


STo Waveform Generator


la Dispersion voltage


Outer Electrode


Heated capillary
to mass spectrometer








ADJUSTABLE


Inner Electrode


Curtai Plt


NOT TO SCALE


Figure 3-3. Drawing of lonalytics orthogonal dome FAIMS cell. Figure 3-3 contains a drawing
of the orthogonal dome FAIMS cell. The red and black dots describe a simple
scenario in which the black dots represent the analyte of interest and the red dots are
noise (interferent ions). Both types of ion enter the FAIMS cell through the curtain
plate and are separated as the red ions collide with the walls of both electrodes when
the proper CV is set. A percentage of the analyte ion is also lost, but this is the only
ion type that enters the mass spectrometer.

































Figure 3-4. Photograph of lonalytics orthogonal dome FAIMS cell. Figure 3-4 is a photograph
of the disassembled orthogonal dome cell used for these experiments. The plate gap
adjuster rotates to change the distance between the inner and outer electrodes.





































Figure 3-5. WSU apparatus. A photograph of the entire WSU apparatus including ESI source,
IMS drift tube, and FAIMS cell can be seen in Figure 3-5. The waveform generator
is not connected to the FAIMS cell so that it is easily visible.

3.2 FAIMS Limitations

As discussed briefly in Chapter 1, one of the primary limitations of FAIMS is fairly low

resolution compared to chromatographic separation methods, hence its use here in conjunction

with mass spectrometry as a separation device rather than a stand-alone instrument. There are

several factors that can affect resolving power of a FAIMS cell.40 The factors that have been

addressed in the following experiments include dispersion voltage, distance between the inner

and outer electrodes in the domed FAIMS cell, carrier gas consistency, and addition of a

desolvation region prior to the FAIMS cell. In addition, another limitation is band broadening

over time as buildup of solvent interferes with ion separation within the FAIMS cell.









3.3 Effects of Addition of Helium to the Carrier Gas

For the entire first set of experiments (chapter 2), the carrier gas in the FAIMS cell was

100% pure nitrogen. However, experimentation has shown that the addition of a percentage of

helium can increase transmission and sensitivity. This effect slightly narrows CV peaks and

therefore increases resolving power. The reason for this behavior is not clear; although, it is

suspected that because helium increases mean free path within the cell, ion dispersion and

desolvation are also increased. It is also possible that the high-Hield is effectively strengthened.

It is not desirable to use too high a percentage of helium carrier gas into the cell, as the

high voltage can induce discharge. It should be noted that because the amount of helium in the

atmosphere is trace, Hield instruments developed with FAIMS technology would most likely not

employ helium in the carrier gas.

3.4 High-resolution FAIMS (HRFAIMS)

High-resolution FAIMS (HRFAIMS) experiments were conducted on the WSU apparatus

described in this chapter. Three maj or differences between this instrument and that used for

characterization in chapter 2 in terms of FAIMS resolution are: FAIMS cell geometry, addition

of an IMS (set to total ion transmission mode) in front of the FAIMS cell, and use of a

percentage of helium (25%) in the nitrogen carrier gas.

As discussed above, the FAIMS cell used for this set of experiments was an orthogonal

dome cell, rather than a line-of-sight cylindrical cell. The dome cell can provide higher

resolution, due in part to the longer path length through the cell, and due in part to the use of two

Hield areas in series; there is a cylindrical area and a spherical area.

The IMS drift cell was set in full ion transmission mode (ie., there was no gating);

although, the heater and electric Hield were still engaged. There is also gas flowing

countercurrent through the IMS, which can dramatically reduce the amount of solvent reaching









the FAIMS cell. The temperature of the heater was set to be as high as possible without

fragmenting the explosive molecules. The IMS had the most significant effect on overall

resolution compared to the other factors. Transmission was increased, and CV peaks were

narrower, demonstrating increased resolution. This is probably due to the fact that the heater

desolvates the sample, and the electric field focuses ions into a tight beam for more efficient

introduction into the FAIMS cell.

3.4.1 Separation of Mixtures of Explosives

Although use of FAIMS as a separation device may decrease total ion transmission, when

the CV is set for a particular ion of interest, the signal-to-noise ratio of that ion can increase

greatly .

The first mixture that was examined consisted of about 20 ppm of HMX and about 15 ppm

of RDX. The DV was set to -2500V, which was the optimal setting when the IMS was installed.

The carrier gas was made up of 75% nitrogen and 25% helium to improve ion count and

resolution. Figure 3-6 is the CV spectrum for the HMX/RDX mixture. CV was scanned from 0-

20V once over a period of 8 minutes. Each peak is distinct and fully resolved. The RDX peak is

significantly smaller because there was not only a slightly smaller concentration in the sample,

but also the boiling point is much less than HMX.












-800 SM 7B
O 96

090_



























O 31


RTOOO0



100-


95 -

as 6



7 0-


65-


50-


2 so


30
25
20



0[


NL
1 38E4
TIC MS
RDXH~M(M
iurelE


H MX
CV 2.4V


RDX
CV 7.2V



200 2 12127,


0 05


10 15 20 25 30 35 40 45
Time(min)


50 55 BO BS 70 75


Figure 3-6. CV spectrum of HMX/RDX mixture.












































65


Solvent
CV 5.0V














































SI I I 'li l l i I l i l l i l l'I '
100 1 50 200 250 300 350 400
m /z



Figure 3-7. Mass spectrum of HMX/RDX mixture. Figure 3-7 contains the mass spectra of the
RDX/HMX mixture, showing the [M+Cl]~ adducts of both RDX (m/z 257) and HMX

(m/z 331) taken from the relative CV peaks (which are seen in Figure 3-6). The
HMX peak at m/z 341 is an [M+NO3]- adduct


256.9


RD~XH MXMidurel 6#
140-143 RT
283-289 AV4NL
231E2T -pms[
50a0-400 00]











R DXH MXMMIurel 6#
40-54 RT' 0 80-1 08
AV 15N L1 18E 3
T -pms[
50a0-400 00]


RDX
CV 7.2V


330 9


1746 2027 i


HMX
CV 2.4V


340 9












RT 0a00-1589 SM 7B
1B5 NL:
100- 8 49E3
95: TIC MS
TNT RDXTNTMI
90- Aurel2
CV 2.1V
1 57



75-




55- 81

so- 1 RDX
451" I Solvent CV 7.2V
40- ~CV 5.0
35-
1 91
30 ~5.76 ,

570


15~ 11.20 5 47

10- 0.67 9 54 6 9 8 42 9 62 10 96 119


23467810 1112 1314 15
Time (mmn)


Figure 3-8. CV spectrum of TNT/RDX mixture. A second mixture examined was torpex, which
consists of TNT and RDX in approximately equal portions. The only difference
between the mixture that was examined here and the military explosive is that in this
case there is no aluminum powder. In this case, one CV scan from 0-20V was done
over 16 minutes. Once again, both components (see Figure 3-8) are fully resolved
from each other and the solvent background in the CV spectrum.











225 9 RDXTNTMlxturel 2#Q8
100 -4-92 RT 1 69-1 85
90- AV 9 NL 5 20E2 T -
TNT p ms i
80- 50.00-400.00]
CV 2.1V






20-2

68- 256 8 RDXTNTMlxturel 2#2
81-285 RT
5.70-5.78 AV. 5 NL.
60-- 3 55E2 T p ms [
50RDX 50 00-400 00]
CV 7.2V





10-8


100 150 200 250 300 350 400
m/z


Figure 3-9. Mass spectrum of TNT/RDX mixture. Figure 3-9 shows the mass spectra of the
respective CV peaks of TNT and RDX in the mixture (Figure 3-8). Both components
(the TNT [M]- ion at m/z 226 and the RDX [M+Cl]- ion at m/z 257) are shown. It
should be noted that because the same DV was used for both of these mixtures
(-2500V), the CV values of RDX and the solvent is identical in the CV spectra of
both mixtures. (Compare figs. 3-6 and 3-8)

3.4.2 Resolving Isomers

The next step in separation is the resolving of two isomers of the same explosive. The


explosive chosen was dinitrotoluene (DNT), because it has two isomers that are precursors to


TNT: 2,4-DNT and 2,6-DNT. The analysis of these isomers can potentially be important to


explosives forensics. Few, if any, technologies deployed today can detect and resolve these


isomers from each other. Because they are isomers, they have identical molecular weight. Both


have similar mobility behavior, but due to the fact that the shapes of the molecules are slightly


different, each behaves uniquely under high field.










RT' 000-1589 SM' 7B
064 NL
100--( 361E2
95: TIC MS 24
9012,4-DNT and 26
85. \CV 110.8V




65- 2,6-DNT
"i CV 12.2V


5079



40 0 7 95
1 445

200



II 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Time (min)



Figure 3-10. CV spectrum of both isomers of DNT. A mixture was created using about 20 ppm
of both isomers in equal proportion. This is a relatively small amount, hence the low
ion transmission. The CV scan was conducted slowly from only 10-15V over 16
minutes. Both isomers are clearly resolved from each other, as shown in Figure 3-10.
It should be noted that there was a [2,6-DNT-NO]- fragment peak that appeared
during the characterization (as discussed in Chapter 2, section 2.6). This fragment did
not appear while using the high-resolution setup in the 10-15V CV range.

3.5 Dispersion Voltage

Another effect that the IMS added was that the optimal DV was no longer the highest

possible (-4000V); the ion signal was nearly zero above a DV of -3000V. The FAIMS carrier

gas flow rate also had to be reduced to about 0.5 L/min from 4.0 L/min for optimum

performance. This is in contrast to the situation with FAIMS by itself (as in chapter 2), where

optimum sensitivity is observed at maximum DV and high carrier gas flow rate. There are

several possible reasons for this unusual behavior. For example, higher flow leaving the curtain

plate may cause ion scattering at the IMS/FAIMS interface. This may be overcome by closing









the system with a nonconductive sleeve that encloses the space between the IMS and FAIMS.

This might also make directing the ion beam into the FAIMS easier. Another possible

explanation is that the small difference in potential between the last electrode of the IMS (-2kV)

and the curtain plate (-1kV) required carrier gas flow to be decreased to allow the potential to

transport ions efficiently into the FAIMS cell.

3.6 Effects of Adjustment of Plate Gap

As discussed above, the distance between the inner and outer electrodes of a FAIMS cell

affects the electric field generated. Since the electrode gap of the dome cell can be adjusted

(figure 3.3), data were collected to determine the optimal electrode gap for explosive detection.

The cylindrical cell electrode gap is fixed at 1 mm; whereas, the gap in the cylindrical

portion of the dome cell is fixed, the gap in the dome portion is adjustable up to 2.4 mm in

increments of 0. 1 mm. Four data sets were taken with a fixed DV at -3000V, each consisting of

two CV scans from 0-20V over a period of 8 min each. The sample selected was a solution of

about 15 ppm RDX, which was chosen because it has a relatively low fragmentation temperature

(about 1700C) compared to the other explosives. The FAIMS/MS behavior of both the [M+Cl]~

ion and the fragment ions were analyzed.

The expected relationship between resolution and electrode gap is that the lowest gap

should provide the highest resolution, as the field is stronger. As expected, the ion count was the

lowest, as there is a lower mean free path; most ions collide with the electrodes.











~adz 310
. as CV = 11.0V


RT 000-1600 SM 7B
10 -

0 RD
a- CV

7-
70




a .


NL
2 46E4
TIC MS


X+ NO3
= 7.8V


m/Z 310


RDX+ NO3~
CV = 7.8V






12 1

10 97 1~131





1or 0 8


RDX+CI-
CV = 16.0V


RDX+CI-
CV = 16.0V


5 67 8 9
Time (mmn)


10 11 12 13 14 15


Figure 3-11. RDX CV spectrum at 0.5 mm. The first trial was at a relatively small gap of 0.5
mm (Figure 3-11). The CV was scanned from 0-20V over 8 minutes twice. On the
figures, ion count is seen after "NL." This figure shows three peaks. One of these
(m/z 310) is probably a solvent cluster. A second (m/z 283) is the nitrate adduct of
RDX. The third (m/z 257) is the RDX chloride adduct. In this case, the RDX analyte
of interest is the smallest peak compared to the solvent. Probability of transmission
of RDX molecules relative to solvent indicates that when the electrode gap is
smallest, most RDX molecules collide with the electrodes and are lost. The
maximum ion signal is 2.46 x 104 COunts.
















4$ X6,~




2~68


100 191 200 250 3131 35i0 4110


~Y~6-S~SI
RT:6~3;rS,71 ~
~ hM 1.16f~ T:-~








~C~42lgl
R51.'SPO3,37 M
4dl Ne I.1BB T.-~








~ItSt9r8
RT. 4.15-4.54 CWF
X NL ~,3r a ;r, p~


20-


~1UB


as9 lo2~9


ssr.~ C PF;3 3557


2509 271.7 /7 310.1


3753


Figure 3-12. RDX mass spectra at 0.5 mm. Figure 3-12 shows the mass spectra for each of the
three CV peaks at an electrode gap of 0.2 mm. Relative to the previous figure, the

top mass spectrum is the right-most CV peak. The RDX mass spectrum has an
additional peak at m/z 267, which is an adduct ion [RDX+NO2]- that co-elutes

through the FAIMS cell with the chloride adduct. This second adduct is not unusual,

though it is usually found in extremely low intensity compared to the chloride adduct.












RT:0 00-12 86 SM 7B
100- m?/Z 310 4.36 4 43 RDX+CI- 2E4
asCV = 10.7V bn CV = 18.0V TIC MS

851i 4 58


71
70- 1 6 69 RDX+ NOs-
esRDX+ NOs~ CV = 7.3V

CV I 477' 26.24





40.32 2 BB4 1 5 89
35- 3 4 4 7.6010 410hl 8 711.5 1.82
30. ~~~2.81771 771 2
3 34 I 7.95 10


25-27 712 2
20-7 265 570


11 2 1 2 2 9
II 1 2 3 4 5 6 7 8 9 10 11 12
Time (mmn)



Figure 3-13. RDX CV spectrum at 0.75 mm. The second trial done was at a gap of 0.75 mm,
which is 1.5 times larger. Again, the CV was scanned twice from OV to 20V over a
total period of 16 minutes. As expected the ion count is slightly higher. With this
larger gap, the resolution is lower (CV peaks are wider); furthermore, the relative
intensities of the different ions appear different (in particular, the RDX [M+Cl]- ion is
significantly more abundant). The maximum ion count is 2.92x104 COunts. Due to an
error, only the first 12 minutes are shown.




















IS* -~jl I


CV = 10.7V



1BS 1326 I10.B 233.0I 287.1 21167 13223 354.I2 3M% 2 391.B


100 150 205 250 300 351) 400


RaT 504~7,2 AYI
id8 NL~1 20E3T -


CV = 18.0V


59NLR958E2T:-p







RT 3 9519 4 N A
83 NL 24~6E T: 0
faistes
soonceaoo


CV = 7.3V


6r19 889


218 7 224t 91 277


328.1 341 8 32 9


Figure 3-14. RDX mass spectra at 0.5 mm. Figure 3-14 displays the mass spectra corresponding
to the three CV peaks in the previous figure. These are similar to those with the
smaller electrode distance, as each ion is well resolved.


1_11














PT 00 -1600 SM 7B
700

95 736


85- 689
80-
75-
70-
65 .
~60
55 55
S503 I
S45 m/Z 310 (
aso0 CV = 8.2V
35 650

30
25
20-
15-I64
10~ 273 27301 617
0518261 359 42594


NL
684E4
RDX+CI~ TIC MS
CV = 18.5V


RDX+CI-
CV = 18.5V


1521
1497
14 94
1486

51


m/z 310
7-83 CV = 8.2V 1470

14 58


794 I1434



O 3 1419


111 234 5 67 8
Time (min)


9 10 11


12 13 14


15 16


Figure 3-15. RDX CV spectrum at 1.0 mm. Next, the electrode gap was increased by a factor of
1.3, to 1.0 mm (Figure 3-15). The ion count was higher, yet the transmission
behavior is completely different from the smaller gaps. This CV spectrum appears
much like those in previous chapters, with a well-resolved peak for the analyte ion
and only small background. The RDX [M+Cl]- ion is much more intense, the peak at
m/z 310 is much less intense, and the peak at m/z 283 is hardly visible. The
maximum ion intensity has increased to 6.84x104 COunts.
























































Figure 3-16. RDX mass spectra at 1.0 mm. The mass spectra for RDX taken at a FAIMS
electrode gap of 1.0 mm can be seen in Figure 3-16. Thus far, both ion count and
resolution are superior to those at smaller electrode gaps. At this gap, resolution and
ion count are well-balanced. Likely, the electric field is at a strength optimal for ion
focusing. Note that the peak at m/z 267 is barely visible.


U jl~je


II ~_ I_ ____ I__I_ I _


II


_11 I


RT 6 7 4.31 AVr
49 NL 5 6'iEJ 3 T p


CV = 18.5V


rF9: lsrs :~0? !I;O


A;;


RT 2514 8 A
148 N 21.51.1EJ
pFlls~21S715
5011-40 00]


CV = 8.2V


11131
15r 9 2T5.


_ta~i.9 3~97 3~2:8 j~7 g


r0D 1M


C "''1
10D 5401 300
me:











RT 0 00 -1600 BM glJ/Z 310*
100, CV = 10.


187E5
TIC MS


Time (min)


Figure 3-17. RDX CV spectrum at 2.0 mm. The last trial was done with an electrode gap of 2.0
mm (Figure 3-17). If the distance is set too high, as in this case, then the electrodes
can no longer carry a waveform. This ion count was the highest of all these trials,
presenting three orders of magnitude greater transmission than the smallest setting at
0.5 mm. The RDX peak was robust and well-resolved; however, the peaks at m/z 283
and m/z 310 were again visible.























"iCV = 10.2V







CV = 8.7V


P/8 NL10EA


CV = 19.6V


~rii


c?,. -sl


WpSlI 4 g9E4Tl





9 00-40900|


I~il IM :Ot! ~W rO
hlP2


Figure 3-18. RDX mass spectra at 2.0 mm. The mass spectra in Figure 3-18 show that the ion
counts are the highest compared to the spectra taken at other electrode distances.
These four trials confirmed that a distance between the inner and outer FAIMS
electrodes in the dome region set at about 1.0 mm yields the optimal transmission and
resolution for explosives. It was observed that the CV for the chloride adduct of
RDX increased as the electrode gap was increased, while the other peak CVs
decreased (allowing the peaks to be resolved), but when the gap was at 2.0mm, the
m/z 283 and m/z 310 peak CVs increased.










Table 3-1 shows the intensities of each of the major ions discussed above. As the electrode

gap was increased, the intensity of the chloride adduct of RDX increased by almost two orders

of magnitude, while the m/z 3 10 and m/z 283 ion counts decreased until the electrode gap was

set to 2.0 mm, where the trend reverses.

Table 3-1. Relative lon Intensities.
Electrode gap (mm) [RDX+ NOJ]" m/z 310 [RDX+CI]~
(counts) (counts) (counts)
0.5 1.5x10 2.5x10 2.5x10
0.75 1.2x10 2.9x10 2.5x10
1.0 5.7x10 6.0 x10 6.8x10
2.0 6.0 x10 2.0 x10 2.0 x10



Table 3-2 lists the CV values for the same ions. The behavior mentioned above occurs for

the CV values as well as the ion intensities. The chloride adduct of RDX increases slightly as

the electrode gap increases, and the other ions decrease until the electrode gap is set to 2.0 mm,

where the CVs increase.

Table 3-2. Relative CV Values.
Electrode gap (mm) [RDX+ NOJ]" m/z 310 [RDX+CI]~
(V) (V) (V)
0.5 7.8 11.0 16.0

0.75 7.3 10.7 18.0
1.0 5.9 8.2 18.5
2.0 8.7 10.2 19.6



To examine the ultimate resolution of the FAIMS system, an attempt was made to resolve

[M]- from [M-H]- ions for individual explosive compounds. Unfortunately, none of the

explosives have both of these ions in approximately equal proportion; this makes it more difficult

to allow for the separation of the ions. Nevertheless, for other types of ions, resolution is

certainly feasible. To maximize resolution, the CV scan would have to be extremely slow or





RDX














CV and DV created using data sets taken using the FAIMS/MS setup described in
chapter 2. The curves indicate that the CV of each ion increases at more negative DV
(hence greater field strength). TNT and other nitroaromatics do not produce ions that
are transmitted at a positive DV, although the chloride adduct ions of RDX and HMX
respond to both positive and negative DVs. The lonalytics instrument will not allow
the DV to be set less than & 2000V or greater than & 4000V; this is indicated by the
vertical red lines. Because of this DV/CV relationship, most of the explosives appear
to exhibit "B" type behavior when high field is applied.


over a narrow CV range, and the gas (and thus ions) entering the FAIMS cell would need to be

completely desolvated. Different adducts of the same explosive, however, have been separated,

for instance, the [M+Cl]- and [M+NO3]~ ions of RDX (see section 3.5).

3.7 Relationship of DV to CV

As discussed earlier, there is no single set of conditions that optimizes separation of every

mixture for every FAIMS-based instrument. In order to characterize a FAIMS-based separation

instrument, CVs must be found based on field strength, which is mainly a function of DV and

cell geometry. There is, however, the same relationship between DV and CV for all FAIMS

instruments, as this relationship is based on the change in mobility for an ion between low and

high field.12









3.8 Sensitivity

Data sets collected on the FAIMS instrument used for characterization (as described in

chapter 2) show that the DV for the optimal separation (highest resolution with lowest

background) for the nitroaromatic explosives is -4000V, and +4000V for the nitramines. The

IMS/FAIMS instrument (as described in this chapter) did not display the same behavior. The

peak DV was at about -2500V, and at any DV set higher than -3000V the ion signal decreased

rapidly to zero. In addition, the flow rate of the FAIMS carrier gas for this instrument could not

be set higher than 1 L/min, or ions were lost. There are several possibilities for this behavior; the

desolvation and focusing attributes of the IMS most likely require that a lower amount of carrier

gas and voltage are needed to achieve the same separation results.

One method to measure sensitivity and resolving power of a FAIMS instrument is to

examine the CV spectrum for both peak width and ion intensity. As discussed earlier, there is a

degree of ion loss through a FAIMS (data gathered during this research shows about two orders

of magnitude compared to no FAIMS cell in the system). Most FAIMS instruments (including

the instrument used for chapter 2) demonstrate an average CV peak width of about 2-3V;40 in

contrast, the best HRFAIMS system CV peaks were as narrow as about 0.75V (for examples, see

Eigs. 3-6 and 3-8). This has both benefits and disadvantages; narrower peaks require slower scan

speed to define the peak, and necessitate precise CV values when Eixed, but the narrower peaks

are better resolved from each other and from background ions.

3.9 IMS/FAIMS of Explosives

In another set of experiments, the IMS was operated in gating mode in order to attempt to

add an additional dimension of separation. This involved manually setting the times at which the

gate would pulse open with a delay to allow ions of a fixed drift time (and, thus, ion mobility) to

pass through the drift tube.









All other ions would be excluded from transmission. Hence, a percentage of all ions were

discarded by the IMS, in addition to those discarded by the FAIMS portion of the instrument.

An attempt was made to separate explosive mixtures by gating the IMS; however, ion counts

were not high enough to discern individual components of the mixture. This is largely due to the

loss in sensitivity due to the low duty cycle of the IMS/FAIMS system. For instance, if a 500 ms

gate width is used, and the gate is only open for 25 ms, then the duty cycle is only 5%, which

results in a significant loss of signal. Furthermore, the lack of synchronization between the IMS

gate pulses and the fill times of the LCQ quadrupole ion trap may result in even greater loss in

sensitivity .

Gating the IMS was not critical in these studies, as data have indicated that individual

explosive peaks as part of a mixture are fully resolved while the IMS is functioning simply as a

desolvation and focusing apparatus for the FAIMS. Any future experiments should include

efforts to increase ion signal when the IMS is gating. (For full discussion see Chapter 4)









CHAPTER 4
CONCLUSIONS AND FUTURE WORK

4.1 Conclusions

The United States is under constant threat from the use of explosives by her enemies.

There is a great need for the development of technologies that can eventually be applied in

instruments that will be used to detect improvised explosive devices. These analytical

instruments mainly will consist of laboratory and intermediate-sized instruments or man-portable

Hield instruments, and all will be utilized both overseas and domestically.

Instruments designed for the detection of explosive compounds can fall into two

categories, depending on function and/or mission. These are either qualitative threat detectors,

or quantitative threat characterizers.41 Factors that affect instrument use are portability, resolving

power, sensitivity, and selectivity.

Mass spectrometry is a widely used analytical methodology. When a separation device is

coupled to a mass spectrometer, all of the analytical figures of merit are drastically improved.

FAIMS is a promising technology that functions well as a separation device and is

compatible with a mass spectrometer. There are many advantages to employing FAIMS as such.

Even though the use of FAIMS may lower overall ion transmission, the signal-to-noise ratio of

the overall instrument may be much higher than a lone mass spectrometer. This can be

extremely important when an instrument is used in a Hield environment, away from the

laboratory, when there is a great deal of background matrix with which to contend.

The goal of this research was to determine the degree to which a FAIMS/MS device can be

employed to detect explosive compounds that are most likely to demonstrate a threat, and to

attempt to ascertain methodologies to eventually do so in a field environment. The experiments

discussed herein consisted primarily of three parts. The first characterized explosive compounds










using a FAIMS/MS, while the second included additional FAIMS/MS experiments conducted to

analyze and improve the analytical factors of the instrument. The third section described a novel

high-resolution FAIMS/MS method, employing an IMS drift tube to desolvate and focus ions

into the FAIMS cell. The HRFAIMS method is sensitive, and the resolution is high enough that

valuable forensic data can be collected.

4.1.1 Characterization of Explosive Compounds

A list of explosive compounds was compiled based on both the frequency of encounter

during terrorist events and the presence in common military explosives. The feasibility of

detection of these compounds was first evaluated; once accomplished, the characteristics of each

compound were analyzed and noted. All of the explosives in this study contained nitro leaving

groups. This list includes three nitroaromatics: TNT, which is one of the oldest known and

commonly encountered explosives, and two isomers of DNT that are precursors to the

manufacture of TNT, and, therefore, are forensically important. TNT is also found in many

military explosive mixtures. All of the nitroaromatics were easily detectable and were readily

characterized using the FAIMS/MS instrument.

There were two nitramines evaluated, RDX and HMX, which are encountered along with

TNT in military explosives and plastic explosives mixtures. The most common explosives

include mixtures of TNT and RDX. Both RDX and HMX are similar molecules that have

similar characteristics. The formation of a molecular-type ion requires adduct formation to be

well detected. The best adduct for this purpose is formed with the addition of a chloride (Cl-) ion

(improved by doping the solvent with a small amount of a source of chloride, such as CCl4),

though all will form adducts with nitrate (NO3 ) aS Well.

There were two nitrate esters evaluated, NG and PETN. Both are extremely unstable. The

molecular ion cannot be detected by MS; PETN will only form adducts in low concentrations.










NG will not form adducts at all. Because of this instability, though, NG and PETN are rarely, if

ever, used without another explosive in a mixture.

The last explosive evaluated in this study was ANFO, a mixture of ammonium nitrate and

a fuel, which is a commonly used commercial explosive. It is also favored by terrorists due to

the ease of its manufacture and acquisition of its components. The nature of these components

creates difficulties for detection, as ammonium nitrate and many of the fuels used (usually

diesel) may be found in high levels in the background. Because there is no single structure,

ANFO has not been well characterized for detection by MS.

4.1.2 FAIMS Characteristics

FAIMS separates ions using a high electric field to affect the change in ion mobility. One

of most important FAIMS characteristics that affect the strength of this field is the dispersion

voltage. Once the optimum DV was determined, then other analytical parameters could be

ascertained for each explosive compound. No two types of FAIMS instruments can be

characterized identically, though there is a basic relationship that applies in all cases between DV

and CV.

Other FAIMS factors that were examined were the addition of a small amount of helium to

the carrier gas, which improves resolution and ion transmission, and the distance between inner

and outer electrodes of a FAIMS cell. The optimal electrode gap for the detection of explosives

on the instrument used corresponds to other FAIMS studies that have been done (a distance of

1mm).

4.1.3 IMS/FAIMS/MS

During these experiments, an IMS was added to the instrument to add another dimension

of separation. When the IMS was gated, only ions with a specific ion mobility were transmitted

on to the FAIMS cell; therefore, only a fraction of the ions were transmitted. The ion counts









during this experiment were not high enough to distinguish individual components of a mixture.

Changes could be made to the instrument to attempt to improve transmission and sensitivity.

This is discussed further in the "recommendations" section of this chapter.

4.1.4 High-resolution FAIMS

The addition of the IMS drift cell was proven to be beneficial to the instrument, however.

When the IMS was in total ion mode, there was no ion gating, and all ions passed through the

drift tube. Because the tube is heated and carries an electric field, the analyte was desolvated,

and the ion beam was focused into the FAIMS. When this method was used along with the

optimal FAIMS parameters, resolving power of the FAIMS, and ion transmission increased

drastically. At certain junctures, the CV peaks were so narrow that the CV scan speed had to be

slowed to allow the entire peak to pass through before the CV changed.

4.1.5 Mixture Separations with FAIMS/MS

Once characterization was completed, then the value of the FAIMS to separate mixtures

was analyzed. If an analytical instrument is unable to resolve mixtures, then its value as a field

instrument is limited. This phase of experiments was greatly enhanced using the HRFAIMS

method. First, two mixtures containing different explosives were separated successfully and the

components were well-resolved. Resolution was high enough that mixtures containing two

isomers of the same compound (DNT) were also successfully separated from each other. This

ability can be valuable for forensics analysis of either post-blast or manufacturing of TNT.

4.1.6 Disadvantages of FAIMS

FAIMS instruments still have limitations, especially pertaining to field use. The resolution

and transmission of the FAIMS cell are strongly affected by water and other solvent, and the cell

must be as dry as possible. After continued use, if the FAIMS cell is not periodically dried out,

the sensitivity and resolving power drop as either the electrodes are hindered from carrying a










high electric field, or the saturation of solvent causes ion-molecule reactions. As such, the

carrier gas must not contain impurities; furthermore, carrier gas is required in large volumes

(between 3-5 L/min). This can provide issues in terms of instrument period of use, maintenance,

and design, for it is undesirable to add large cylinders of gases to an apparatus if the obj ective is

portability.

The lonalytics prototype instruments are extremely delicate, and are not compatible with

field use. For instance, the DV is transported from the waveform generator to the FAIMS cell

via an insulated cable. Any environmental changes in the area around the cable can significantly

affect the waveform and field generated.

4.2 Recommendations

Analysis of data sets collected during this series of experiments has determined that

FAIMS/MS is a feasible method to use for the detection of explosive compounds. A high-

resolution methodology for the detection of explosives with FAIMS/MS has been identified

during this research study, and should be further explored.

4.2.1 Further HRFAIMS Improvements

The use of the IMS as an ion focusing/desolvation device could be further improved,

because its use during this study was not intended to be so. The interface between the IMS and

FAIMS consisted of an empty space between the two components. The IMS drift tube must be

lined up with the FAIMS cell manually, and both the electric fields as well as carrier gas may

cause ions to leave the system. A nonconductive sleeve could be used to enclose the system, and

may decrease ion loss, and may improve transfer into the FAIMS cell.

4.2.2 FAIMS Carrier Gas Experiments

Further, experimentation could be conducted into the composition of the FAIMS carrier

gas. Future field instruments will probably not include carrier gas cylinders. Other gas mixtures









could be tested, and the effects determined. A desirable approach would be, first, to use

laboratory-provided pure, dry "zero air," and if this is effective as a FAIMS carrier gas, then to

use ambient air run through a scrubber/dessicant just prior to introduction to the FAIMS cell.

4.2.3 Practical Separations and Analysis

Because the separation of mixtures was done successfully, the next logical step is to

separate unknown mixtures by, for example, having another individual prepare the samples.

Also, efforts should be made to separate explosive compounds from various background

matrices, thus determining the lowest possible concentrations at which explosives can be

detected. After this is done, then a study should be done to identify matrices (if any) that

strongly interfere with detection, and attempt to ameliorate them.

4.2.4 IMS Gating

The IMS/FAIMS/MS method requires more efforts to be made in terms of synchronizing

times when the gates open to the FAIMS CV scanning, and the inj section times of the quadrupole

ion trap. This will further ensure that the lowest number of ions possible is lost.

4.2.5 CV Scanning

A practical instrument would require scan speeds that are as fast as possible, so that

possible threats can be detected and warnings can be administered. CV scan speed can be

adjusted to any length of time, though the higher resolution the instrument, the slower the scan

speed that is necessary. For practical use, an optimal length of time should be set such that

mixtures are fully resolved, but the instrument can scan as quickly as possible. The CV scan

range can also be adjusted so that all of the explosives of interest are covered, yet it is set to the

minimum amount of time needed. If a qualitative threat warning is all that is required, then the

scan time can be minimized to allow just enough of the analyte ions to pass through the system

to be detectable above the noise level, thus further reducing the time required for a scan.









4.2.6 Instrument Maintenance

Both IMS/FAIMS instruments used in this research introduced samples into the system as

liquids. As such, the sensitivity and resolution of the FAIMS degraded over time, and regular

cleaning of the cell with a volatile solvent that did not leave a residue, such as methanol,

followed by drying of the cell were needed. Sampling explosives in a vapor or particulate

sample, rather than in liquid solution, may reduce this problem. Practical instruments should

include a small amount of solvent for regular cleaning; and note that carrier gas should always be

run through the system when in standby.

4.3 Future Work

Now that feasibility has been established, and the first step toward method development

for explosives detection via FAIMS/MS has been taken, the next step includes working toward

developing an instrument employing this method. A great deal of effort is required, however, to

design and create an apparatus that can be employed for use in a Hield environment. The best

way to approach this would be to Birst design and test an intermediate instrument that would be

used specifically for explosives and other related compounds. This work would likely include

further research in interfaces between each component of the instrument, FAIMS cell

geometries, and other ways to allow the instrument to serve its purpose. Also, a method of

practical sample introduction to the system is needed, as explosives have low vapor pressures

and are not conducive to collection via gas collection.

4.3.1 Ionization Sources

One possible way to move toward development of a Hield instrument was actually

attempted during this research study: the use of a more portable ionization source. ESI is

designed for ionizing liquid solutions, and APCI requires a heater and corona discharge; both of

these limit the capabilities of an instrument.










A novel type of ionization source was evaluated during this research for use with an

IMS/MS for the detection of explosives. Developed by Blanchard & Co., the distributed plasma

ion source (DPIS) is designed to provide ionization at atmospheric pressure using a simple,

portable, low-power device. The DPIS is constructed with two overlapping electrodes of

different sizes that are separated with a dielectric. A potential applied across the electrodes

creates a distributed plasma around the edge of the small electrode, which ionizes the

surrounding air and sample vapor. This type of source can be useful because sample can be

introduced as a vapor. In this case, liquid sample was kept in a bottle, across which a nitrogen

carrier gas was blown, and the headspace was directed over the DPIS source, into one of WSU' s

homebuilt IMS tubes.





















Figure 4-1. Distributed plasma ionization source. Figure 4-1 is a photograph of the DPIS
extended through the ion screen of the IMS drift tube. The plasma discharge is a
distinct blue color. The sample line is taped to the insulated wire that powers the
source.

The DPIS circuit has limitations, however. There is no diode to prevent overload of the

circuit, and the maximum potential cannot exceed -1kV. Data were collected with several

explosives. The DPIS source did ionize the samples, but the signal-to-noise ratio was extremely









low. There may be many reasons for this. First, the source voltage is low compared to the high

voltage of the IMS, which may limit transmission of ions into the IMS drift tube. Also, the only

sample directed into the instrument is headspace, which contains significantly less analyte than a

liquid which is vaporized.

This source is not compatible with laboratory-scale IMS/MS instruments, as the ionization

voltage is not scaled with the other voltages of the system. Nevertheless, the DPIS did ionize the

explosive samples, and can potentially be used with low-power, portable Hield instruments.

4.3.2 Development of Field Instruments

The first step in this research has been to determine the feasibility of the IMS/FAIMS/MS

method for detecting explosives. Now that this has been accomplished, improvements can be

made to the overall method (as discussed above). Once this is done, then work can proceed to

reduce size and power consumption of each component, and then durable packaging can be

developed for a Hieldable instrument.










REFERENCES


1. Country Reports on Terrorism, US State Department, 2005.

2. Opportunities to Improve Airport Passenger So I emling~ \ ithl Mass Spectrometry,
Committee on Assessment of Security Technologies for Transportation, National
Materials Advisory Board, 2004.

3. Shea, Dana A.; Morgan, Daniel, Detection of Explosives on Airline Passengers:
Recommendation of the 9/11 Commission and Related Issues, Congressional Research
Service, Library of Congress, 2006.

4. Eiceman G. A.; Borsdorf H., lon M~obility Spectrometry: Principles and Applications,
Applied Spectrometry Reviews, 2006, vol. 41, no. 5, pp. 323-375.

5. Hill, Herbert H.; Simpson, Greg. Capabilities and Limitations oflon M~obility
Spectrometry for Field Screening Applications, Field Analytical Chemistry, 1997, vol.1,
no. 3, pp. 119-134.

6. Lee, Harold G.; Lee, Edgar D.; Lee, Milton L., Atmospheric Pressure Ionization Time-
of-flight Mass Spectrometer for Real-time Explosives Vapor Detection, Proceedings of
the First International Symposium on Explosive Detection Technology, pp. 619-63 3,
1992.

7. Osorio, Celia; Gomez, Lewis M.; Hernandez, Samuel P.; Castro, Miguel E., Time-of-
flight Ma~ss Spectroscopy M~easurements of TNT and RDX on Soil Surfaces, Detection and
Remediation Technologies for Mines and Minelike Targets, 2005, v ol. 5794, pp. 803-
811.

8. B ottrill, A.R., High-energy Collision-induced Dissociation of2acromolecules using
Tandem Double-focusing/Time-of-f light Ma~ss Spectrometry, PhD thesis, University of
Warwick, Coventry, UK., 2000.

9. Gruber, Ludwig. Characteristics of Different Ma~ss Analyzers, Fraunhofer-Institute
for Process Engineering and Packaging, http://www. ivy.fraunhofer. de/ms/ms-
analyzers.html, 2000.

10. Purves, Randy W.; Guevremont, Roger; Day, Stephen; Pipich, Charles W.;
Matyj aszczyk, Matthew S., Ma~ss Spectrometric Characterization ofa High-field
Asymmetric Waveform lon M~obility Spectrometer, Review of Scientific Instruments,
1998, v ol. 69, no. 12, pp. 4094-4106.

11. Kolakowski, Beata M.; McCooeye, Margaret A.; Mester, Zoltan; D'Agostino, Paul
A., Rapid Separation and Identif ication of Chemical and Biological Warfare Agents in
Food and Consumer Matrices Using FAIM~S-M~S Technology, National Research Council
of Canada, 2006.










12. Guevremont, Roger, High-Field Asynametric Waveform lon M~obility Spectronsetry
(FAIM~S), lonalytics Corp., Ottawa Canada, 2004.

13. Reich, Richard, Fundamentalsd~~~dd~~~ddd~~ and Applications ofAtnzospheric Pressure Chemical
ionization Quadrupole lon Trap M~ass Spectronzetry for the Analysis ofExplosives, MS
thesis, University of Florida, 2001.

14. Finnigan2 LCQ Series APPL APCI Combination Probe Operator 's Manual, Rev. A,
Thermo-Electron Corp., 2004.

15. Buryakov, I.; Krylov, E.; Nazarov, E.; Rasulev, U., A New M~ethod ofSeparation of
M~ulti-atontic Ions by M~obility at Atmospheric Pressure Using a High-Frequency
Anaplitude-a;synametric Strong Electric Field, International Journal of Mass Spectrometry,
1993, vol. 128, pp. 143-148.

16. Kanu, Abu B.; Hill, Herbert H., lon M~obility Spectronzetry: Recent Developments
andNovel Applications, Washington State University, LabPlus International, 2004.

17. Atkins, Peter; dePaula, Julio, Physical Chentistly, 7th Ed., Chap. 24, Oxford Univ.
Press, Oxford, UK, 2001.

18. Welling, M.; Schuessler, H.A.; Thompson, R.I.; Walther, H.; lon M~olecule
Reactions, Ma~ss Spectronzetry and Optical Spectroscopy in a Linear lon Trap,
International Journal of Mass Spectrometry and lon Processes, 1998, vol. 172, no. 1, pp.
95-114.

19. De Hoffman, Edmond and Stroobant, Vincent, Ma~ss Spectronzetry Principles and
Applications, 2nd Ed., 2005

20. Paul, W., Electromagnetic Traps for Charged and Neutral Particles,~~PPP~~~PP~~~PP Modern
Physics, 1992, vol. 62, no. 3, pp. 531-540.

21. Finnigan2 LCQ Series Hardware Manual, Rev. A, Thermo-Electron Corp., 2003.

22. Mathieu, E., 2Menoire sur Le M~ouvenzent Vibratoire d'une M~embrane de fornze
Elliptique, Journal of Pure Mathematics, 1868, vol. 13, pp. 137-203.

23. McLachlan, N. W., Theory and Application of2athieu Functions, Dover, 1947.

24. March, Raymond E.; Todd, John F. J., Quadrupole lon Trap Ma~ss Spectronsetry,
Wiley, 2005.

25. Murray, K. K.; Boyd, Robert K.; et. al., Standar~dd dddD efinitionsd~~d~ of Terms Related to
Ma~ss Spectronzetry, IUPAC, 2006.

26 Sparkman, O.D.; Klink, F. E., LC M~S: Fundamentalsd~~~dd~~~ddd~~ and Applications, ACS, 2001.










27. Annual Threat Assessment of the Director ofNational Intelligence for the Senate
Armed Services Committee, Statement by the Director of National Intelligence, John D.
Negroponte to the Senate Armed Services Committee, transcript available on
www.globalsecurity. org, Feb. 2006.

28. Davis, Tenney L., The Chemistry of Powder and Explosives, Ch. 1, AngriffPress,
Hollywood, CA, 1972.

29. Wade, L. G., Organic Chemistry, 6th Ed., Ch. 15-16, Prentice Hall, 2005.

30. Hannum, David; et. al., M~iniaturized Explosives Preconcentrators for Use in Man-
portable Explosives Detection Systems, Sandia National Laboratory, 1999.

31. Committee on the Review of Existing and Potential Standoff Explosives Detection
Techniques, Existing and Potential StandoffExplosives Detection Techniques, National
Research Council, National Academies Press, 2004.

32. Yinon, Jehuda, Toxicity and Metabolism ofExplosives, CRC Press, Boca Raton, FL,
1990.

33. Cooks, R. Graham; et. al., Direct, Trace Level Detection of Explosives on Ambient
Surfaces by Desorption Electrospray ionization Ma~ss Spectrometry, Royal Society of
Chemistry, 2005, vol. 15, pp. 1950-1952.

34. Federal Facilities Assessment Branch, Public Health Assessment, US Army Umatilla
Depot Activity, Centers for Disease Control and Prevention, 1997.

35. Cooper, Paul W., Explosives Engmneering, Ch. 3, Wiley, New York, NY, 1996.

36. Fenn, John B., Electrospray Ionization, Principles and Practices, Mass Spectrometry
Reviews, 1990, vol. 9, no. 1, pp. 37-70.

37. United States patent 7071465: lon mobility spectrometry method and apparatus,
inventors: Hill, Herbert H. et. al., Washington State University research foundation,
2006.

38. Bradbury, Norris E., and Nielsen, Russell A., Absolute Values of the Electron
Mobility in Hydrogen, Physical Review, 1935, vol. 49, no. 5, pp. 388-393.

39. Guevremont, R.; Thekkadath, G.; Hilton C. K., Compensation Voltage (CV) Peak
Shapes Using a Domed FAIM~S 0 ithr the Inner Electrode Translated to Various
Longitudinal Positions, Journal of the American Society of Mass Spectrometry, 2005,
vol. 16, no. 6, pp. 948-956.

40. Schvartsburg, A.; et. al., High-Resolution FAIM~S Using New Planar Geometry
Analyzers, Analytical Chemistry, 2006, vol. 78, no. 11, pp. 3706-3714.










41. Transportation Security: Systematic Plan2ning Needed to Optimize Resources,
statement of Cathleen A. Berrick to the US Senate Committee on Commerce, Science,
and Transportation, USGAO, 2005.









BIOGRAPHICAL SKETCH

Jared J. Boock was born in Hazleton, PA, in 1980. He grew up in Torrington, CT,

and Panama City, FL. He completed his B.S. in chemistry from Florida State University

in 2002 and received a commission in the US Air Force.

His first duty station was the Molecular Sciences Division, Materials Technology

Directorate, Air Force Technical Applications Center, Patrick AFB, FL, where he served

as the division Test Director. His primary duty was to plan and execute field tests, and

evaluate data for the test and evaluation of chemical weapons detection equipment. He

wrote several classified test and technical reports.

In 2005, he was selected for the Air Force Institute of Technology Civilian

Institutions program, to study for his M. S. degree. He chose to pursue his M. S. degree in

analytical chemistry at the University of Florida. Upon completion of his M.S. degree he

will be stationed at the High Explosives Research and Development Facility at Eglin

AFB, FL.