ORNL research on glow discharge ionization, electrospray (see Fig. 1), and quadrupole ion trap mass spectrometry has led to tools that could help detect hidden explosives, monitor hydrocarbon levels in low-emission vehicles being developed, and determine structures of biomolecules such as DNA and proteins.
|
|
|
Fig. 1. Scott McLuckey adjusts the first electrospray/ion trap mass spectrometry instrument that he helped develop at ORNL in the late 1980s. |
Mass spectrometry can trace its roots to Sir J. J. Thomson, who discovered the electron in 1897. Thomson measured the charge-to-mass ratio of the electron, thus pioneering mass spectrometry over 100 years ago. Thomson’s student Francis William Aston invented a device that demonstrated the existence of stable isotopes. Called a mass spectrograph, the device sensitively separated isotopes of the same element based on differences in the deflection of ions in a magnetic field; because the ions of each isotope differ in mass and charge, they can be magnetically separated, collected, and counted in a specific part of the instrument. Using an early mass spectrograph, Thomson showed in 1919 that neon has at least two isotopes—one with a mass number of 20 and the other with a mass number of 22. Then, other scientists used mass spectrometry to measure the positions and sizes of the peaks in the mass spectrum to identify isotopes in the periodic table and determine their relative abundances.
Since then, the power of the analytical technique, in its many forms, has been exploited in a wide range of scientific disciplines, from atomic physics to molecular biology. Its applications include dating of ancient specimens, drug testing, nuclear fuel analysis, environmental monitoring, pharmacokinetic measurements, and peptide sequencing, to list but a few. Indeed, when mass spectrometrists gather today, the collection includes physicists, chemists, and a rapidly growing contingent of biologists.
The history of mass spectrometry at Oak Ridge National Laboratory goes back a little over 50 years to its role in the Manhattan Project. It is frequently pointed out that the calutron mass separator, invented by E. O. Lawrence in Berkeley, California, and used at the Oak Ridge Y-12 Plant to separate the isotopes of uranium in the war effort, is something of a "preparative" mass spectrometer. Most mass spectrometers, however, are used not on a production scale but rather as sensitive tools for analysis. The first "analytical" mass spectrometers at Oak Ridge were contemporary with the calutrons and were used to analyze their output. Such analyses continue to be performed today in support of the Department of Energy’s Stable Isotopes Program.
While the history of elemental mass spectrometry, as represented by the measurement of isotope ratios of materials produced by the calutrons, goes back to the early 1940s and the very beginnings of Oak Ridge, mass spectrometry research at ORNL has expanded into new areas over the years as the needs and missions of DOE have changed. Significant effort, for example, has been devoted to research in organic mass spectrometry over the past 20 years. A line of work that began in the mid-1980s is highlighted here as an example of how basic research in organic mass spectrometry has been integrated into several applied programs within DOE and other federal agencies and into both the general basic research and applied analytical chemistry communities.
Glow Discharge, Ion Traps, and TNT
|
|
|
Fig. 2. Front view of an atmospheric sampling glow discharge ionization source. |
At this same time, Sam McDowell of DOE’s Office of Safeguards and Security (OSS) approached Joel Carter, then head of the mass spectrometry effort at ORNL, about possible approaches to the detection of trace vapors of hidden explosives. Carter recognized the potential applicability of the BES research to this problem and presented it to the staff. McLuckey and Glish proposed a combination of the ORNL-developed and patented atmospheric sampling glow discharge ionization source and tandem mass spectrometry, a technique whereby an ion of interest is mass-selected for reaction (usually dissociation) and a second stage of mass analysis is performed on the charged products. The proposal was subsequently approved by the OSS, which supported our efforts to explore this approach.
|
|
|
Fig. 3. A plot of the logarithm of the signal strength vs the logarithm of the quantity of TNT admitted to a mass spectrometer employing atmospheric sampling glow-discharge ionization. |
Ionization to form negative ions combined with mass spectrometry provides good specificity, which translates into low false alarm rates, relative to most other analytical techniques. However, good specificity in the laboratory does not necessarily translate into good specificity in the field. For this reason, ORNL researchers chose to use tandem mass spectrometry to analyze anions produced by the glow discharge. In tandem mass spectrometry, the ions produced by the glow discharge source, often referred to as parent ions, are mass-selected and caused to dissociate prior to a second stage of mass analysis. This procedure requires a molecule to pass three tests before it is identified as an explosive. First, the molecule must form a stable negative ion. Second, the parent ion must have the mass known to be formed from a targeted explosive. And third, the masses of the product ions resulting from dissociation of parent ions must match those of the targeted explosive indicated by the parent ion mass. For example, TNT captures an electron in the glow discharge to form the molecular anion at m/z 227. The molecular ion is used as the parent ion for detection of TNT. The TNT parent ion fragments to a variety of products, but the most abundant product ions arise from loss of the hydroxyl radical to give a product ion at m/z 210 and loss of nitric oxide to give a product ion of m/z 197. These two product ions are typically monitored for detection of TNT.
Although, in principle, the combination of glow discharge with tandem mass spectrometry was attractive for the detection of trace vapors of explosives, fundamental issues that had to be addressed included the efficiency of converting parent ions to product ions and the extent to which ion chemistry both in the ion source and upon ion activation yields useful information. In the early work with explosives, the glow discharge source was coupled with a conventional beam-type tandem mass spectrometer in which two discrete mass analyzers with an intermediate collision region were used in series. In beam-type tandem mass spectrometry, each step of the procedure is carried out in a discrete region of space. Significant ion losses can result from transport from one region to another. This arrangement yielded parent-to-product ion efficiencies of about 1% as a result of relatively poor ion transport efficiency and electron detachment from the parent ion in competition with fragmentation. This situation resulted in a trade-off between specificity and detection limit of nearly two orders of magnitude.
During the mid-1980s, the quadrupole ion trap was beginning to emerge as a mass analyzer with interesting characteristics for tandem mass spectrometry. The quadrupole ion trap operates on the principle that ions can be stored within an oscillating electric field. With appropriately shaped electrodes, an oscillating electric field (usually a quadrupole field or a variation thereof) can be created that stores ions in three dimensions. Furthermore, the amplitude of the electric field can be varied in such a fashion that ions are ejected from the ion trap and into a detector in a mass-dependent fashion, thereby allowing the ion trap to serve as a mass spectrometer. Of particular interest for the explosives detection application were the high parent-to-product ion conversion efficiencies (up to 100%) that had been reported for positive ions. The quadrupole ion trap is one of the so-called "tandem-in-time" instruments, as opposed to the conventional "tandem-in-space" beam-type instruments. As such, all steps of the tandem mass spectrometry experiment are carried out in sequence in the same region of space. Therefore, ion transport losses are avoided. Motivated by the potential indicated in the early ion trap results, the first experiments at Oak Ridge using the quadrupole ion were initiated, marking the beginning of a now decade-old effort at ORNL on research and applications using the quadrupole ion trap which now spans many groups and many sponsors.
|
|
|
Fig. 4. Tandem mass spectrum obtained from 500 fg of the plastic explosive RDX. The peak at m/z 102 corresponds to the loss of CH2 NNO2 from the parent ion and is highly diagnostic for RDX. The inset shows the total ion current obtained as the RDX sample was admitted into the ion source. |
The explosives detection work illustrates the synergy that can develop between applied projects and fundamental research. To make novel approaches succeed often requires basic research in order to understand the underlying phenomena associated with critical aspects of the process. In this case, for example, basic research was needed in electron capture under air glow discharge conditions, ion injection into a quadrupole ion trap from an external ion source, and unimolecular dissociation chemistry of explosives-derived ions. As a result, the data of Fig. 4 could be generated, demonstrating the capability of tandem mass spectrometry to detect ultralow levels of explosives vapors. Figure 3 shows the results of a procedure whereby the plastic explosive RDX (rapid detonating explosive) was monitored. A 500 femtogram (fg), or 5 ´ 10–13 g, sample of RDX was placed at the inlet of the glow discharge ion trap, which was continuously monitoring for RDX by admitting ions, mass-selecting ions at m/z 176, collisionally activating any ions that may appear at m/z 176, and analyzing the products. The parent ion for RDX appears at m/z 176 and its diagnostic product ion appears at m/z 102 (arising from the loss of the neutral carbon-hydrogen-nitrogen-oxygen fragment CH2 NNO2 from the parent anion). Therefore, the appearance of ions at m/z 176 and m/z 102 simultaneously indicates the presence of RDX.
The capability to detect targeted compounds at such low levels with the extremely high specificity afforded by the glow discharge/tandem mass spectrometry experiment has applications beyond that of explosives detection. Teledyne Corporation, for example, has licensed the glow discharge technology and markets a glow discharge–ion trap quadrupole mass spectrometer for both environmental analyses and explosives detection. Mass spectrometry-based explosives detection technology is being considered both for security applications, such as airport security, and for detection of land mines and buried munitions.
Glow Discharge, Ion Traps, and BTX
Roughly five years ago, scientists from the automobile industry and DOE national laboratories conducted exploratory discussions to determine the extent to which national laboratory expertise could help the U.S. auto industry develop "low-emission vehicles" (LEVs). Recent federal regulations, as well as more stringent regulations enacted in California, require auto manufacturers to market LEVs in the near future. Rapid, rugged, cost-effective vehicle emissions instrumentation is required both for developing LEVs and ensuring that they meet regulatory requirements when they come to market. The U.S. automobile manufacturers expressed a need to evaluate quantitatively a variety of trace components in LEV exhaust during vehicle development. Of principal interest was the suite of hydrocarbon compounds known to contribute to smog generation, including benzene, toluene, and the xylenes (these common aromatic molecules are commonly referred to as BTX). No satisfactory technologies were available for providing rapid, high-quality, cost-effective characterization of trace pollutants expected from such vehicles being developed.
ORNL researchers Scott McLuckey, Michelle Buchanan, Kevin Hart, Keiji Asano, and Doug Goeringer proposed a methodology based on glow discharge—quadrupole ion trap mass spectrometry whereby positive ions derived from the hydrocarbons would be subjected to tandem mass spectrometry. This proposal captured the interest of Mark Dearth, lead scientist on engine exhaust analysis for the Environmental Research Consortium (ERC) involving Ford, Chrysler, General Motors, and Navistar. A cooperative research and development agreement (CRADA) between the ERC and DOE’s Advanced Energy Projects and Technology Division, then part of BES, was formulated, and a research plan was executed to implement a strategy for the rapid analysis of a variety of targeted hydrocarbons at parts-per-billion levels in engine exhaust.
Unlike explosives, it is very difficult to convert hydrocarbons to stable negative ions by electron capture. Furthermore, they are notoriously difficult to ionize by most conventional means without inducing extensive fragmentation. Most hydrocarbons fragment to a fairly common set of products so that extensive fragmentation upon ionization precludes speciation of the parent compounds. The ORNL and Ford scientists, therefore, chose to exploit the ion-molecule reaction chemistry of nitrous oxide (NO+ ). This ion is interesting chemically because it reacts with hydrocarbons but not with major components of engine exhaust, such as carbon dioxide and water. The NO+ ion tends to react either by attaching itself to a hydrocarbon molecule (M) to yield (M + NO+ ) ions, by transferring one of its electrons to a hydrocarbon molecule to yield a molecular ion (M+), or by removing a hydrogen atom from a hydrocarbon molecule (hydride abstraction) to yield (M – H) + ions. The research team studied NO+ ion chemistry with the hydrocarbons in the glow discharge environment and found conditions conducive to the efficient conversion of the targeted molecules of interest to molecular ions or pseudo-molecular ions [i.e., (M + NO+), M+, and (M – H) +].
 |
|
Fig. 5. ( a) Glow discharge mass spectrum obtained from NO+ ionization in which only the (M + NO+) ions of butadiene, methyl butadiene, benzene, toluene, and the xylenes were allowed to accumulate in the ion trap. (b) Spectrum acquired after exciting the parent ions of (a) so that they dissociate to the respective molecular ions. |
|
|
|
Fig. 6. A photograph of the ORNL-developed ion trap instrumentation (white box on the left) that forms the heart of the low-emissions engine exhaust analyzer now being tested in Detroit. |
The Organic Mass Spectrometry Group is now trying to determine the best methods for identifying and quantifying oxygenated hydrocarbons (e.g., alcohols, aldehydes, and ketones) in vehicle emissions. This work is supported by the BES initiative in the Partnership for a New Generation of Vehicles. This analysis problem is similar to the hydrocarbon problem in that it also requires selective ionization combined with rapid and highly specific analysis. Doug"Goeringer and Greg Hurst are examining single-photon photoionization combined with ion trap tandem mass spectrometry as a way to address this problem.
Electrospray, Ion Traps, and DNA Analysis
With the discovery in the past decade of several new ionization methods capable of yielding gas-phase ions derived from large biopolymers, including proteins and deoxyribonucleic acid (DNA), progress in biological mass spectrometry has been nothing short of revolutionary. Mass spectrometry has become, in just a few short years, an important tool for peptide sequencing, protein identification, identification and location of post-translational modifications of proteins, analysis of modified DNA and ribonucleic acids (RNAs), and for many other biological applications. These and other applications have become possible only through the capability to form ions from large biomolecules and through the research community’s growing understanding of the chemistry of the bio-ions.
|
|
|
Fig. 7. A photograph showing light scattered from the fine mist of an electrospray formed at the end of a hypodermic needle. |
In the latter part of the 1980s, our Organic Mass Spectrometry Group began to study electrospray, one of the important new ionization methods, as a way to form gaseous ions from polar species in solution. When a solution is passed through a hypodermic needle held at a potential of several thousand volts, the solution exiting the needle can form a fine mist of charged droplets from which ions emerge. Figure 7 is a photograph of an electrospray at the end of a hypodermic needle which shows light being scattered by the fine mist. A very attractive aspect of this process is that the solute species need not be heated to form gaseous ions. As such, this method has proved to be very effective for the highly involatile and polar molecules that cannot be ionized using conventional gas-phase ionization techniques. Gary Van Berkel of our group was primarily interested in fundamental aspects of the chemistry taking place in the condensed phase while Scott McLuckey was interested in studying the chemistry of the multiply charged ions so frequently produced by electrospray. Together with Gary Glish, they were the first group to couple electrospray with the ion trap (see Fig. 1). Since then, several other group members, including Doug Goeringer, Rose Ramsey, Keiji Asano and Jim Stephenson, have also contributed heavily to the electrospray–ion trap effort.
The group’s expertise with the quadrupole ion trap placed it in an excellent position to exploit the ion trap as a tool for studying the chemistry of polyatomic multiply charged ions. For example, the slow heating nature of the ion trap’s collisional activation process was particularly useful in the group’s heavily cited work on the unimolecular decomposition of small pieces of DNA. This work constituted the first studies of multiply charged anions derived from DNA and led to a detailed mechanistic understanding of nucleobase loss from DNA and subsequent cleavage of the DNA backbone at the base-loss site. This line of work has subsequently been expanded upon by the group at ORNL and others in the mass spectrometry community to provide a way to sequence small oligomeric nucleic acids (both DNA and RNA) via tandem mass spectrometry. This capability is particularly valuable in sequencing short pieces of DNA and RNA and in locating and identifying modifications to DNA and RNA.
 |
|
Fig. 8. ( a) Electrospray mass spectrum of horse heart cytochrome c acquired using the quadrupole ion trap with dimethylamine present in the analyzer at a pressure of 1.2 ´ 10–6 torr after the ions were stored for roughly 20 ms. (b) Electrospray mass spectrum resulting from essentially identical conditions as (a), except that an ion storage period of 1.06 s was used prior to mass analysis. |
Because of its ability to store ions for several seconds, the quadrupole ion trap is particularly well suited to the study of ion-molecule reaction chemistry. For example, Fig. 8 shows data derived from the first study of the ion-molecule reaction chemistry of a multiply charged biopolymer. Figure 8(a) shows the electrospray mass spectrum of horse heart cytochrome c acquired using the quadrupole ion trap with dimethylamine present in the analyzer at a pressure of 1.2 ´ 10–6 torr after the ions were stored for roughly 20 milliseconds (ms). Figure 8(b) shows the mass spectrum resulting from essentially identical conditions except that an ion storage period of 1.06 s was used prior to mass analysis. Comparison of the two spectra shows that the highly charged bio-ions transfer protons to dimethylamine in the gas phase. It has been shown that the kinetics associated with these reactions are sensitive to differences in the high-order (i.e., three-dimensional) structures of proteins. This protein-base study opened up a new and now very active area in gas-phase ion chemistry research which has been expanded to include hydrogen-deuterium exchange chemistry, protein-acid chemistry, DNA-acid chemistry, DNA-nucleophilic substitution chemistry, among others.
All of the ion/molecule reactions alluded to above are sensitive to the three-dimensional structures of biologically derived ions. It is, of course, well known that the shapes of biomolecules have much to do with their functions. It is unclear at this time the extent to which the structures of biological ions in the gas phase relate to their condensed-phase structures. However, the fact that there are chemical probes of gas-phase ion structure opens up the possibility that mass spectrometry may be capable of providing important information regarding issues related to three-dimensional structure. This type of information would add to that already available from mass spectrometry—molecular weight and primary structure (sequence) information.
The most recent research undertaken by our Organic Mass Spectrometry Group involves both electrospray and glow discharge ionization. The ion trap has the unique capability to store oppositely charged ions simultaneously in overlapping regions of space. This ability enables the study of the ion-ion chemistry of multiply charged positive ions derived from electrospray with singly charged negative ions derived by glow discharge. In addition to their fundamental usefulness in the study of ionic interactions in the dilute gas phase, ion-ion reactions have already been shown to be useful in bioanalysis applications. For example, it is now being developed for use in analyzing complex protein mixtures, with the goal of detecting and identifying pathogens. Of particular interest are the signature proteins of bacteria, virus capsid proteins, and protein toxins.
As a result of these initial fundamental studies supported by the BES program, the National Institutes of Health (NIH) became interested in the evolution of quadrupole ion trap instrumentation in biomedical applications. The application of the electrospray–ion trap combination to bio-ion analysis has, therefore, been supported by the NIH over the past five years. Also, DOE’s Office of Nonproliferation and National Security (NN) recently initiated a project to explore the usefulness of electrospray–ion trap mass spectrometry for biological pathogen detection in its Chem/Bio Nonproliferation Program. Perhaps the greatest impact of the ORNL electrospray–ion trap work has been the recent commercial introduction by the firms Bruker and Finnigan of electrospray–ion trap combinations. Both instrument designs were influenced by the early Oak Ridge work, and much of the interest in this combination in the mass spectrometry community.
Conclusion
Mass spectrometry began its many contributions to science and society 100 years ago with the discovery of the electron and hardly remained a static science since then. From the measurement of the charge-to-mass ratio of the electron to the identification of the isotopes and the measurement of their abundances to the molecular weight determination of proteins, mass spectrometry continues to evolve. Research at ORNL with glow discharge ionization, electrospray, and quadrupole ion trap mass spectrometry is a reflection of the diversity of this field and the wide-ranging impact that mass spectrometry research can make. Time and again, the fundamental understanding of the often complex chemistry and physics associated with a mass spectrometry experiment has opened up the possibility for significantly improved analytical measurements. At the same time, the needs associated with a particular goal have often clarified areas where basic research is needed.
|
|
|
