Article pubs.acs.org/ac
Nondestructive Tandem Mass Spectrometry Using a Linear Quadrupole Ion Trap Coupled to a Linear Electrostatic Ion Trap Ryan T. Hilger, Robert E. Santini, and Scott A. McLuckey* Department of Chemistry, Purdue University, West Lafayette, Indiana 47907-2084, United States ABSTRACT: A novel hybrid tandem mass spectrometer is presented that combines a linear quadrupole ion trap (QLIT) with a linear electrostatic ion trap (ELIT), which is composed of opposing ion mirrors. The QLIT is used both as an accumulation device for the pulsed injection of ions into the ELIT and as a collision cell for ions released from the ELIT and back into the QLIT. Ions are subjected to mass analysis in the ELIT via Fourier transformation of the time-domain signal obtained from an image current measurement using a pick-up electrode in the field-free region of the ELIT. The nondestructive nature of ion detection and the relatively straightforward axial entrance and exit of ions into and from the ELIT allow for the execution of nondestructive tandem mass spectrometry experiments whereby both the initial mass spectrum and the product ion spectrum are obtained on the same initial ion population. The timed pulsing of a deflection electrode, in conjunction with the release of ions from the ELIT, allows for the selection of precursor ions for recapture by the QLIT. The transfer of ions back and forth between the QLIT and ELIT is illustrated with Cs ions, the selection of precursor ions is demonstrated with isotopes of tetraoctylammonium cations, and complete nondestructive tandem mass spectrometry experiments are demonstrated with a mixture of angiotensin II and bradykinin cations. With the current apparatus, the efficiency for the process of recapturing ions and then reinjecting them into the ELIT is 35%−40%. The instrument is capable of isolating an ion from a neighbor with a mass as close as 1 part in 500, with negligible loss of the desired species. spectrometry (MSn). In MSn experiments, the same sequence associated with tandem mass spectrometry (precursor ion analysis, isolation, processing, and product ion analysis) is repeated as many times as desired, with a product ion from the previous dimension becoming the precursor ion for the next dimension. The sequence may be repeated until the ion population is diminished due to charge partitioning among several product ions during each dimension and/or inefficiencies (ion losses) in the numerous ion manipulations.11 In most MSn experiments, (n − 1) ionization and destructive detection steps are required to determine the sequence of ions selected over the course of the first (n − 1) stages of ion isolation. Nondestructive ion detection is necessary to execute an MSn experiment following a single ionization event with both mass analysis and isolation at each stage. For an MS/MS experiment, nondestructive ion detection allows for both the initial mass spectrum and the MS/MS spectrum to be acquired from the same initial ion population. Such a capability could be important for analytes that are very expensive or timeconsuming to produce, such as the products of slow and/or inefficient gas-phase reactions. It could also be useful when the precursor ions can be produced only transiently. In such a
T
andem mass spectrometry1 (also MS/MS or MS2) is sequential mass spectrometric analysis. For many studies involving MS/MS, an initial MS experiment is executed to identify potential precursor ions for MS/MS. The initial MS experiment is almost always performed in a destructive sense in that a separate subsequent ionization event is required for MS/ MS. In the MS/MS experiment, the first “MS” step involves the isolation of a single precursor ion, which is then subjected to some physiochemical process that results in the formation of new (product) ions. These product ions are then analyzed in the second stage of mass spectrometry. The identities of the product ions are often informative, with regard to the identity and structure of the precursor ion. Because of this, tandem mass spectrometry is widely used for the analysis of important classes of molecules, such as drugs,2,3 proteins,4,5 carbohydrates,6,7 and lipids.8,9 Tandem mass spectrometry is typically performed one of two ways,10 depending on the type of instrument: tandem-in-time (e.g., quadrupole ion traps, linear ion traps, and magnetic/ electrostatic ion traps) or tandem-in-space (e.g., tandem timeof-flight, triple quadrupole, quadrupole/time-of-flight, and quadrupole/Orbitrap). One convenient aspect of the tandemin-time approach, in contrast with the tandem-in-space approach, is that it is straightforward to extend the dimensionality beyond two, because no hardware modifications are required. This is referred to as multidimensional mass © 2013 American Chemical Society
Received: March 8, 2013 Accepted: April 17, 2013 Published: April 17, 2013 5226
dx.doi.org/10.1021/ac4007182 | Anal. Chem. 2013, 85, 5226−5232
Analytical Chemistry
Article
remain trapped. The same applies to transverse kinetic energy (energy orthogonal to the trap axis). Fragment ions that gain significant transverse kinetic energy may cease to be stable. We have attempted to subvert these difficulties by performing the dissociation externally and reinjecting the product ions. This ensures that the kinetic energy (both axial and transverse) is determined by the injection optics and not influenced by the fragmentation process. Multireflection ToF devices, which are principally identical to our linear electrostatic trap, have been previously developed for the purpose of ion isolation prior to analysis via high-resolution mass spectrometry,27−30 although simultaneous image charge detection was not utilized. Schemes in which ions are recaptured in a QLIT after isolation using a multireflection ToF device have been described in the patent literature. Makarov presented a scheme in which ions are ejected from a QLIT, a single species is isolated using multireflection ToF, fragmented, and then the fragments are returned to the QLIT via a separate entrance such that the overall ion path is cyclical.31 Verentchikov and co-workers described an instrument that involves recapturing ions in a QLIT after isolation by a multireflecting ToF analyzer having a folded ion path.32 In this work, we demonstrate a nondestructive MS/MS experiment that couples a relatively compact linear electrostatic ion trap for mass analysis and mass selection with a linear quadrupole ion trap for collision-induced dissociation. The process is enabled by the nondestructive ion detection approach employed in the electrostatic ion trap and is facilitated by the relative ease with which ions can be ejected from and readmitted into linear ion traps (i.e., both linear electrostatic ion traps and linear quadrupole ion traps).
situation, the analyst might not have enough time to perform multiple analyses before the precursor ions become unavailable. Nondestructive analysis could also simplify monitoring of gasphase reactions. The reaction could be monitored as a function of time without having to reinitiate the reaction after each data point is collected. The Marshall group, for example, used Fourier transform ion cyclotron resonance (FTICR) instruments to perform sequential stages of MS on ions generated from the same initial population of precursors.12 They demonstrated the ability to provide the mass spectrum to the user after each stage, allowing the user to choose the precursor for the next stage.13 Nondestructive detection has also been applied to quadrupole ion traps.14−17 Soni and co-workers, for example, used such an instrument to perform multiple stages of MS without refilling the trap.16 The examples of nondestructive detection in MS/MS and MSn have been restricted to magnetic/electrostatic (e.g., FTICR) and electrodynamic ions traps. Interest in the mass spectrometry field has recently been directed to electrostatic ion traps, which do not require the use of either high magnetic fields or radio-frequency voltages. The highly prominent example of an electrostatic ion trap is the Orbitrap mass analyzer,18 which, similar to FTICR instruments, also uses a nondestructive ion detection approach and Fourier transformation of the time-domain signal. MSn experiments using an Orbitrap as anything other than the analyzer for the final MS step are not straightforward to perform; however, because it is difficult to induce chemistry in the trap without losing ions and it is not straightforward to eject ions efficiently into an external device. An alternative form of electrostatic ion trap that allows for nondestructive detection and relatively straightforward ion ejection and readmission is the electrostatic linear ion trap of the type pioneered by Benner19 and Zajfman,20−23 which is comprised of opposing ion mirrors and a central field-free region. In this device, all ions are accelerated to a known energy before being trapped between the two reflectrons. Each time they pass through the field-free region, they produce an image charge signal on a centrally placed detection electrode. Fourier transformation is then applied to the time domain data and the m/z values of the ions can then be determined from the frequencies.22,23 Advantages of this device include its physical compactness relative to the long flight tubes used in time-offlight (ToF), its tolerance for temporally broad ion pulses, its compatibility with nondestructive detection, and relatively simple axial ion admission and ejection. Several studies have recently been undertaken that involve dissociation of single macromolecular ions in an electrostatic linear ion trap. In these experiments, the ions are heated by exposure to an infrared laser beam and both the charge24−26 and frequency spectrum26 of the trapped ions are monitored as functions of trapping time. While interesting information about the dissociation of the macromolecules is obtained, these studies highlight the difficulties involved with conducting MS/ MS experiments directly in an electrostatic linear ion trap. In this device, the frequency of ion motion is determined by both the m/z and kinetic energy of the trapped ions. If either of these parameters is fixed, the frequency can be used to determine the other. However, a typical dissociation event affects both of these parameters, so that frequency information alone is insufficient to determine the m/z of the products. Also, the mirrors within an electrostatic linear ion trap have a finite kinetic energy acceptance, so fragment ions whose kinetic energy differs substantially from that of the precursor will not
■
EXPERIMENTAL SECTION Materials. Cesium bromide, tetraoctylammonium bromide, bradykinin, and angiotensin II (human) were purchased from Sigma−Aldrich (St. Louis, MO). Methanol and glacial acetic acid were purchased from Mallinckrodt (Phillipsburg, NJ). Water was purified with a water purifier (D8961, Barnstead, Dubuque, IA) prior to use. The solvent for all samples was 49.5% H2O, 49.5% methanol, 1% acetic acid. Mass Spectrometry. A schematic diagram depicting the ion path of the instrument is shown in Figure 1. Sample solution is loaded into a pulled glass capillary, placed 3−10 mm in front of the sampling orifice (100 μm diameter). An electrospray is created by applying high voltage to a platinum wire in contact with the solution.33 Ions and charged drops enter the interface region, which is evacuated to 600 mTorr by a rotary vane pump. Here, desolvation occurs before the ions pass through a 250-μm orifice into the high-vacuum (1 × 10−4 Torr) region, which is pumped by a 400 L/s turbomolecular pump. The ions are transported through an RF-only quadrupole ion guide and a three-element Einzel lens to a quadrupole bender that turns the ions into the linear quadrupole ion trap, abbreviated here as QLIT to distinguish it from the electrostatic ion trap, which is referred to as ELIT. The QLIT consists of four stainless steel rods (radius = 4.69 mm, length = 127 mm, field radius = 4.17 mm,), two end-caps (aperture diameter = 2.4 mm, spaced 2.5 mm from the ends of the rods), and four auxiliary electrodes inserted between the rods that are used to create an axial electric field. These will be referred to as the linac (linear acceleration) electrodes. The linac electrodes34 have a T-shaped cross section, and the length of the stem of the “T” varies along the axis of the trap. We have 5227
dx.doi.org/10.1021/ac4007182 | Anal. Chem. 2013, 85, 5226−5232
Analytical Chemistry
Article
ELIT. This chamber is pumped to 1 × 10−8 Torr by a 700 L/s turbomolecular pump. The design of the ELIT analyzer was inspired by those of Benner19 and Dahan and co-workers.21 It consists of a field-free region (1.300 in inner diameter, 2.000 in. long) with an ion mirror on each side. Each mirror is composed of seven parallel plates (2.000 in. by 2.000 in. by 0.025 in. thick, Kimball Physics, Wilton, NH) with holes 0.225 in. in diameter drilled through the center. The innermost plate is attached directly to the cylinder that encloses the field-free region. The remaining six plates are spaced apart by placing ceramic balls (diameter = 0.250 in., Reid Supply, Muskegon, MI) in the 0.125 in. holes around the edges of each plate. This produces a plate-to-plate spacing of 0.217 in. The field-free region is held at ground potential. DC power supplies (Models P02HA30 and N02HA30, Acopian Technical, Easton, PA; and Model Ortec 556, Advanced Measurement Technology, Oak Ridge, TN) apply potentials to the remaining plates. The same potentials are used on both mirrors, so we will define them for the front mirror only. Numbering the plates 1 through 6 (the seventh plate being attached to the field-free region), the typical potential profile was 2376, 1700, 900, 600, −1526, and −2060 V. The first two plates of the front mirror are connected through switches (Behlke). These switches are used to ground these two plates so that ions may pass into and out of the ELIT. The image charge signal is created as the ions pass through a brass tube (0.325 in. i.d., 0.375 in. o.d., 0.787 in. long) located in the center of the field-free region. The tube is connected to the input of a charge-sensitive preamplifer (CoolFET, Amptek, Bedford, MA). The output of the preamplifier is fed into a Gaussian shaping amplifier (CR-200−500 ns-INST, Cremat, Watertown, MA) that has been adjusted to provide a pulse height of 1 V when the preamplifier detects 16 000 charges. The output of the shaping amplifier is digitized at a rate of 10 MS/s by a PCI-based digitizer (CS1621, Gage Applied Technologies, Lachine, Quebec, Canada) within a personal computer. The overall noise on the detection system corresponds to 250−400 charges on the brass tube (root mean square, rms). The time domain signal is zero-filled to the next power of two and Fourier-transformed with a rectangular window, using a program written in LabVIEW 11.0 (National Instruments, Austin, TX). In the event that the ions are released from the ELIT before the signal has had time to decay into the noise, an exponential window is applied to the time domain data prior to Fourier transformation. The time constant of this window is adjusted to reduce spectral leakage without unnecessarily degrading the resolution. After acquisition of the time domain signal in the ELIT, any remaining ions can be released back toward the QLIT by grounding the first two plates of the front mirror. The ions then pass back through the deflector, where a properly timed voltage pulse applied to the steering plates can be used to isolate a species of interest. After these ions re-enter the QLIT, a switch rapidly increases the potential on the exit lens, trapping the ions inside, where they are thermalized by collisions with the helium bath gas. At this point, the axial field from the linac electrodes is removed by ramping up their potential so that it is equal to the DC offset of the rods. The isolated precursor ions can then be subjected to collision-induced dissociation. This is done by triggering a function generator to produce a sinusoidal signal of proper amplitude and frequency, which is applied across one pair of rods (dipolar excitation) using a transformer. The potential applied to the linac electrodes is then ramped back
Figure 1. Schematic diagram depicting the ion path of the mass spectrometer (not to scale). Legend: A, electrospray emitter; B, interface region; C, RF-only ion guide; D, Einzel lens; E, DC quadrupole bender; F, linear quadrupole ion trap (QLIT); G, electrostatic deflector; and H, electrostatic linear ion trap (ELIT) comprised of opposing ion mirrors.
oriented the trap such that the stem is shortest at the entrance (source) end and longest at the exit (ELIT) end. All four linac electrodes are connected to the same DC potential; a negative potential (relative to the DC offset of the rods) is used to attract the positive ions toward the exit end of the QLIT. The rods are driven by a home-built RF power supply at 1.09 MHz with amplitude adjusted to provide optimal trapping of each analyte. Helium gas is used to raise the pressure inside the QLIT to ∼1−10 mTorr. When the QLIT has been filled with the desired number of precursor ions, the incoming ions are blocked using an upstream lens. At this point, the DC offset of the linear ion trap along with the potentials on the entrance lens, exit lens, and linac electrodes are ramped up 2 kV over the course of 80 ms using programmable, high-voltage power supplies (2HVA24-BP1, UltraVolt, Ronkonkoma, NY). Five milliseconds (5 ms) before the ions are released from the linear ion trap, the potential on the exit lens is ramped down to a value 1−2 V above that which would allow the trapped ions to spill out. This allows the ions to bunch as tightly as possible near the exit lens, which yields the tightest ion pulses for injection into the ELIT. At this point, a fast, high-voltage switch (GHTS 30, Behlke Power Electronics, Billerica, MA) releases the ion pulse toward the ELIT. On the way, the ions pass through an electrostatic deflector with a pair of steering plates (1 cm axial length, 1 cm apart) that are used to align the ion pulse with the ELIT. A DC bias is also applied to the steering plates so that the deflector assembly acts as an Einzel lens, helping to focus the ions into the ELIT. One of the steering plates is connected to a switch (Behlke) that is used to select a portion of the ions for ELIT analysis. The selected ions travel through a 1 mm orifice into the chamber that contains the 5228
dx.doi.org/10.1021/ac4007182 | Anal. Chem. 2013, 85, 5226−5232
Analytical Chemistry
Article
Figure 2. Analysis, recapture, and reanalysis of Cs cations: (a) time domain data, (b) mass spectrum generated from the first (5 ms) analysis, and (c) mass spectrum generated from the second (20 ms) analysis. The abundance values in the two spectra are not directly comparable, because of differences in acquisition time and signal processing. The overall efficiency of the recapture and reinjection process is 37%.
trap at t = 5 ms, the amplitude of the signal was 1.31 V. When the ions re-entered the ELIT at t = 20 ms, the amplitude was 0.48 V, which corresponds to an efficiency of 37%. Figure 2 also shows mass spectra corresponding to the initial injection (Figure 2b) and the second injection (Figure 2c). The peak is narrower in Figure 2c, as a result of the longer acquisition time. The relative intensity of the (Cs + H2O)+ peak at 151 Th also appears diminished somewhat in Figure 2c. The reduced relative abundance of the hydrated Cs ion in Figure 2c may be due exclusively to desolvation of the ion during the course of the recapture process. However, instrumental discrimination effects may also play a role. For example, while the ions are in the ELIT, ions of different m/z are dispersed in space. Since ejecting the ions from the ELIT involves grounding the first two plates of the front mirror, if ions of a specific m/z happen to be in the vicinity of these plates when their voltage abruptly changes, the trajectory of those ions may be perturbed, leading to ion loss. Second, recapture efficiency is maximized by properly choosing the time at which the voltage on the exit lens of the QLIT is raised to
down and the product ions are injected back into the ELIT for analysis following the procedure outlined above.
■
RESULTS AND DISCUSSION The ability to recapture ions after FT-ELIT analysis is demonstrated in Figure 2. Here, Cs ions were injected into the ELIT, where they were trapped for 5 ms. During this time, the ions traveled ∼220 m, as calculated from their frequency (373.16 kHz) and the path length across the ELIT (118 mm). The ions then were transferred back into the QLIT, where they were stored for 15 ms before being reinjected into the ELIT, where they were analyzed until the signal decayed into the noise (∼20 ms; see Figure 2a). Cesium was chosen because there is only one naturally occurring isotope, so the amplitude of the image charge signal is not affected by the separation of isotopes while the ions are in the ELIT. This allows for the use of the amplitude of the image charge signal to gauge the number of ions in the device as the analysis progresses. With this knowledge, we can measure the efficiency of the recapture process. When the ions were ejected back toward the linear ion 5229
dx.doi.org/10.1021/ac4007182 | Anal. Chem. 2013, 85, 5226−5232
Analytical Chemistry
Article
Figure 3. Isolation and reanalysis of individual isotopes of tetraoctylammonium cations: (a) time domain data for the tetraoctylammonium analysis, followed by isolation and reanalysis of the 12C isotope; (b) mass spectrum generated from the reanalysis of both isotopes; (c) mass spectrum generated from the reanalysis of the 12C isotope after it was isolated and recaptured; and (d) mass spectrum generated from the reanalysis of the 13C isotope after it was isolated and recaptured.
Tandem mass spectrometry of recaptured ions is illustrated in Figure 4. A mixture of two peptidesangiotensin II (MW = 1046 Da) and bradykinin (MW = 1060 Da)was analyzed in the ELIT for 1 ms to separate the different species from each other. (Angiotensin II + 2H)2+ was then isolated using the steering plates and recaptured in the QLIT and held at a qvalue of 0.24. Collision-induced dissociation was then performed on the isolated precursor ions by applying dipolar resonant excitation for 100 ms. The product ions were then injected into the ELIT for analysis. The time domain data are shown in Figure 4a. Figure 4b shows a mass spectrum generated from the data acquired during the initial ELIT analysis (precursor ions). The spectrum shows all of the ions produced by the source: singly, doubly, and triply charged peptides. The peaks are broad as a result of the extremely short analysis time. Figure 4c shows a mass spectrum generated from the second analysis (product ions). Many sequence-informative fragment ions are seen. Importantly, these fragment ions were generated from the very same ions that produced the precursor ion peak in Figure 4b. Figure 4d shows a mass spectrum generated in a manner similar to that shown in Figure 4c, but without application of the dipolar excitation signal while the recaptured precursor ions were stored in the QLIT. This corresponds to simply reinjecting the isolated precursor ions and serves to demonstrate the quality of the isolation. The resolution is much better than Figure 4b, because of the much longer analysis time. Figures 4e and 4f are similar to Figures 4c and 4d, respectively, but the other peptide, (bradykinin +2H)2+, was isolated and fragmented. Although water loss dominates the product ion spectrum, many sequence-informative fragment ions are also observed.
trap the recaptured ions inside. Without proper care, it is possible to increase the voltage before all of the ions have entered the QLIT, thus producing unwanted discrimination. The latter potential source of discrimination could be eliminated using improved ion optics that would allow us to use a higher voltage on the exit lens during ion admission. This would form a potential well inside the linear ion trap, where the ions could be captured without needing to increase the voltage on the exit lens. Presently, attempts to use a higher voltage on the exit lens reduce recapture efficiency, presumably because of defocusing. In Figure 3 we demonstrate the intentional discrimination against ions using the electrostatic deflector located between the QLIT and the ELIT. In this experiment, we used tetraoctylammonium cations, because they are comprised of prominent 12C and 13C isotopes. The ions were injected into the ELIT and stored for 4.4 ms to disperse them in space. The time required to achieve maximum dispersion is the period of the beat frequency between the two populations to be separated. The ions then were ejected toward the QLIT, but a pulse was applied to one of the steering plates to divert either the 12C or 13C isotope. The remaining ions were then stored in the QLIT for 15.6 ms before being reinjected into the ELIT for analysis (see Figure 3a). Note that the beating is eliminated in the second analysis, because only one isotope is present. Figure 3b shows a mass spectrum in which no pulse was applied to the steering plates during recapture. It shows both isotopes and is identical to mass spectra of the ions as they are delivered from the source (data not shown). Figure 3c shows a mass spectrum of the recaptured ions when the pulse applied to the steering plates was timed to remove the 13C isotope. Figure 3d is similar, but the pulse was timed to eliminate the 12C isotope. It is evident from a comparison of the peak heights that this isolation can be performed without affecting the abundance of the desired isotope.
■
CONCLUSIONS The objective of this effort is to develop nondestructive multidimensional mass spectrometry (MSn) capabilities using a linear electrostatic (ELIT) ion trap for all stages of mass 5230
dx.doi.org/10.1021/ac4007182 | Anal. Chem. 2013, 85, 5226−5232
Analytical Chemistry
Article
Figure 4. Tandem mass spectrometry of recaptured ions using both components of a mixture of bradykinin and angiotensin II: (a) time domain data for the analysis of the mixture, followed by isolation and recapture of (angiotensin +2H)2+, which was then collisionally activated for 100 ms and reanalyzed; (b) mass spectrum generated from the initial analysis of the mixture (abundance values cannot be compared to those in other panels, because of differences in acquisition time and signal processing); (c) product ion spectrum generated after isolation, recapture, and collisional activation of (angiotensin +2H)2+; (d) reanalysis of isolated and recaptured (angiotensin +2H)2+ demonstrating the quality of the isolation; (e) product ion spectrum generated after isolation, recapture, and collisional activation of (bradykinin +2H)2+; and (f) reanalysis of isolated and recaptured (bradykinin +2H)2+, demonstrating the quality of the isolation.
■
spectrometry (MS). To this end, we have coupled a linear quadrupole ion trap (QLIT) to a linear electrostatic ion trap. The former is used to accumulate ions from an external ion source (e.g., an electrospray ion source) for subsequent pulsed injection into the ELIT and for recapturing ions injected back from the ELIT after mass analysis for subsequent collisional activation. The ELIT is used for a first stage of mass analysis, for dispersion of precursor ions to allow for timed ejection of unwanted ions upon release of ions back into the QLIT, and for the final stage of mass analysis of the product ions. The transfer of ions between the QLIT and ELIT with intervening MS stages is demonstrated with Cs ions. Nondestructive tandem mass spectrometry (MS/MS) experiments using ion trap collisional activation is demonstrated with a mixture of angiotensin II and bradykinin ions. These proof-of-principle experiments exhibit transfer efficiencies of a few tens of percent and storage times in the ELIT that are limited by the relatively high pressures in the system. Higher overall efficiencies and higher resolution in the ELIT are anticipated with improved ion transfer optics between the QLIT and the ELIT and lower pressures in the ELIT.
AUTHOR INFORMATION
Corresponding Author
*Address: Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, IN 47907-2084, USA. Tel.: (765) 494-5270. Fax: (765) 494-0239. E-mail: mcluckey@purdue. edu. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by AB Sciex. We thank Randy Replogle, Rob Oglesbee, Phil Wyss, Cathy McIntyre, Bob Fagan, and Hartmut Hedderich (Jonathan Amy Facility for Chemical Instrumentation) for their help with construction of the mass spectrometer. We also acknowledge Mircea Guna and Dr. James W. Hager (AB Sciex) for helpful discussions and for providing the QLIT.
■
REFERENCES
(1) Gross, J. H. Mass Spectrometry: A Textbook, 2nd Edition; Springer: New York, 2011; pp 415−478. (2) Niessen, W. M. A. Mass Spectrom. Rev. 2012, 31, 626−665. 5231
dx.doi.org/10.1021/ac4007182 | Anal. Chem. 2013, 85, 5226−5232
Analytical Chemistry
Article
(3) Niessen, W. M. A. Mass Spectrom. Rev. 2011, 30, 626−663. (4) Hernandez, P.; Müller, M.; Appel, R. D. Mass Spectrom. Rev. 2006, 25, 235−254. (5) Meng, F.; Forbes, A. J.; Miller, L. M.; Kelleher, N. L. Mass Spectrom. Rev. 2005, 24, 126−134. (6) Zhou, W.; Hakansson, K. Curr. Proteomics 2011, 8, 297−308. (7) An, H. J.; Lebrilla, C. B. Mass Spectrom. Rev. 2011, 30, 560−578. (8) Cui, Z.; Thomas, M. J. J. Chromatogr. B 2009, 877, 2709−2715. (9) Zehethofer, N.; Pinto, D. M. Anal. Chim. Acta 2008, 627, 62−70. (10) Johnson, J. V.; Yost, R. A. Anal. Chem. 1990, 20, 2162−2172. (11) McLuckey, S. A.; Glish, G. L.; Van Berkel, G. J. Int. J. Mass Spectrom. Ion Processes 1991, 106, 213−235. (12) Solouki, T.; Paša-Tolić, L.; Jackson, G. S.; Guan, S.; Marshall, A. G. Anal. Chem. 1996, 68, 3718−3725. (13) Guan, S.; Marshall, A. G. Anal. Chem. 1997, 68, 1−4. (14) Syka, J. E. P.; and Fies, W. J., Jr. U.S. Patent 4,755,670, 1988. (15) Goeringer, D. E.; Crutcher, R. I.; McLuckey, S. A. Anal. Chem. 1995, 67, 4164−4169. (16) Soni, M.; Frankevich, V.; Nappi, M.; Santini, R. E.; Amy, J. W.; Cooks, R. G. Anal. Chem. 1996, 68, 3314−3320. (17) Nappi, M.; Frankevich, V.; Soni, M.; Cooks, R. G. Int. J. Mass Spectrom. 1998, 177, 91−104. (18) Makarov, A.; Denisov, E.; Kholomeev, A.; Baischun, W.; Lange, O.; Strupat, K.; Horning, S. Anal. Chem. 2006, 78, 2113−2120. (19) Benner, W. H. Anal. Chem. 1997, 69, 4162−4168. (20) Zajfman, D.; Heber, O.; Vejby-Christensen, L.; Ben-Itzhak, I.; Rappaport, M.; Fishman, R.; Dahan, M. Phys. Rev. A 1997, 55, R1577− R1580. (21) Dahan, M.; Fishman, R.; Heber, O.; Rappaport, M.; Altstein, N.; Zajfman, D.; van der Zande, W. J. Rev. Sci. Instrum. 1998, 69, 76−83. (22) Ring, S.; Pedersen, H. B.; Heber, O.; Rappaport, M. L.; Witte, P. D.; Bhushan, K. G.; Altstein, N.; Rudich, Y.; Sagi, I.; Zajfman, D. Anal. Chem. 2000, 72, 4041−4046. (23) Zajfman, D.; Rudich, Y.; Sagi, I.; Strasser, D.; Savin, D. W.; Goldberg, S.; Rappaport, M.; Heber, O. Int. J. Mass Spectrom. 2003, 229, 55−60. (24) Doussineau, T.; Bau, C. Y.; Clavier, C.; Dagany, X.; Kerleroux, M.; Antoine, R.; Dugourd, P. Rev. Sci. Instrum. 2011, 82, 084104. (25) Doussineau, T.; Antoine, R.; Santacreu, M.; Dugourd, P. J. Phys. Chem. Lett. 2012, 3, 2141−2145. (26) Antoine, R.; Doussineau, T.; Dugourd, P.; Calvo, F. Phys. Rev. A 2013, 87, 013435. (27) Ishida, Y.; Wada, M.; Matsuo, Y.; Tanihata, Y.; Casares, A.; Wollnik, H. Nucl. Inst. Meth. Phys. Res. B 2004, 219−220, 468−472. (28) Plass, W. R.; Dickel, T.; Czok, U.; Geissel, H.; Petrick, M.; Reinheimer, K.; Scheidenberger, C.; Yavor, M. I. Nucl. Inst. Meth. Phys. Res. B 2008, 266, 4560−4564. (29) Wolf, R. N.; Eritt, M.; Marx, G.; Schweikhard, L. Hyperfine Interact. 2011, 199, 115−122. (30) Wolf, R. N.; Beck, D.; Blaum, K.; Böhm, Ch.; Borgmann, Ch.; Breitenfeldt, M.; Herfurth, F.; Herlert, A.; Kowalska, M.; Kreim, S.; Lunney, D.; Naimi, S.; Neidherr, D.; Rosenbusch, M.; Schweikhard, L.; Stanja, J.; Wienholtz, F.; Zuber, K. Nucl. Inst. Meth. Phys. Res A 2012, 686, 82−90. (31) Makarov, A. A. U.S. Patent 7,829,842, 2010. (32) Verentchikov, A. N.; Yavor, M.; Mitchell, J. C.; Artaev, V. International Patent WO 2005/001878, 2005. (33) Van Berkel, G. J.; Asano, K. G.; Schnier, P. D. J. Am. Soc. Mass Spectrom. 2001, 12, 853−862. (34) Loboda, A.; Krutchinsky, A.; Loboda, O.; McNabb, J.; Spicer, V.; Ens, W.; Standing, K. Eur. J. Mass Spectrom. 2000, 6, 531−536.
5232
dx.doi.org/10.1021/ac4007182 | Anal. Chem. 2013, 85, 5226−5232
