Efficient Ion Remeasurement Using Broadband Quadrupolar

Capacitive coupling of the excitation signal to the trapping plates of the open-ended cylindrical analyzer cell is required for high remeasurement eff...
8 downloads 5 Views 452KB Size
Download PDF
Anal. Chem. 1996, 68, 3732-3739

Efficient Ion Remeasurement Using Broadband Quadrupolar Excitation FTICR Mass Spectrometry Cynthia C. Pitsenberger, Michael L. Easterling, and I. Jonathan Amster*

Department of Chemistry, University of Georgia, Athens, Georgia 30602-2556

Quadrupolar excitation is achieved over a wide mass range by using repetitive chirp excitation, filtered noise excitation, and high-amplitude, single-frequency excitation for matrix-assisted laser desorption FTICR mass spectrometry. These methods efficiently axialize ions over a wide range of masses in a 4.7 T FTICR spectrometer. Remeasurement efficiencies are >99.5% for broadband repetitive chirp excitation, 99% with filtered white noise, and 99.4% with high-amplitude, single-frequency broadband quadrupolar excitation. Results indicate that z-axis ejection of ions during detection is the primary mechanism for ion loss during remeasurement experiments. Capacitive coupling of the excitation signal to the trapping plates of the open-ended cylindrical analyzer cell is required for high remeasurement efficiency. Quadrupolar excitation, introduced by Savard et al.1 and first adapted to analytical Fourier transform ion cyclotron resonance (FTICR) spectrometry by Marshall and co-workers,2 has attracted considerable recent attention for increasing the performance and capabilities of FTICR spectrometry. In contrast to dipolar excitation, which is used to increase the radii of the cyclotron orbits of ions prior to detection, with quadrupolar excitation, cyclotron and magnetron motions interconvert. When ions undergo collisional damping during quadrupolar excitation (QE), they move to the center of the analyzer cell, a process called axialization. QE can be used to refocus ions that are scattered throughout an analyzer cell after the detection process, so that they can be remeasured. Remeasurement using quadrupolar excitation was first reported by Speir et al., who demonstrated that ions of a selected massto-charge ratio could be remeasured more than 500 times, with axialization efficiencies >99.5% per measurement cycle.3 More recently, Hendrickson and Laude demonstrated single-frequency QE remeasurement of electrosprayed proteins, emphasizing the S/N improvements and increased spectral sensitivity.4 Marshall et al. have demonstrated attomole sensitivity for peptide ions by using ion remeasurement.5 Hendrickson et al. have also shown remeasurement using a simplified two-plate quadrupolar excitation.6 Aside from remeasurement, QE has led to improvements (1) Savard, G.; Becker, S.; Bollen, G.; Kluge, H.-J.; Moore, R. B.; Schweikhard, L.; Stolzenberg, H. Wiess, U. Phys. Lett. A 1991, 158, 247-252. (2) Schweikhard, L.; Guan, S. H.; Marshall, A. G. Int. J. Mass Spectrom. Ion Processes 1992, 120, 71-83. (3) Speir, J. P.; Gorman, G. S.; Pitsenberger, C. C.; Turner, C. A.; Wang, P. P.; Amster, I. J. Anal. Chem. 1993, 65, 1746-1752. (4) Hendrickson, C. L.; Laude, D. A. Anal. Chem. 1995, 67, 1717-1721. (5) Solouki, T.; Marto, J. A.; White, F. M.; Guan, S.; Marshall, A. G. Anal. Chem. 1995, 67, 4139-4144. (6) Hendrickson, C. L.; Drader, J. J.; Laude, D. A. J. Am. Soc. Mass Spectrom. 1995, 6, 448-452.

3732 Analytical Chemistry, Vol. 68, No. 21, November 1, 1996

in mass resolution, mass selectivity, and selected ion accumulation of electrosprayed ions.3,7-9 As originally described, QE was applied to ions of a single m/z value. However, many applications can be envisioned that would benefit from axialization of a range of masses. In principle, broadband axialization could be used to refocus the products of collisionally activated dissociation (CAD), so that ions can be reexcited for further stages of CAD (i.e., MSn). Techniques for performing multiple-stage mass spectrometry (MSn) experiments hold great promise for the structural characterization of complex biological compounds. Biopolymer analysis is often accomplished with quadrupole mass spectrometers, time-of-flight mass spectrometers, or sector-based instruments, yet these methods are limited in the number of stages of MSn that can be achieved. The unique features of Fourier transform ion cyclotron resonance mass spectrometry (FTICR MS), ultrahigh resolution, high mass accuracy, simultaneous detection of all ions, and the nondestructive nature of ion detection make it particularly well-suited for multiple stages of tandem mass spectrometry.10-20 In FTICR, ion detection is based on the measurement of an image current that is induced on the receiver plates of the ICR cell by coherently excited ions. Unlike other types of mass spectrometers, ions remain in the FTICR analyzer cell after detection and can be used for further analysis.3,21-26 With this advantage, the potential exists for (7) Guan, S.; Wahl, M. C.; Wood, T. D. Marshall, A. G. Anal. Chem. 1993, 65, 1753-1757. (8) Pasˇa-Tolic´, L.; Huang, Y.; Guan, S.; Kim, H. S.; Marshall, A. G. J. Mass Spectrom. 1995, 30, 825-833. (9) Bruce, J. E.; Van Orden, S. L.; Anderson, G. A.; Hofstadler, S. A.; Sherman, M. G.; Rockwood, A. L.; Smith, R. D. J. Mass Spectrom. 1995, 30, 124133. (10) Buchanan, M. V.; Comisarow, M. B. In Fourier Transform Mass Spectrometry: Evolution, Innovation, and Applications; Buchanan, M. V., Ed.; ACS Symposium Series 359; American Chemical Society: Washington, DC, 1984; pp 1-20. (11) Marshall, A. G.; Grosshans, P. B. Anal. Chem. 1991, 63, 215A-229A. (12) Marshall, A. G.; Schweikhard, L. Int. J. Mass Spectrom. Ion Processes 1992, 118/119, 37-70. (13) Jacoby, C. B.; Holliman, C. L.; Gross, M. L. In Mass Spectrometry in the Biological Sciences: A Tutorial; Gross, M. L., Ed.; Kluwer Academic Publisher: Dordrecht, 1992; pp 93-116. (14) Buchanan, M. V.; Hettich, R. L. Anal. Chem. 1993, 65, 245A-259A. (15) Pastor, S. J.; Castoro, J. A.; Wilkins, C. L. Anal. Chem. 1995, 67, 379-384. (16) Pasˇa-Tolic´ L.; Huang, Y.; Guan, S.; Kim, H. S.; Marshall, A. G. J. Mass Spectrom. 1995, 30, 825-833. (17) McIver, R. T.; Li, Y.; Hunter, R. L. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 4801-4805. (18) Li, Y.; Hunter, R. L.; McIver, R. T. Nature 1994, 370, 393-395. (19) McLafferty, F. W. Acc. Chem. Res. 1994, 27, 379-386. (20) Speir, J. P.; Senko, M. W.; Little, D. P.; Loo, J. A.; McLafferty, F. W. J. Mass Spectrom. 1995, 30, 39-42. (21) Williams, E. R.; Henry, K. D.; McLafferty, F. W. J. Am. Chem. Soc. 1990, 112, 6157-6162. (22) Guan, Z.; Hofstadler, S. A.; Laude, D. A. Anal. Chem. 1993, 65, 15881593. S0003-2700(96)00532-X CCC: $12.00

© 1996 American Chemical Society

sequencing large biopolymers of femtomole sample size by MSn. Examples of quadrupolar excitation over a range of masses have been published.27-34 Broadband axialization has been used to move ions that are introduced off-axis through external ion injection or MALDI (matrix-assisted laser desorption/ionization) to the center of an ICR analyzer cell.35 Guan and Marshall have used SWIFT (stored waveform inverse Fourier transform) to axialize ions by quadrupolar excitation, to transfer them from the source to the analyzer of a dual cell.30,31 They demonstrated that repeated, low-amplitude SWIFT waveforms could be used to convert magnetron motion to cyclotron motion in a linear fashion for a range of masses.30 Later, Wood and co-workers provided a systematic study of parameters that affect SWIFT QE, establishing the optimum conditions necessary for efficient transfer of parent ions for CAD.32 Since then, studies have been reported using SWIFT broadband QE,33,34 to axialize ions over a range of 1000 mass units for transferring ions from the high-pressure side to the low-pressure side of a dual analyzer cell. There have been no reports of broadband QE ion remeasurement. Moreover, the utility of this technique has not been fully developed for MS/MS studies or for ICR systems lacking SWIFT capabilities. Broadband quadrupolar excitation has been under investigation in our laboratory for several years. Previously, we have shown broadband axialization of small peptides using a repetitive chirp excitation with a 1 T FTICR mass spectrometer.28 This type of excitation successfully axialized product ions from CAD of small peptides over a mass range of m/z 50-406. Software and hardware limitations of the data system used for these earlier experiments led to a duty cycle of no more than 75% for repetitive chirp quadrupolar excitation, reducing the efficiency of this experiment. The addition of an external function generator with a 100% duty cycle for repetitive chirp excitation improved the efficiency of this experiment.29 With this function generator, we demonstrated one of the first examples of axialization using filtered white noise quadrupolar excitation.29 Here we describe methods that allow efficient quadrupolar excitation over a much wider mass range than has been previously demonstrated. The efficiency of broadband QE is examined using several types of excitation waveforms, including chirp excitation, (23) Guan, S.; Hyun, S. K.; Marshall, A. G.; Wahl, M. C.; Wood, T. D.; Xiang, X. Chem. Rev. 1994, 94, 2161-2182. (24) Guan, Z.; Drader, J. J.; Campbell, V. L.; Laude, D. A. Anal. Chem. 1995, 67, 1453-1458. (25) Campbell, V. L.; Guan, Z.; Vartanian, V. H.; Laude, D. A. Anal. Chem. 1995, 67, 420-425. (26) Campbell, V. L.; Guan, Z.; Laude, D. A. J. Am. Soc. Mass Spectrom. 1995, 6, 564-570. (27) Guan, S.; Marshall, A. G.; Wahl, M. C. Anal. Chem. 1994, 66, 1363-1367. (28) Wang, P. P.; Pitsenberger, C. C.; Easterling, M.; Amster, I. J. Proceedings of the 42nd ASMS Conference on Mass Spectrometry and Allied Topics, Chicago, IL, May 29-June 3, 1994; p 35. (29) Pitsenberger, C. C.; Amster, I. J. Proceedings of the 43rd ASMS Conference on Mass Spectrometry and Allied Topics, Atlanta, GA, May 21-26, 1995; p 798. (30) Guan, S.; Marshall, A. G. J. Chem. Phys. 1993, 98, 4486-4493. (31) Guan, S. H.; Wahl, M. C.; Marshall, A. G. J. Chem. Phys. 1994, 100, 61376140. (32) Wood, T. D.; Ross, C. W.; Marshall, A. G. J. Am. Soc. Mass Spectrom. 1994, 5, 900-907. (33) Huang, Y.; Pasˇa-Tolic´ L.; Guan, S.; Marshall, A. G. Anal. Chem. 1994, 66, 4385-4389. (34) Marto, J. A.; Guan, S.; Marshall, A. G. Rapid Commun. Mass Spectrom. 1994, 8, 615-620. (35) Guan, S.; Pasˇa-Tolic´ L.; Marshall, A. G.; Xiang, X. Int. J. Mass Spectrom. Ion Processes 1994, 139, 75-86.

frequency burst excitation, and filtered white noise excitation. Broadband axialization is used to efficiently axialize ions over a mass range 2000 m/z units wide. EXPERIMENTAL SECTION A 4.7 T FTICR spectrometer that was designed and fabricated at the University of Georgia was used for all experiments. The instrument consists of a custom-designed vacuum chamber, a shielded 4.7 T superconducting magnet, and a commercial data system (IonSpec, Irvine, CA). The main vacuum chamber is constructed of aluminum and is pumped to a base pressure of 5 × 10-10 Torr with a 330 L/s turbomolecular pump, backed by a 150 L/s diffusion pump. Housed in the main vacuum chamber, in the homogeneous field of the magnet, is an open-ended, capacitively coupled analyzer cell with a diameter of 2.25 in. and an aspect ratio of 1. An ancillary chamber, evacuated by a 1100 L/s cryopump, is used to pump down sample-covered stubs to a base pressure of 1 × 10-7 Torr before introduction to the main chamber. This two-stage sample introduction permits laser desorption analysis under ultrahigh vacuum. Nitrogen (>99.999% purity) for collisionally induced axialization is introduced through a pulsed valve. The pulsed valve is opened for less than 20 ms, and the analyzer pressure rises to approximately 1 × 10-4 Torr. The pressure returns to 5 × 10-9 Torr within 20 s. Two-plate quadrupolar excitation, recently described by Hendrickson et al.,6 was used for all experiments. With two-plate quadrupolar excitation, only one of the electrical connections to the analyzer cell is switched from its normal connection. With four-plate quadrupolar excitation, three electrical leads must be switched, including the two leads to the detection electrodes. By using two-plate quadrupolar excitation, a preamplifier for the image current can be mounted directly on the detection plates, inside the vacuum chamber, which provides a considerable improvement in the signal-to-noise ratio for this instrument. Switching between normal dipolar excitation and QE is performed by an electromechanical relay.3 Data accumulation and processing are controlled by an IonSpec data system. For these experiments, the transient signals were digitized at 100 kHz, and the data set size was 32 kB. The instrument is equipped with an internal preamplifier (IonSpec, Irvine, CA) mounted directly on the analyzer cell plate to reduce the adverse effects of distributed capacitance on the image current. Samples are desorbed using the 355 nm output of a Nd:YAG laser (New Wave Research, Sunnyvale, CA), focused with telescopic optics. The sample target can be rotated to examine different samples without moving the focal point of the laser beam. Samples are prepared using standard MALDI methods. The PEG samples are dissolved at a concentration of 1 nmol/µL in 70:30 water (0.1% TFA)/acetonitrile solution. Two microliters of a saturated solution of sinapinic acid in a 70:30 solution of water (0.1% TFA)/acetonitrile is first placed on the probe, followed by 2 µL of the PEG solution. The sample is allowed to air-dry. Ions are formed by matrix-assisted laser desorption and detected at a trapping potential of 0.5 V. Broadband quadrupolar excitation of ions was accomplished with the set of events depicted in Figure 1. At the beginning of the experimental sequence used for ion remeasurement, the trapping potential is raised to 5 V, and a relay is activated to switch the electronics from normal dipolar excitation to quadrupolar excitation. A buffer gas is introduced at a peak pressure of 1 × 10-4 Torr, and the ions undergo quadrupolar excitation for 3.9 s. The trapping potential is then reduced to 0.5 Analytical Chemistry, Vol. 68, No. 21, November 1, 1996

3733

Figure 1. Typical sequence for broadband repetitive chirp, filtered white noise, and single-frequency, high-amplitude QE remeasurement studies at 4.7 T. Ions are formed by MALDI and trapped in the analyzer cell by a separate experimental sequence. When the pulsed valve is opened, the pressure rises to 1 × 10-4 Torr. The synthesizer provides the rf signal used for quadrupolar excitation. The trapping potential is raised to 5 V during QE and reduced to 0.5 V during the rest of the experiment. Capacitive coupling can be enabled or disabled at any point during the experiment. Quadrupolar excitation by either the data system’s frequency synthesizer or by an external frequency synthesizer can be selected by the software (Ext/Int Excite).

V, and the ions in the analyzer cell are detected approximately 25 s after the start of the sequence. The quadrupolar excitation signal originates from a function generator with a 3 MHz bandwidth (Stanford Research Systems, Model DS 335). The generator provides several types of waveforms, including fixed frequency, linear and logarithmic frequency sweeps, and white noise. The data system has been modified so that the operator can select between QE excitation sources, i.e., the external function generator or the data system’s frequency synthesizer, by activating an additional relay from the software (see Figure 1). For these experiments, the external function generator was used for chirp and white noise quadrupolar excitation. The data system’s frequency synthesizer was used for single-frequency quadrupolar excitation. The selected excitation waveform is then amplified and applied to the excitation plates of the analyzer cell. Linear frequency sweeps of sinusoidal waveforms were used for chirp quadrupolar excitation. These waveforms, originating from the function generator, typically span a frequency range of 28 767-47 924 Hz, and the sweeps are repeated at a rate of 1 kHz. For burst excitation, a single frequency is generated by the data system and is applied to excitation plates. For broadband-filtered white noise excitation, white noise from the external frequency generator is passed through an active bandpass filter, amplified, and applied to the cell excite plates. The bandpass filter is a statevariable filter, shown schematically in Figure 2a, that was constructed on a breadboard. This filter was selected for its ease of tuning and minimal component sensitivities. The values of the three resistors, Rf (2 K), Rq (560 Ω), and Rg (100 K), determine the center frequency, Q value, and band-center gain. The frequency spectrum of a typical signal used for QE is shown in Figure 2b and was obtained by digitizing the excitation signal and computing its Fourier transform. Table 1 outlines the specific conditions used for remeasurement in this report. As will be 3734 Analytical Chemistry, Vol. 68, No. 21, November 1, 1996

Figure 2. (a) Schematic of the state-variable, active bandpass filter used to filter white noise produced by a frequency synthesizer. LF411 operational amplifiers were used, and Rf ) 2 K, Rg ) 100 K, and Rq ) 560 Ω. (b) The frequency spectrum of the white noise after filtering.

illustrated later, we have examined the effects of remeasurement with and without the use of capacitive coupling. Capacitive

Table 1. Experimental Parameters for Quadrupolar Excitation

frequency of applied excitation (Hz) QE duration (ms) QE voltage (V) trapping voltage during QE (V) trapping voltage during detection (V) sweep repetition rate (Hz) detection excitation voltage (V) a

chirp excitation

filtered noise excitation

burst excitation

28 767-47 924 3900 27 5 0.5 1000 40, 70

38 947-31 836 3900 20-30 5 0.5 naa 40, 70

28 767 3900 29 5 0.5 naa 70

Not applicable.

Figure 3. Flow chart illustrating the manner in which the electronics for quadrupolar excitation and capacitive coupling are integrated into the experiment. SPDT relays (shown as X) are used to switch between internal and external rf sources for QE, and to switch between dipolar and quadrupolar excitation, and are activated by TTL pulses originating from the data station.

coupling extends the rf excitation to the ends of the analyzer cell to linearize the excitation field and reduce the axial ejection of ions.36 A schematic of the electronics used to integrate QE and capacitive coupling into the experiment is shown in Figure 3. RESULTS AND DISCUSSION Frequency Chirp Broadband QE. In the presence of QE, ions that experience a resonant, alternating azimuthal quadrupolar field convert magnetron motion to cyclotron motion, the latter being damped more rapidly by the buffer gas. Likewise, in the presence of a broadband QE field, ions with cyclotron frequencies that fall within the range of applied frequencies will also undergo axialization. We have previously reported this type of broadband QE,28 but software limitations required delay times between repetitions of the chirp waveforms, reducing the efficiency of axialization. We have overcome this limitation by using an external signal generator which can produce continuous repetitive frequency sweeps, thereby eliminating the delay between frequency sweeps and the consequential magnetron expansion. By this method, chirp excitation has been used to efficiently axialize ions of poly(ethylene glycol) (PEG) 2000. In the mass spectrum shown in Figure 4a, ions formed by MALDI of PEG 2000 were axialized by a continuous linear frequency sweep from 28 767 to 47 924 Hz, corresponding to a mass-to-charge range of 1500-2500. The frequency range was swept at a rate of 1000 Hz, i.e., 1000 (36) Beu, S. C.; Laude, D. A. Anal. Chem. 1992, 64, 177-180.

sweeps/s or 1 ms/sweep. The ions formed by a single laser desorption event were then remeasured 50 times with signal averaging to produce the mass spectrum of Figure 4b. For this experiment, capacitive coupling was possible for the entire experimental cycle, and a normal dipolar excitation amplitude of 70 V was used for ion detection. The polymer distribution in Figure 4a is nonstatistical due to the small number of ions that were present in the cell. Remeasurement of the ions maintains the same distribution, as can be seen in Figure 4b, demonstrating that the efficiency of remeasurement is constant over the entire mass range. Figure 4c,d shows an expansion of the mass scale around m/z 2500, and demonstrates that remeasurement increases the S/N while maintaining isotopic resolution. An individual scan of ions formed from MALDI of PEG 2000 yields virtually no signal at the high end of the axialization range. There is signal present, but it is buried in the noise. Signal averaging for 50 scans using chirp excitation over m/z 1500-2500 produces an improvement in S/N of approximately a factor of 5. Signals for the polymer M + Na+ and M + K+ ions appear from out of the noise with isotopic resolution. Previously, with a 1 T FTICR spectrometer, we had attained remeasurement efficiencies of ∼98.5% using broadband chirp excitation. However, these prior experiments were limited by the maximum m/z range that could be axialized, typically up to 400 mass units. Also, high trapping voltages (3 V) and reduced detection (dipolar) excitation amplitudes were necessary to achieve high efficiency. We can now achieve efficient remeaAnalytical Chemistry, Vol. 68, No. 21, November 1, 1996

3735

Figure 4. (a) Ions formed by one MALDI event remeasured 50 times using a frequency chirp (28 767-47 924 Hz) and broadband quadrupolar excitation and detected with a dipolar excitation amplitude of 70 V, typical of that used for normal FTICR experiments. (b) The ion signal after 50 remeasurement cycles using frequency chirp broadband QE of ions formed from one laser desorption event of PEG 2000. The data in (a) are scaled to the same intensity as the ion signal in (b). The inset in (a) is a 38.7 times amplification. (c) An expansion of the mass axis of (a). (d) An expansion of the mass range of (b), illustrating the improved S/N and constant isotopic resolution from broad-band QE remeasurement.

surement over a much wider mass range, using normal detection conditions. Figure 5a shows a plot of the summed total ion intensity versus the number of remeasurement cycles and compares these with calculated signal intensities. From this plot, we see that the remeasurement efficiency is 98.5% or better. We observe noise to increase in direct proportion to the square root of the number of remeasurements, as expected. Plots of S/N versus remeasurement number therefore show the same efficiency as do the plots of Figure 5. By reducing the rf dipolar excitation amplitude to 40 V, the remeasurement efficiency can be increased to 100%, as shown by the data for 50 remeasurement cycles in Figure 5b, obtained for PEG 2000 under experimental conditions otherwise identical with those for Figure 4. We have been able to show that axial ejection is the principal mechanism for ion loss in the remeasurement experiment, by disabling capacitive coupling 3736

Analytical Chemistry, Vol. 68, No. 21, November 1, 1996

during the experiment. Under these conditions, the remeasurement efficiency was found to decrease to 99% (Figure 5) for ions over a 400 m/z range. The efficiency of filtered noise broadband QE was found to be approximately the same (99%-99.5%) at both 70 and 40 VDET (VDET is the amplitude of the excitation voltage used for ion detection). When capacitive coupling was disabled for the entire experiment, ions (37) McLuckey, S. A.; Goeringer, D. E.; Glish, G. L. Anal. Chem. 1992, 64, 14551460. (38) Kelley, P. E. U.S. Patent 5,134,286, 1992. (39) Goeringer, D. E.; Keiji, G. A.; McLuckey, S. A.; Hoekman, D.; Stiller, S. W. Anal. Chem. 1994, 66, 313-318. (40) Kenny, D. V.; Callahan, P. J.; Gordon, S. M.; Stiller, S. W. Rapid Commun. Mass Spectrom. 1993, 7, 1086-1089. (41) Asano, K. G.; Goeringer, D. E.; McLuckey, S. A. Anal. Chem. 1995, 67, 2739-2742. (42) Ernst, R. R. J. Magn. Reson. 1970, 3, 10-27. (43) Kaiser, R. J. Magn. Reson. 1970, 3, 28-43. (44) Marshall. A. G. Presented at the 5th East Coast ICR and Ion-Molecule Symposium, University of Delaware, Newark, DE, Sept. 24, 1983; Presented at the 10th Federation of Analytical Chemistry and Applied Spectroscopy Societies Meeting, Philadelphia, PA, Sept. 27, 1983. (45) Ijames, C. F.; Wilkins, C. L. Chem. Phys. Lett. 1984, 108, 58-62. (46) Marshall, A. G.; Wang, T. L.; Ricca, T. L Chem. Phys. Lett. 1984, 108, 6366. (47) Bruce, J. E.; Anderson, G. A.; Chen, R.; Cheng, X.; Gale, D. C.; Hofstadler, S. A.; Schwartz, B. L.; Smith R. D. Proceedings of the 43rd ASMS Conference on Mass Spectrometry and Allied Topics, Atlanta, GA, May 21-26, 1995; p 682. (48) Bruce, J. A.; Anderson, G. A.; Smith, R. D. Anal. Chem. 1996, 68, 534541.

Figure 5. (a) Remeasurement efficiency curves for chirp, filtered white noise, and frequency burst quadrupolar excitation, with dipolar detection voltages of 70 VDET. The solid lines represent theoretical curves, calculated for efficiencies of 60%-100%. (b) Remeasurement efficiency curves for repetitive chirp and filtered noise broad-band QE at 40 VDET.

could not be remeasured efficiently, presumably due to axial ejection. The smaller range of ions axialized by filtered noise QE is due to the excitation waveform whose frequency spectrum is shown in Figure 3b. As can be seen, the power peaks sharply at m/z 2000. Modifying the active bandpass filter, so that the power is flat over the range of interest, should increase the bandwidth of axialization. Burst Excitation Broadband QE. Isolation of ions of a single m/z can be achieved with a low-amplitude (99% when optimized. Filtered white noise excitation is easier to implement than chirp excitation, as only one variable, the amplitude, needs to be tuned. Variables in chirp excitation include not only the amplitude but also the sweep rate. The results show that capacitive coupling of the excitation signal to the trapping plates of the open-ended cylindrical analyzer cell reduces z-axis ejection and leads to high remeasurement efficiency. Experiments are in progress to demonstrate the utility of these methods for performing multiple stages of tandem mass spectrometry.

Figure 9. Axialization of high-mass ions from (a) PEG 4600, (b) PEG 6000, and (c) PEG 8000. The mass spectra in (a) and (c) were collected by using frequency burst quadrupolar excitation. The mass spectrum of (b) was obtained by using frequency chirp quadrupolar excitation.

ACKNOWLEDGMENT We gratefully acknowledge the financial support of the National Science Foundation, CHE-9412334. Received for review May 30, 1996. Accepted August 15, 1996.X AC960532R

(50) Dey, M.; Castoro, J. A.; Wilkins, C. L. Anal. Chem. 1995, 67, 1575-1579.

X

Abstract published in Advance ACS Abstracts, September 15, 1996.

Analytical Chemistry, Vol. 68, No. 21, November 1, 1996

3739