Anal. Chem. 1997, 69, 105-107
Synchronized Scanning of the First and Second Mass Spectrometers on Tandem Double-Focusing Mass Spectrometers Fred H. Strobel
Emory University Mass Spectrometry Center, Department of Chemistry, Emory University, Atlanta, Georgia 30322
A method is described to synchronize the scans of the first and second double-focusing mass spectrometers of a tandem double-focusing mass spectrometer. The scans are synchronized by scanning the first mass spectrometer with voltages instead of the magnetic field. The method is demonstrated with scans that provide all of the precursor ions that produce a selected product ion (precursor ion scan). These precursor ion scans are compared to the previous method of obtaining precursor ion scans on a tandem double-focusing mass spectrometer. The precursor ion scans using the synchronized scanning allow for the use of high-energy collisions on any tandem double-focusing mass spectrometer, while limiting the mass range possible in a single scan. The limited mass range may be corrected by obtaining several spectra with different magnetic fields on the first mass spectrometer. Tandem mass spectrometers with two double-focusing mass spectrometers provide for high precursor and product ion resolutions. In addition, tandem double-focusing mass spectrometers allow kilovolt collision energies to be used for collision-induced dissociation (CID). One problem with these tandem doublefocusing mass spectrometers is that they have lacked a flexibility in their normal use. Typically, the tandem double-focusing mass spectrometers have been used for determining the product ions of a single mass-selected precursor ion (product ion scan). A few unique experiments have been performed on tandem doublefocusing mass spectrometers, but these experiments involve scanning only one of the double-focusing mass spectrometers.1-3 The main reason for the inability to perform the complex scans has been the inability to scan two magnets simultaneously. This difficulty is caused by the hysteresis in the magnets. A method to switch precursor ions by controlling two magnets has been demonstrated; however, the time involved in setting and stabilizing the magnetic field would not allow for simultaneously scanning the mass spectrometers.4 An alternate method of controlling the mass transmitted through a double-focusing mass spectrometer is by the voltages applied to the electrostatic analyzer (ESA) and the source. This method is called accelerating voltage scans, which are commonly used in high-resolution measurements. (1) Ballard, K. D.; Gaskell, S. J.; Jennings, R. K. C.; Scrivens, J. H.; Vickers, R. G. Rapid Commun. Mass Spectrom. 1992, 6, 553-559. (2) Wu, Z.; Bordas-Nagy, J.; Fenselau, C. Org. Mass Spectrom. 1991, 26, 908911. (3) Scrivens, J. H.; Rollins, K.; Jenning, R. C. K.; Bordoli, R. S.; Bateman, R. H. Rapid Commun. Mass Spectrom. 1992, 6, 272-277. (4) Takao, T.; Gonzalez, J.; Yoshidome, K.; Sato, K.; Asada, T.; Kammei, Y.; Shimonishi, Y. Anal. Chem. 1993, 65, 2394-2399. S0003-2700(96)00863-3 CCC: $14.00
© 1996 American Chemical Society
Controlling the first mass spectrometer with voltages could allow for the scan of the first mass spectrometer to be synchronized with scans performed on the second mass spectrometer. The synchronization of the scans on the two mass spectrometers would give the flexibility of triple quadrupole mass spectrometers to tandem double-focusing mass spectrometers; however, the tandem double-focusing mass spectrometers will have the advantage of high mass resolution and the ability to use high-energy collisions in CID. Specifically, the developed scans could be used to determine the ions containing specific functional groups in a mixture or in a chromatograph. These scans determine all the precursor ions that either form a specific product ion (precursor ion scan) or lose a specific neutral (neutral loss scan). These scans could only be performed on instruments that couple scanning mass spectrometers (triple quadrupoles, hybrids, or multisector mass spectrometers).5 Currently, the precursor ion scans can be performed on a tandem double-focusing mass spectrometer; however, high-energy CID for precursor ion scans are limited to experiments performed on a ZAB-T or similar instrument that allows a large kinetic energy spread through the ESA on the second mass spectrometer.3 This method of precursor ion scans (Scrivens method) was developed by Scrivens et al.3 Scrivens’ method used a floating cell to reduce the kinetic energy spreads in the product ions to less than the kinetic energy spread transmitted through the ESA on the second mass spectrometer. The low kinetic energy spread allows the second mass spectrometer to be set to transmit the product ion. The first mass spectrometer is scanned to produce the precursor ion scan.3 High-energy collisions could be used on any tandem doublefocusing mass spectrometer for precursor ion and neutral loss scans by synchronizing the scans of the two mass spectrometers. Implementation of the synchronized scans is described in terms of a traditional four-sector mass spectrometer (contains two identical double-focusing mass spectrometers with one magnet and one ESA each). Appropriate changes may be necessary for nontraditional geometries. The synchronized scans are implemented by scanning the magnetic field of the second mass spectrometer. The appropriate voltages, which need to be applied to the source and the ESAs, are shown in Table 1. These voltages are a function of the magnetic fields, instrumental constants, and the m/z of the scan. By maintaining a constant magnetic field on the first mass spectrometer, the voltages need to be synchronized only to the scan of the second magnetic field. Many methods (5) Hoffmann, E. de J. Mass Spectrom. 1996, 31, 129-137.
Analytical Chemistry, Vol. 69, No. 1, January 1, 1997 105
Table 1. Voltages Needed To Be Applied on a Four-Sector Mass Spectrometer to Implement Various Scansa ESA voltage scan type product ion scan (m1 ) constant) precursor ion scan (m2 ) constant) neutral loss scan (m3 ) constant)
source voltage 2/m
k (cB1
1)
first 2/m
cB1
second 1
cB2B1/m1
k (cB1B2/m2)
cB1B2/m2
cB22/m2
k (cB1(B1 B2)/m3)
cB1(B1 B2)/m3
cB2(B1 B2)/m3
a Definitions: m , m/z of precursor ion passed through MS-1; m , 1 2 m/z of product ion passed through MS-2; m3, m/z of the “neutral loss” ()m1 - m2); B1, magnetic field of MS-1 magnet; B2, magnetic field of MS-2 magnet; E1, voltage on first ESA; E2, voltage on second ESA; Vacc, source voltage ()kE1); k, constant that makes the previous statement true when a source formed ion is transmitted though MS-1; c, instrumental constant that takes into account the geometry of the instruments makes equations; m1 ) cB12/E1 and m2 ) cB22/E2 true (assumes same geometry on both mass spectrometers).
have been previously described to synchronize the appropriate voltages with the scan of a magnetic field.6-10 Any of these methods may be used. In this paper, only precursor ion scans will be demonstrated. These precursor ion scans (linked precursor ion scans) will be compared to the Scrivens method. The precursor ions scans will be evaluated on the basis of the criteria proposed by Scrivens et al.3 Their criteria evaluated the scans in terms of the (a) selectivity of product ion, (b) resolution of the precursor ion, (c) available mass range, (d) collision energy, and (e) mass accuracy. EXPERIMENTAL SECTION The experiments were performed on a JEOL JMS-SX102/ SX102A/E (B1E1/B2E2/E3).11 The scan of the third ESA was synchronized with the second ESA, and the collector slit on the third ESA was left open. This procedure allows the five-sector mass spectrometer to behave identically to a four-sector mass spectrometer. The two double-focusing mass spectrometers were set at 1000 resolution. Ar collision gas at 75% beam attenuation was used for CID. Ions were formed by fast-atom bombardment (FAB) using 6-kV Xe atoms. The spectra are an average of 50200 scans. The precursor scans were performed by using a computercontrolled precursor ion scan of the second double-focusing mass spectrometer (MS-2), while the first double focusing mass spectrometer (MS-1) was controlled by an external circuit. The external circuit used the reference voltage of the second ESA as input to produce a signal proportional to the square root of the reference voltage. The proportionality of the above circuit was variable through a 10-turn potentiometer in the circuit to allow for setting the magnet field on different values. The output of the external circuit was used as a reference voltage to control the accelerating voltage and the first ESA. The data from the (6) Bruins, A. P.; Jennings, K. R.; Evans, S. Int. J. Mass Spectrom. Ion Phys. 1978, 26, 395-404. (7) Haddon, W. Anal. Chem. 1979, 51, 983-988. (8) Haddon, W. F. Org. Mass Spectrom. 1980, 15, 539-543. (9) Shushan, B.; Boyd, R. K. Anal. Chem. 1981, 53, 421-427. (10) Sato, K.; Asada, T.; Ishihara, M.; Kunihiro, F.; Kammei, Y.; Kubota, E.; Costello, C.; Martin, S. A.; Scoble, H.; Biemann, K. Anal. Chem 1987, 59, 1652-1659. (11) Strobel, F. H.; Adams, J. J. Am. Soc. Mass Spectrom. 1995, 6, 1232-1242.
106 Analytical Chemistry, Vol. 69, No. 1, January 1, 1997
Figure 1. Linked precursor ion scans of FAB produced ions of leucine enkephalin in glycerol (insets show (M + H)+ region): (A) precursors of m/z 279; (B) precursors of m/z 278.
computer was exported to a 486 PC and converted to a correct mass scale. RESULTS AND DISCUSSION (A) Selectivity of Product Ion. It is important to have good product ion selectivity. In the linked precursor ion scans, the selectivity is determined by the resolution of MS-2. The ability to apply the correct voltages for the magnetic field on MS-2 will limit the resolution of MS-2 in link precursor ion scans. In practice, the Scrivens method should allow for higher product ion selectivity, because the voltages do not need to be changed with the magnetic field. The product ion selectivity of linked precursor ion scans is adequate for most purposes. For example, Figure 1 shows the precursor ion scans of m/z 278 and 279 from FAB-produced ions of leucine enkephalin using a glycerol matrix. The (M + H)+ ion is naturally found in both spectra; however, ions from C-terminal sequence fragments (Y′′)12 are only observed in the m/z 279 (Y2′′) spectrum. In addition, the N-terminal sequence fragment A4 ion is only observed in the m/z 278 (B3) spectrum. Both spectra do contain peaks that can be attributed to the glycerol matrix background, most notably a precursor ion 92 m/z above each product ion. (12) The nomenclature for the peptide fragments is that proposed in: Roepstorff, P.; Fohlman, J. Biomed. Mass Spectrom. 1984, 11, 601.
accelerating voltage. Precursor ions are observed out to m/z 2211. This limited mass range in a single scan is a disadvantage over the Scrivens method, where a single scan could acquire the entire mass range. (D) Collision Energy. The collision energy for each precursor ion is not a constant in the linked precursor ion scans but decreases linearly with the mass of the precursor ion; however, the collision energies can be maintained in the kilovolt energy range. The Scrivens method provides a constant energy; however, the collision energy must be less than the kinetic energy spread transmitted through the ESA on MS-2. For the ZAB-T used by Scrivens et al., the kinetic energy spread transmitted through the ESA was ∼2 kV3. For the JEOL JMS-SX102/SX102A/E used in these experiments, the kinetic energy spread transmitted through the ESA will be ∼100 V.11 The ESA on most tandem doublefocusing mass spectrometers would limit the precursor ion scans using the Scrivens method to low-energy collisions. (E) Mass Measurement Accuracy. There should be no significant advantage in mass measurement accuracy to either of the precursor ion scans. It should be noted that a correction factor of 1.0018 was used in converting the spectra in the paper to the correct mass scale. The correction factor was determined by using the correction factor needed for two precursor ions in the CsI spectra and was found to be applicable to the other ions in the CsI spectra. This correction factor was found to be applicable to other spectra.
Figure 2. Linked precursor ion scans of the precursors of Cs2I+ formed by FAB of CsI: (A) MS-1 magnetic field set to transmit m/z 393 at 10-kV accelerating voltage; (B) MS-1 magnetic field set to transmit m/z 1172 at 10-kV accelerating voltage.
(B) Precursor Resolution. Precursor resolution is determined by the resolution of MS-1. In Figure 1, the isotopes are separated for the (M + H)+ ion in each spectrum, indicating a resolution greater than 556; however, attempts to reproduce the precursor ion scan of m/z 152 in dibenzosuberone, which Scrivens et al. used to demonstrate a resolution of greater than 5000, were not successful.3 This failure may be related to electronic noise in the external circuit used to control MS-1. (C) Available Mass Range. The available mass range is limited by the sensitivity of the source, decreasing dramatically with the decrease in the accelerating voltage. In addition, the product ions’ kinetic energy is inversely proportional to the square of the precursor ion mass, which would cause the product ions’ kinetic energy to decrease dramatically with the precursor ion mass. The lower kinetic energy of the product ion will reduce transmission through MS-2. Examination of CsI clusters produced by FAB dissociating to form Cs2I+ demonstrates the limits of the linked precursor ion scans. A precursor ion scan with the MS-1 magnetic field set to transmit m/z 393 at 10-kV accelerating voltage is shown in Figure 2a. No precursors above m/z 1172 were observed. Higher mass ranges can be observed by changing the MS-1 magnetic field to transmit a higher m/z at 10-kV accelerating voltage. For example, Figure 2b is a precursor ion scan of Cs2I+ with the MS-1 magnetic field set to transmit m/z 1172 at 10-kV
CONCLUSIONS Precursor ion scans on tandem double-focusing mass spectrometers can be obtained by scanning of MS-1 with voltages, while scanning MS-2. In addition, a similar technique should be possible for neutral loss scans. The mass range obtainable in a single spectrum is limited based on the minimum collision energy needed; however, a larger mass range may be obtainable by getting several spectra. The spectra would be obtained by setting the MS-1 magnetic field to transmit progressively higher masses at full accelerating voltage. The problem associated with the precursor ion scans is the kinetic energy of the product ions will become extremely low. The low kinetic energy of the product ion reduces their transmission through MS-2. The problem may be corrected by changing the magnet field to obtain a satisfactory spectrum over different ranges. Another solution will be to use a floated collision cell. In conclusion, this paper demonstrated the possibility of simultaneously scanning of two double-focusing mass spectrometers. ACKNOWLEDGMENT I acknowledge the use of the JEOL JMS-SX102/SX102A/E, which was funded as a shared instrument by NIH (Grant 1S10RR06276) and NSF (Grant CHE-9119862). I thank the Chemistry Department’s electronic shop for their advice on the external circuit used in these experiments, the Chemistry Department for giving me the opportunity to pursue this work, and Dr. Kermit Murray for some helpful comments. Received for review August 23, 1996. Accepted October 16, 1996.X AC9608634 X
Abstract published in Advance ACS Abstracts, November 15, 1996.
Analytical Chemistry, Vol. 69, No. 1, January 1, 1997
107
