FOCUS: COOKS, 2006 DISTINGUISHED CONTRIBUTION IN MASS SPECTROMETRY AWARDEE ACCOUNT AND PERSPECTIVE Hybrid Mass Spectrometers for Tandem Mass Spectrometry Gary L. Glisha and David J. Burinskyb a Department of Chemistry, University of North Carolina, Chapel Hill, North Carolina, USA b GlaxoSmithKline, Five Moore Drive, P.O. Box 13398, Research Triangle Park, North Carolina, USA Mass spectrometers that use different types of analyzers for the first and second stages of mass analysis in tandem mass spectrometry (MS/MS) experiments are often referred toas “hybrid” mass spectrometers. The general goal in the design of a hybrid instrument is to combine different performance characteristics offered by various types of analyzers into one mass spectrometer. These performance characteristics may include mass resolving power, the ion kinetic energy for collision-induced dissociation, and speed of analysis. This paper provides a review of the development of hybrid instruments over the last 30 years for analytical applications. (J Am Soc Mass Spectrom 2008, 19, 161–172) © 2008 American Society for Mass Spectrometry andem mass spectrometry (MS/MS), in a very “high” ion kinetic energies (5–20 keV), whereas “low” generic description, is a process in which an ion ion kinetic energies (Ͻ50 eV) are typical in quadrupole Tformed in anion source is mass-selected in the mass filters and ion traps. The ion kinetic energy is an firststage of analysis, reacted, and thenthe charged important parameter in MS/MS experiments because products from the reaction are analyzed in the second the most common reaction involves colliding the ion of stage of analysis. The type and quality of data that is interest with a target gas atom or molecule. When obtained can vary greatly depending upon the type of performing ion-neutral collision experiments, the pos- analyzer used in the first and second stages of analysis, sible reactions that can be accessed (e.g., collision- and the type ofreaction performed betweenthe stages induced dissociation, collisional cooling, charge permu- of analysis. The reactions that can be done also can tation, ion/molecule reaction) depends upon the ion depend upon the type of analyzer. Over the years there kinetic energy.The appearance of the MS/MS spectrum have been a variety of means developed to measure the can change drastically as a function of the collision (ion mass-to-charge ratio of gas-phase ions.The most com- kinetic) energy. A related factor that is equally impor- mon methods involve: dispersion based on ion momen- tant, but often not considered, is the time frame of the tum or kinetic energy (magnetic and electric sector experiment, that is, the elapsed time between the colli- instruments); separation in time based onion velocity sion event and the second (or next) stage of analysis. (time-of-flight); transmissionthrough an electrody- With high kinetic energies, the experimental time frame namic field (quadrupole mass filter); and periodic mo- is typicallyquite short (i.e., microseconds). Short disso- tion in a magnetic or electrodynamic field (ion traps). ciation times mean a greater kinetic shift. A large kinetic There are differences in the experimental parameters shift typically favors dissociations that occur faster, but associated with these various analysis methods that are that also require more internal energy to dissociate fast pertinent to the MS/MS experiment. Some parameters enough to be observed on the time scale of the experi- are obvious, typically related to the performance of the ment. Not only does this lead to differences in the mass analyzer while others are more subtle, related to appearance of high-energy and low-energy MS/MS the reactions/chemistry occurring between the stages spectra, but also to reduced MS/MS efficiency for of analysis. Many times these different parameters are higher kinetic energy CID. used to categorize MS/MS experiments. Another distinguishing feature by which tandem mass One parameter is the ion kinetic energy. Sector and spectrometry experiments can be classified or categorized time-of-flight (TOF) instruments typically operate at is whether the analysis and excitation events are separated spatially or temporally. Techniques in which the ions are separated “in space”—sector, TOF analyzers, and quad- Address reprint requests to Dr. G. L. Glish, Department of Chemistry, University of North Carolina-Chapel Hill, CB# 3290, Chapel Hill, NC rupole mass filters—are commonly referred to as “beam” 27599-3290, USA. E-mail: [email protected] techniques because ions traverse the analyzer as a contin- Published online January 9, 2008 © 2008 American Society for Mass Spectrometry. Published by Elsevier Inc. Received October 17, 2007 1044-0305/08/$32.00 Revised November 7, 2007 doi:10.1016/j.jasms.2007.11.013 Accepted November 8, 2007 162 GLISH AND BURINSKY J Am Soc Mass Spectrom 2008, 19, 161–172 Table 1. Comparison of mass analyzers with respect to key performance attributes Resolving power Mass accuracy Analyzer type Kinetic energy MS-1 MS-2 MS-1 MS-2 Measurement timea (seconds) Magnetic (B) sector High Medium Medium ppt ppt 10Ϫ5 Electric (E) sector Highnab Low nab ppt10Ϫ5 Electric and magnetic sectorc High High Medium ppm ppt 10Ϫ5 Time-of-flight (TOF) High Low Mediumd ppm ppm10Ϫ5 Quadrupole mass filter (Q) Low Low Low ppt ppt10Ϫ4 Quadrupole ion trap(QIT) Low Low Low ppt ppt10Ϫ2 FTICR Low High High ppm ppm 10Ϫ1 Orbitrap High nab High nab ppm 10Ϫ1 aThe time, after activation, for the ions to react prior to the second stage of analysis. bna: not applicable; this analyzer has not been used for the first stageofanMS/MS experiment. There is currently no mode of operation in which a parent ion can be mass selected for a subsequent stage of analysis. cWhen electric and magnetic sectors are used as a double focusing mass spectrometer for one stage of analysis, either prior to (MS-1) the reaction, or subsequent to (MS-2) the reaction. dThe resolving power is medium if the TOF ion axis is orthogonal to the first stage of MS; if the axes are coaxial, the resolving power is low. uous or pulsed beam (analogous to a laser beam). For the of an MS/MS experiment is no different than that beam techniques, every stage of mass-to-charge analysis is measured for a normal mass spectrum when using that done in a separate analyzer, such that the two (or more) analyzer. However, that is not necessarily true for the stages of analysis are separated (physically) in space. resolving power in the second stageof mass analysis. Conversely, trapping instruments such as quadrupole ion For example, because of the kinetic energy release traps, Fourier transform ion cyclotron resonance (FTICR) associated with dissociation of an ion, a single magnetic spectrometers, and orbitraps, characterize ions based sector or electric sector exhibits low mass resolving upon the frequency of their motion in a defined space. In power if CID occurs at high kinetic energies. Con- theMS/MS experiment using quadrupole ion traps and versely, if a double focusing electric/magnetic sector FTICR spectrometers, all stages of the experiment can combination is used medium resolving power is ob- performed in the same analyzer, with the different stages served. Table 1 summarizes the performance parame- being separated in time rather than spatially. The overall ters for the various analyzers. experimental efficiency is almost alwayshigher because 1) Hybrid MS/MS instruments use various combina- the ions do not have to be transferred from one analyzer to tions of these analyzers to obtain desirable performance another, and 2) there is typically a much longer time characteristics for the phenomena being investigated. period for reaction and dissociation, so the kinetic shift is Typically, resolving power and measurement accuracy small. However, the overall experiment can take signifi- in each stage of the mass spectral analysis are of cantly more time to complete compared withbeam-type primary importance. However, the kinetic energy used instruments. for CID can also be a critical parameter, depending on The last two attributes that can be used to classify the the information being sought. Additionally, the speed various mass analysis techniques are resolving power of analysis (duty cycle) can also be an important param- Ϫ [(m/z)(⌬m/z) 1] and measurement accuracy. Mass mea- eter. The development of many early hybrid instru- surement accuracy is either “high”, expressed in parts ments was motivated by the desire for fast MS/MS per million (ppm),or“low,” expressed in parts per spectral acquisition times required by the time scale of thousand (ppt). FTICR spectrometers, orbitrapsand a chromatographic separation. The early hybrid instru- TOF analyzers can routinely achieve low- or sub-ppm ments (1980s) all usedbeam type analyzers (beam-beam mass accuracies, while the other analyzers (i.e., quad- instruments) because these were the standards of the rupole mass filters or ion traps) are typically in the ppt day, with trapping instruments just being introduced as range. In terms of mass resolving power, analyzers of commercially viable mass analyzers. Two of these various types are capable of either low, medium, or beam-beam type hybrid instruments evolved into com- high resolving power. Clearly, relative terminology mercial instruments, sector/Q and Q/TOF, with the such as low, medium, andhigh can be debated.That latter one still being a popular instrument today. Sub- being said, for the purpose of this discussion, the sequently (1990s), beam-trap type instruments were following definitions apply to those terms: “low”, less explored, followed by trap-beam instruments. A hand- than 1000; “medium”, 1000 to 10,000; “high”, greater ful of beam-trap instruments (sector/QIT) have been than 10,000. It should also be noted that whilethe terms commercialized. The trap-beam instrument closely re- “resolving power” and “resolution” are often used lated to the Q/TOF, the QIT/TOF, has become commer- interchangeably in the mass spectrometry literature, cially available.