Vanderbilt University
spacer.jpg (674 bytes)
spacer.jpg (674 bytes)


 

Mass Spectrometry Tutorial

2. Instrumentation


  1. Basic Concepts

    A mass spectrometer is an instrument that produces ions and separates them in the gas phase according to their mass-to-charge ratio (m/z). Today a wide variety of mass spectrometers is available, ranging from benchtop detectors for gas chromatography to warehouse sized instruments such as accelerator mass spectrometers. All of these share the capability to assign mass-to-charge values to ions, although the principles of operation and the types of experiments that can be done on these instruments differ greatly.

    Basically, a mass spectrometric analysis can be envisioned to be made up of the following steps:

    Sample Introduction ® Ionization ® Mass Analysis ® Ion Detection/Data Analysis

    Samples may be introduced in gas, liquid or solid states. In the latter two cases volatilization must be accomplished either prior to, or accompanying ionization. Many ionization techniques are available to produce charged molecules in the gas phase, ranging from simple electron (impact) ionization (EI) and chemical ionization (CI) to a variety of desorption ionization techniques with acronyms such as FAB, PD, ES and MALD.

    Mass spectrometers are operated at reduced pressure in order to prevent collisions of ions with residual gas molecules in the analyzer during the flight from the ion source to the detector. The vacuum should be such that the mean free path length of an ion, i.e., the average distance an ion travels before colliding with another gas molecule, is longer than the distance from the source to the detector. For example, at a pressure of 5x10-5 torr for instance, the mean free path length of an ion is approximately one meter, i.e., about twice the length of a quadrupole instrument. Thus, the introduction of a sample into a mass spectrometer usually requires crossing of a rather large pressure drop, and several means have been devised to accomplish this. Gas samples may be directly connected to the instrument and metered into the instrument via a needle valve. Liquid and solid samples can be introduced through a septum inlet or a vacuum-lock system. However, when connecting continuous introduction techniques like gas chromatography (GC), high performance liquid chromatography (HPLC) or capillary electrophoresis (CE), special interfacing becomes imperative to prevent excessive gas load.

    Once ions are formed, they can be accelerated, focused or brought to resonance by electrical and magnetic fields. The separation and detection of biologically-relevant ions of different m/z ratios will be discussed below in section 2.C.

  2. Ionization

    1. Electron Impact and Chemical Ionization

      Volatile substances can be ionized by electron (impact) ionization in a process involving the interaction of the gaseous sample with an electron beam generated by a heated filament in the ion source (see Figure 1 ). The electron energy is defined by the potential difference between the filament and the source housing and is usually set to 70 eV (~1.12x10-17J). A magnetic field keeps the electron beam focussed across the ion source and onto a trap. Upon impact with a 70 eV electron, the gaseous molecule may lose one of its electrons to become a positively charged radical ion,

      M + e- ® M + 2e-

      where M is termed the molecular ion. It carries an unpaired electron and can occupy various excited electronic and vibrational states. If these excited states contain enough energy, bonds will break and fragment ions and neutral particles will be formed. With an electron energy of 70 eV, enough energy is transferred to most molecules to cause extensive fragmentation.

      All ions are subsequently accelerated out of the ion source by an electric field produced by the potential difference applied to the ion source and a grounded electrode. A 'repeller' serves to define the field within the ion source. Depending on the lifetime of the excited state, fragmentation will either take place in the ion source giving rise to stable fragment ions, or on the way to the detector, producing metastable ions. If no fragmentation occurs before the ion strikes the detector, a signal for the molecular ion is generated. The mass spectrum obtained from recording all of these ions contains signals of varying m/z and intensities, depending on the numbers of ions that reach the detector. The fragmentation pathways of the molecular ion depend on the structure of the molecule, such that similar structures give similar mass spectra. A special monograph is available for those wishing to learn more about ion fragmentation and the interpretation of mass spectra ( 3 ). Also, mass spectral libraries are available today that contain over 160,000 spectra which can be used to help identify unknowns and search for common substructures. The sensitivity of 70 eV EI varies widely because the signal arising from an analyte can be spread over many fragment ions in some cases. Fragmentation can be reduced by choosing an electron energy close to the ionization potential of the neutral molecule (typically 10 - 12 eV for simple organic molecules).

      Chemical ionization (CI) relies on the interaction of the molecule of interest with a reactive ionized reagent species. Many such reactants are gaseous Bronsted acids. For example, some of the most widely used reactant species are generated by EI from methane. The first ion formed is CH4 which then reacts to give the Bronsted acid CH5 + according to the following reaction

      CH4 + CH4 ® CH5 + + CH3·

      If a neutral molecule M in the source has a higher proton affinity than CH4, the protonated species MH+ will be formed in an exothermic reaction.

      M + CH5+ ® MH+ + CH4

       

      The instrumentation used for CI is very similar to that used for EI. The major difference in the design of the source is that it is more gastight so that the reactant gas is retained at higher pressures in order to favor ion/molecule reactions. The pressure inside the CI ion source is typically of the order of 0.1 - 1 torr. A wide variety of reagents have been reported in the literature, some of which lead to fragmentation (e.g. hydrogen (H2) for which the reactant ion is H3+). Other reagents give rise to mass spectra that, depending on the analyte, mainly show the protonated molecule (e.g. i-butane (C 4H10) for which the reactant ion is C4 H9+). Negative ions can be generated in a similar manner by a reaction involving an election capture process ( 4 ).

    2. Plasma Desorption Mass Spectroscopy (PDMS)
    3. A breakthrough in the analysis of biomolecules came in 1974 with the introduction of PDMS ( 5 ). This technique uses 252Cf fission fragments to desorb large molecules from a target. The target is made of a thin aluminum foil, often covered with a layer of nitrocellulose, to which a droplet of the sample solution is applied. The adsorption of proteins to nitrocellulose is believed to be based on hydrophobic interactions and allows salts to be washed off and chemical reactions to be carried out on the target. Alternatively, the sample can be electrosprayed directly onto Ni or Al foil, a technique which is more effective for smaller peptides.

      Two atomic particles are produced by the 252Cf fission reaction, one causing desorption of the analyte and the other providing the start signal for the time-of-flight measurement. A time-of-flight mass analyzer, described below, is generally used for ion separation. The general arrangement of the instrument is shown in Figure 2 . Similar to other desorption techniques, cluster formation and multiple charging is observed. Suppression effects, i.e. the inability to ionize some molecules due to the presence of other compounds present, are also sometimes observed.

      PDMS has a reasonably good sensitivity with peptides and small proteins, and typically about 10 picomoles is needed for molecular mass determinations of such a protein. Perhaps its major virtue is that it is a simple method, both easy to use with easily interpretable spectra, and well-suited for the protein chemist and others not expert in mass spectrometry who wish to utilize mass specific analyses in their research programs.

    4. Fast Atom Bombardment (FAB)
    5. FAB is another soft ionization technique, i.e., one that yields minimal fragmentation, that performs well for polar and thermally labile compounds (6 ). In a typical FAB analysis, the sample is usually dissolved in an appropriate matrix, a viscous solvent, in order to keep the sample in the liquid state. Some of the more common liquid matrices are glycerol, 1-thioglycerol, a mixture of dithiothreitol and dithioerythritol, 3-nitrobenzyl alcohol, and triethanolamine. One major role of the matrix, because of its low freezing point, is to keep the sample in a liquid state as it enters the high vacuum ion source. This matrix also reduces damage to the analyte caused by the high energy bombarding particle.

      In FAB ionization, the sample droplet is bombarded with energetic atoms (Ar, Xe) of 8-10 keV kinetic energy. Ions (e.g., Cs+) can be used as the bombarding particle in a similar technique termed liquid secondary ion mass spectrometry (LSIMS). The general process leading to the formation of the molecular ion is depicted in Figure 3 and involves several different mechanisms including ejection of preformed ions. The phase transition, i.e., the conversion of liquid sample to gaseous ions on bombardment, is believed to result from sputtering, a redistribution of momentum from the initial high energy particle by way of a cascade of collisions within the matrix. The formation of protonated molecular ions (M+H)+ or other cationized species such as (M+Na)+ in the positive ion mode, and (M-H) - in the negative ion mode, can be attributed to both gas phase reactions and solution chemistry. Doubly-charged ion species and dimeric cluster ions of the analyte are occasionally observed. The major advantage of FAB is that it is easy and fast to operate, the spectra are simple to interpret, and the source itself is easily retrofitted on most mass spectrometers. However, one of the major disadvantages of the FAB technique is that it requires a high concentration of the organic liquid matrix (typically 80 to 95% glycerol), overall giving only moderate sensitivity. Matrix cluster ions can, in some cases, dominate the mass spectrum. In addition, damage to the matrix caused by particle bombardment gives intense chemical background. It has been shown that discrimination in ionization efficiency of one analyte over another, due to differences in hydrophobicity and surface activity, causes problems when analyzing mixtures especially in quantitative applications. In some cases, the matrix also directly reacts with the analyte, forming radical anions or causing reduction of the analyte. The desorption process also produces a great many sputtered neutral molecules in addition to ions.

      With the development of continuous-flow FAB (CF-FAB), several of the above mentioned problems have been eliminated. Instead of applying the matrix with the sample to the probe tip, a probe was designed that continuously delivers sample solution to the target at flow rates of up to about 10 m L/min. As a consequence, less organic matrix is necessary to keep the sample liquid, which results in an increase in the signal-to-noise ratio due to lower chemical background. A major advantage of CF-FAB is its usefulness for flow-injection analysis and on-line reaction monitoring, as well as coupling to HPLC and capillary electrophoresis (discussed later in this chapter). Additional details of the operation and applications of CF-FAB can be found in a specialist monograph ( 7 ).

      Several mixtures have been reported for mass calibration in FABMS. For low mass applications, one can simply use the organic matrix clusters as an internal calibrant (e.g. glycerol [Glyn+H]+ or [Glyn-H]-), where n is a positive integer. Cesium iodide clusters of the form [Cs nIn-1]+ give very stable signals up to 10,000 Da. For better accuracy, mixtures of different salts, nonionic surfactants or even gold clusters (Aun)-, generated from internal standards are preferred. Today, FAB is widely used on quadrupole and sector instruments.

    6. Thermospray and Particle Beam
    7. Thermospray ionization was introduced in 1983 for the coupling of HPLC at conventional flow rates (0.5 to 1.5 ml/min.) to a mass spectrometer (8 ). The effluent from the HPLC column is vaporized under reduced pressure by heating a stainless steel tube of 0.10 to 0.15 mm inner diameter, as shown in Figure 4 . The resulting supersonic jet contains small droplets that vaporize further due to the hot gas in this low pressure region of the ion source. Complete evaporation of the solvent from the liquid droplets produces gas phase ions from ionic compounds in the sample solution or from gas phase chemical ionization when an auxiliary filament or low-current discharge device is used. Ionization requires polar or charged species and volatile buffers; the filament arrangement is used for semivolatile samples, and the discharge device for highly aqueous effluents. The temperature of the vaporizer is critical and has to be adjusted for a given solvent composition to give best results. Ions are drawn into the analyzer by electric fields and enter through an orifice of about 0.5 mm diameter. Thermospray is considered a soft ionization technique and induces only limited fragmentation of the analyte.

      The particle beam (PB) interface, derived from the MAGIC interface (Monodisperse Aerosol Generation Interface for Chromatography), has elements in common with thermospray, but gives spectra with more fragment ions (9 ). Again, formation of an aerosol is the initial step, followed by dispersion caused by a gas stream (usually helium), and desolvation. In the original interface, a momentum separator then separates the lighter dispersion gas and vaporized solvent from the sample particles which have higher momentum, and finally ions enter the mass spectrometer ion source as a beam of charged particles. An EI source can be used to produce mass spectra that compare well to those recorded for prevaporized samples in conventional EI sources. The PB mass spectrum, however, has a relatively intense background at low mass due to the solvent and this limits sensitivity. PB is less sensitive than thermospray or electrospray and is not best suited for the analysis of ionic components, high molecular mass samples, and thermally labile compounds.

    8. Electrospray (ES)
    9. ES ionization has had a tremendous impact over the last few years on the use of mass spectrometry in biological research. It was the first method to extend the useful mass range of instruments to well over 50,000 Da. Although introduced in its present form in 1984, the technique goes back to investigations of the electrically assisted dispersion of liquids at the beginning of this century. In fact, a major discovery took place almost unnoticed in 1968, when Malcolm Dole and coworkers were able to bring macromolecules into the gas-phase at atmospheric pressure (10 ). This was done by spraying a sample solution from a small tube into a strong electric field in the presence of a flow of warm nitrogen to assist desolvation and then measuring the ions by ion mobility techniques. Further innovative experiments in this field led to the introduction of the ES ionization source (11 ). Since then, a wide range of biomolecules has been investigated by ES. The sample is usually dissolved in a mixture of water and organic solvent, commonly methanol, isopropanol or acetonitrile. It can be directly infused, or injected into a continuous-flow of this mixture, or be contained in the effluent of an HPLC column or CE capillary.

      The ES source design is simple, with spray formation occurring in a high voltage field as shown in Figure 5 . In one proposed mechanism, ion formation is believed to result from an ion evaporation process, first proposed in 1976 ( 12 ). A spray of droplets is caused by electrostatic dispersion from the liquid ejected from the capillary tip. Aided by the heated bath gas (usually nitrogen), the droplets undergo declustering, losing solvent molecules in the process and eventually producing individual ions. In another proposed mechanism, desolvation of the droplets leads to an increasing charge density on the droplet surface that will eventually cause a coulombic explosion that leads to individual ions. Whatever the detailed mechanism, ions are formed at atmospheric pressure and enter a cone shaped orifice, which acts as a first vacuum stage where they undergo free jet expansion. A skimmer then samples the ions and guides them to the mass spectrometer.

      Spray formation is the crucial part of the ES technique. It is usually advisable to filter all the solvents and high concentrations of electrolytes should be avoided because they can lead to electrical breakdown and unstable operating conditions. High flow rates, compatible with full-bore and microbore HPLC, can be accommodated today by using aids such as a heated nebulizing gas to assist spray formation. Often an organic solvent, sometimes referred to as sheath liquid, is added to the HPLC effluent via a coaxial arrangement to establish a more stable spray. It has been pointed out that the electrospray chamber can be considered an electrolysis cell, where one half-cell reaction takes place at the stainless steel needle / liquid interface with transport of electrons via the power supply to the counter electrode (positive mode), and transport of positively charged droplets and ions through the gas phase to the counter electrode. This suggests the possibility of redox reactions occurring in the source, although under normal conditions this does not appear to be a major problem.

      For macromolecules, ions entering the mass spectrometer usually each have a high number of charges. Because mass spectrometers measure mass-to-charge ratios rather than mass itself, it is possible for high molecular mass molecules to carry sufficient numbers of charges to fall within the m/z range of a quadrupole mass filter, typically m/z 2500 (see below). As shown in Figure 6A , high molecular mass ions often have a wide distribution of charge states. The figure shows the mass spectrum of horse myoglobin (m.w. 16951.5) taken at low resolution with charge states from +12 to +24 readily observed. This distribution of ions permits the calculation of the molecular mass of the original analyte from any two neighboring ions at m/z values m1 and m2, with n1 and n2 charges, respectively. If m1 < m2, and n2 = n1-1, then

      M = n1(m 1-mA) = n2 (m2-mA)          (1)

      n2 = (m1 -mA)/(m2-m1)                       (2)

      where M represents the molecular mass of the uncharged molecule and mA the mass of the charged adduct A (e.g. H+, Na+, NH4+). Thus the equations can be solved to give both n2 and M. The measurement of the charge state distribution of large molecules is not always easy or reproducible because it can readily be changed by relatively small changes in analysis conditions, changing pH, adding solvents or salts, partial denaturation of the protein, breaking disulfide bonds, etc. Further details are given later in specific applications.

      If high resolution capabilities are available, one can in suitable cases resolve the individual carbon isotope peaks of a given charge state. Figure 6B shows the high resolution ES ion distribution of a single charge state of horse myoglobin. Since isotope peaks should be whole integer m/z units apart, whereas in Figure 6B they are approximately 0.06 m/z units apart, then these ion species are (approximately) in the +17 charge state. The average molecular mass of the protein can then be easily calculated. The attainable mass accuracy for measuring molecular masses of biomolecules with ES on a magnetic sector instrument is typically about 0.001% (10 ppm) and somewhat better when using internal calibration. Common calibration compounds are nonionic surfactants (with sodium adducts) or the different charge states of standard proteins, in which case average molecular masses have to be used if the isotope peaks are unresolved.

      Electrospray has been used in conjunction with all common mass analyzers. One advantage of ES over PD and MALD (see below) is that as a consequence of the multi-charging phenomenon, the instrument can be calibrated in the low m/z range, using singly-charged calibrants with known exact masses. Major disadvantages are that spray formation is adversely affected even by moderate buffer and salt concentrations, and that mixtures of high mass samples can give overlapping charge state distributions that may be difficult to assign to individual components.

      Another atmospheric pressure ionization technique, termed ion spray, basically works in a manner similar to ES, but the ion spray formation occurs by a pneumatically assisted process at atmospheric pressure. Although both techniques give qualitatively the same results, ES and ion spray each appears to have some unique advantages in specific applications. Ion spray, for example, has been used more successfully to study non-covalent interactions, probably attributed to its ease of use with 100% aqueous solutions. This aspect is discussed in more detail below in the Applications section.

    10. Matrix-Assisted Laser Desorption (MALD)
    11. The use of lasers in mass spectrometry goes back to the early 1960's. A wide range of wavelengths, from UV to IR, have been used with many different types of mass spectrometers for isotope measurements, elemental composition analysis, and for pyrolysis of small organic and inorganic molecules.

      A major breakthrough came in 1988 with the introduction of MALD (13 ), a technique which now is able to detect biomolecules over 300,000 Da in size. The technique involves embedding the analyte in a solid matrix which absorbs energy at the wavelength of the laser. Common matrices are listed in Table I . The actual mechanism of MALD, a combination of desorption and ionization, is still being investigated. One hypothesis proposes a hydrodynamic model for the desorption of matrix and embedded biomolecules. The laser energy absorbed by the matrix, typically on the order of 106 watts/cm2, leads to intense heating and generation of a plume of ejected material that rapidly expands and undergoes cooling. The phase transition (evaporation and sublimation) is probably the rate determining step in ion formation. Generation of ions is believed to arise through ion/molecule reactions in the gas phase. However, depending on the matrix used, enough energy can be transferred to the molecule to break weak bonds. Generally, the [M+H]+ ion, or [M+Na]+, [M+K]+, etc., are preferentially formed in the positive ion mode, and [M-H]- ion in the negative ion mode. However, the technique also generates singly- and multiply-charged clusters of the analyte of low intensities, an undesirable situation in that these tend to complicate the spectrum. Mass resolution obtainable with MALD is highest when using laser power close to the threshold level required to produce ions from the solid sample, i.e., at low signal strength.

      MALD produces a relatively intense matrix background, generally below m/z 1000, that can be minimized electronically. This chemical background depends on the matrix and laser wavelength chosen. Calibration of the instrument can be accomplished either using internal standards of similar chemical behavior or by using an external calibration. A mass accuracy of ± 0.01% (± 1 Da at a molecular mass of 10,000) is the best that can be achieved under favorable conditions with today's commercially available instruments. In practice, other factors contribute to error in mass measurement accuracy. For example, many of the proteins used as internal standards for calibration have not themselves been rigorously characterized. Errors in their molecular masses can be caused by errors in sequence, chemical changes in side chains (oxidation/reduction), imbedded molecules from sample preparation such as phosphate and sulfate, or covalently-attached small molecules like acyl groups, metal cations, etc., and other changes due to sample work-up procedures. Nevertheless, MALD is a very fast and sensitive technique, implemented on small, relatively inexpensive instruments that do not require extensive expertise in mass spectrometry. Such instruments are ideally suited for biological scientists that need molecular mass information more quickly and more accurately than can be obtained by gel electrophoresis.

  3. Mass Analysis

    1. Sector Instruments

      The first mass spectrometers were built in the early 1900's. J.J. Thomson (14 ) used fixed magnetic and electric fields to separate ions of different mass and energy and was able to prove that the noble gas Ne was composed of two different isotopes of masses 20 Da and 22 Da providing the final proof of the atomic theory of matter. To distinguish between the two Ne isotopes, Thomson made use of the different behavior of charged particles of varying momentum and energy in an electromagnetic field.

      With acceleration out of the ion source, an ion acquires kinetic energy

      Ekin = zV acc = mv2 / 2,          (3)

      where Vacc represents the potential difference that defines the acceleration region, m is the mass of the ion and v its velocity. When entering a homogenous magnetic field on a trajectory perpendicular to the magnetic field strength B, the ion experiences the Lorentz force (F) that is perpendicular both to B and v. The resulting trajectory of the ion in the magnetic field is a circle with a radius r because the Lorentz force just balances the centripetal force,

      F = zvB = mv2 / r.                  (4)

      Thus, one obtains the relationship between the m/z value of an ion and the magnetic field strength

      m / z = r2B 2 / 2VAcc.              (5)

      On a magnetic sector instrument, r is a fixed value that is given by the geometry of the magnet and V acc is usually kept constant during the acquisition of a mass spectrum by scanning the magnetic field strength B in a reproducible manner. This implies that the magnetic field strength B has to be adjusted (scanned) in order to successively transmit ions of different m/z values through optical slits defined by the magnet radius r at constant Vacc (see Figure 7A ). Another effect of a magnetic sector field is focusing of a diverging ion beam i.e., it acts as an ion optical lens. For that reason a magnetic sector field is called a directional focusing device. However, ions of the same mass, but having different energies, will not be focused at the same point. Thus, to improve the focusing, i.e., improve the resolving power of the mass spectrometer, a device termed an energy analyzer is placed in the ion optic pathway, often before the magnet. Such an energy analyzer, also called an electrostatic analyzer (ESA), as shown in Figure 7B , commonly consists of two parallel cylindrical electrodes. The ESA results in overall directional focusing, but ions of different energy will still have different foci. However, if the ESA is designed so that the dispersion of the ions due to their velocity spread is exactly equal and opposite to that of the magnetic sector, the result of the combination is zero net velocity dispersion, i.e., ions of the same m/z but different velocities are focused at the same point. We can derive the equation for the ESA radius by equating the electrical force zE to the centripetal force:

      r = mv2 / zE = 2Vacc / E.        (6)

      The combination of an ESA and magnetic sector, then, is both direction (line A in Figure 7C ) and energy (line B in Figure 7C ) focusing, and is for that reason called double-focusing at the intercept of lines A and B.

      Such an apparatus is capable of a mass resolving power exceeding 100,000. The resolving power R is defined as

      R = m / Dm                                (7)

      where D m is the mass difference of two neighboring masses, m and (m + D m), of equal intensity with signal overlap of 10%. A resolving power of 100,000 allows one to clearly distinguish an ion of mass 100.000 Da from one of mass 100.001 da, or 100,000 Da from 100,001 da, i.e., to 10 parts per million. Such accurate mass measurement at low mass can be used to obtain the empirical formula of an unknown ion, by determining the compatible combinations of carbon, hydrogen, nitrogen and other atoms at the measured exact mass within the experimental uncertainty.

      Compared to an instrument that integrates the ion beam, i.e., records all of the ions all of the time, an instrument that scans the ion beam only records a fraction of the ions for each m/z generated. Thus, at any given time, only one ion species is focused on the detector. As a consequence, rapid scanning will result in poor ion statistics, and hence poor accuracy for mass determination. Furthermore, most magnets require a short settling time after each scan to ensure return to the same starting field strength. Selected ion monitoring is one way to increase sensitivity by increasing the time spent recording ions of interest. In this method, the acceleration and ESA voltages are stepped with a fixed magnetic field to permit a few specific ions to pass in sequence, rather than scanning the magnet over the entire range. However, the ions to be monitored must be known in advance. Increased sensitivity and speed, but loss of spectral information, are the consequences. Another sensitivity-increasing technique is the use of array detectors that can simultaneously record a certain percentage of the mass range. Typically, these collect ions within 5-20% of the mass range simultaneously, but these devices are exceedingly expensive. Also, to record the entire mass range, these must be stepped in consecutive small ranges.

      The necessary geometry and size of high resolution sector instruments depend on the mass range, sensitivity and resolving power that the analyst wishes to achieve. Some advantages, although not unique, are their relatively high mass range, sensitivity and resolving power, and their compatibility with a wide range of ionization techniques. The disadvantages are their size and cost compared to most other mass spectrometers.

    2. Quadrupole Mass Filter and Quadrupole Ion Trap Instruments
    3. The basic principles of the quadrupole mass filter were published in the early 1950's by Paul and Steinwedel (15 ). It has now become one of the most widely used types of mass spectrometers because of its ease of use, small size and relatively low cost. Mass separation in a quadrupole mass filter is based on achieving a stable trajectory for ions of specific m/z values in a hyperbolic electrostatic field. An idealized quadrupole mass spectrometer consists of four parallel hyperbolic rods. To one pair of diagonally opposite rods a potential is applied consisting of a DC voltage and an rf voltage. To the other pair of rods, a DC voltage of opposite polarity and an rf voltage with a 180° phase shift is applied. The potential f o applied to opposite pairs of rods (See Figure 8 ) is given by;

      ± f o = U + V cos wt              (8)

      where U is a DC voltage and Vcos w t, the time-dependant rf voltage in which V is the rf amplitude and w, the rf frequency.

      At given values of U, V and w , only certain ions will have stable trajectories through the quadrupole. The range of ions of different m/z values, capable of passing through the mass filter, depends on the ratio of U to V. All other ions will have trajectories which are unstable (i.e., they have large amplitudes in x- or y-direction) and will be lost. The equation of motion for a singly-charged particle can be expressed as a Mathieu equation from which one can define expressions for the Mathieu parameters au, and qu,

      au = ax = -ay = 4zU / mw 2r o2         q u = qx = -qy = 2zV / m w 2ro2            (9)

      with m/z, the mass-to-charge ratio of the ion, and ro, half the distance between two opposite rods. There is no z parameter, because the ac field only acts in the x/y-plane (z is the main axis of the linear quadrupole). Scanning the mass range on a quadrupole means changing the values U and V at a constant ratio a/q=2U/V, while keeping the rf frequency w fixed.

      The resolving power of a quadrupole mass filter also depends on the number of cycles experienced by an ion within the rf field, which in turn depends on its velocity. Thus, the resolution will increase with increasing mass, as ions of higher mass have lower velocity. However, the transmission efficiency will decrease, due to the longer time ions of higher masses spend in the quadrupole.

      One of the advantages of a quadrupole mass filter over a sector instrument is the low voltage applied to the ion source, i.e., the kinetic energy of the ions is on the order of 5-10 eV, compared with several keV for a sector instrument. This eliminates high voltage problems and makes interfacing to GC and LC easier. Other advantages are its good transmission efficiency, high scan speed, and wide acceptance angle to give high sensitivity.

      The quadrupole ion trap is based on the same principle as the quadrupole mass filter, except that the quadrupole field is generated within a three-dimensional trap. This trap consists of a ring electrode and two end caps as shown in Figure 8 . In the original design, the potential f o was applied to the ring electrode and - f o to the end caps (16 ). With this arrangement, ions were detected with resonance techniques. As is the case with quadrupole mass filters, the quadrupole field is closest to the theoretical ideal in the center of the trap. For this reason, a moderator gas like helium is often introduced into the trap in addition to the sample, to dampen the oscillations of the ions and hence concentrate them in the center of the trap. As indicated by the name, the ion trap can store ions over a long period of time making it possible to study gas phase reactions. In particular, the ion trap has excellent MS/MS capabilities (see below). The mass range of commercial instruments is 650 da, scanned at over 5000 da/sec. By reducing the scan speed to 0.015 Da/sec, a resolving power of 1.2x107 (full width at half maximum, FWHM) can be achieved. Also, under special conditions, i.e., resonance with external field to cause ejection, a mass range of up to 45,000 Da and sensitivities in the attomole range have been obtained. The very high sensitivity of the ion trap is a consequence of the fact that all ions formed can in theory be detected. However, space charge effects (ion-ion coulombic interactions) reduce the accuracy of mass assignment for an ion trap. Even though ion/molecule reactions take place within the trap, EI spectra generally compare well to EI spectra acquired on quadrupole mass filters. The applications potential of the ion trap is very great. Size, speed, sensitivity, MS/MS capabilities and compatibility with most ionization techniques favor further development of this mass analyzer.

    4. Time-of-Flight
    5. Time-of-flight (TOF) mass spectrometry was first successfully used as an ion analyzer in the 1950's ( 17 ). With the development of PDMS and later MALD MS, TOF analyzers have experienced a resurgence of use for instruments devised for the life sciences. The principal of mass analysis in a TOF analyzer is based on the principle that ions of different m/z values have the same energy, but different velocities, after acceleration out of the ion source. Thus, the time required for each to traverse the flight tube is different: high mass ions take longer to reach the detector than low mass ions. From equation (3 ), we can derive the expression;

      v = (2zVacc / m)1/2,          (10)

      t = (m / 2zVacc )1/2L,          (11)

      for the velocity of an ion of mass m/z and the time, t, spent in the drift region of length L. A schematic drawing of the TOF mass spectrometer set up with MALD ionization is given in Figure 9 . With an accelerating voltage of 20,000 V and a drift tube of 1 meter, a singly-charged ion of mass 500 Da will have a velocity of approximately 9x104 m/s and the time spent in the drift tube will be 1x10-5 s.

      An equation relating the flight time of an ion with its m/z value can also be derived;

      t = a (m/z)1/2 + b             (12)

      where a and b are constants for a given set of instrument conditions, and are determined experimentally from flight times of ions of known masses.

      Several ionization techniques are suitable for TOF mass analyzers by which ions are generated or ejected from the ion source over very short periods of time. This can be achieved with a laser pulse, 252Cf fission fragments, and introduction of ions from continuous ionization sources (EI, ES, FAB, etc.) with pulsed deflection or pulsed extraction. This pulse also gives the start signal for the data acquisition.

      From equation (11 ) it follows that m/z is proportional to t2, which leads to the formula for the resolving power;

      m / Dm = (1/2) (t / Dt)        (13)

      Today's linear TOF MS instruments are capable of attaining a resolution of 1 part per 1000. A factor affecting instrument resolving power is the time resolution of the detector and electronic circuitry, which with current technology is capable of handling a D t of a few nanoseconds. A major limitation in achieving high resolution is the consequence of the spread in time, space and kinetic energy of the initial ion packet. The time difference in formation of two ions of the same mass caused, for instance, by the length of a laser pulse, will remain the same during the flight to the detector. Increasing the flight time by reducing the acceleration voltage or increasing the length of the drift tube would lead to higher resolution since D t would remain the same but t become greater (see equation (13 )). Ions that are not formed at the same location, due to the spread of the sample on the target or width of the ion beam, will be accelerated to different kinetic energies. Also, ions can be generated with different initial kinetic energy. This spread in kinetic energy can be partially compensated by using a device called a reflectron as shown in Figure 10 . Ions of higher energy will penetrate deeper into an electrostatic repeller field (ion mirror) and will be turned around and arrive at the detector at the same time as ions of lower initial energy that penetrate the field less. In a linear TOF (no reflection), neutral and charged fragments generated through fragmentation of ions in the drift region cannot be distinguished from the original ion, because their velocity remains the same. Fragments of metastable ions generated prior to reflection have the same velocity as their parent ion, but a reduced energy and hence can be distinguished when spectra acquired in linear and reflectron mode are compared, due to their time difference. Reflectron TOF instruments are capable of a resolving power of over 10,000.

      TOF analyzers have some special advantages. In contrast to sector instruments and quadrupoles, all ions accelerated out of the ion source of the TOF instrument will reach the detector, giving it a relatively high sensitivity. Of course, the ions must be pulsed into the analyzer, usually at rates of 10-10,000 Hertz, depending on the particular instrument arrangement, and so, some dead time exists when ions are not being analyzed. Also, the mass range of TOF analyzers is virtually unlimited and any practical upper limit is dictated by the ionization process and by detector efficiency. The combinations of time-of-flight mass spectrometry with MALD, PD and ES have produced effective tools in the laboratory of biochemists, due to their relatively low cost, high sensitivity, speed and ease of operation.

    6. Ion Cyclotron Resonance Mass Spectrometry
    7. The ion cyclotron resonance (ICR) mass spectrometer is based on technology developed in the 1950's, Fourier transform techniques (FT-MS) (18 ) and the development of external ion sources ( 19 ,20 ). Like the ion trap, FT-MS is capable of storing ions within a cell. It consists of three pairs of parallel plates arranged as a cube, used for trapping, excitation or detection, respectively, as shown in Figure 11 . The cell lies within a strong magnetic field which is perpendicular to the trapping plates.

      A trapped ion, having an initial velocity, v, will, according to equation (4 ), travel in a circular path of radius r

      r = mv / zB                  (14)

      rotating with a frequency, w ,

      w = 2p / t = zB / m       (15)

      The frequency w is called the cyclotron frequency. The ion is kept within the cell by applying a dc-voltage to the trapping plates and can remain trapped for hours, provided the pressure is < 10-8 torr. According to equation (15) , the cyclotron frequency is inversely proportional to the mass of an ion. Because of the initial spatial distribution of ions, an excitation pulse is applied prior to detection, so that all ions with identical cyclotron frequency absorb energy and move together coherently. This then induces a current in the detector plates (image current) that is proportional to the numbers of ions and m/z values of the excited ions. The decay of the image current is mainly caused by collisional damping and occurs more rapidly at higher pressure in the cell. An outstanding feature of FT-MS is the extremely high resolving power that can be attained; for example, using electrospray ionization, a resolution of over 2x106 has been achieved. Another important feature of FT-MS is its MS/MS capability, which makes it an important tool for basic research in gas phase chemical research.

Back Up Next