Introduction to Principles and Practice of Mass Spectrometry

1 History of Mass Spectrometry

According to the International Union of Pure and Applied Chemistry definition, 'Mass spectrometry is the branch of science dealing with all aspects of mass spectroscopes and the results obtained with these instruments.' It evolved from research in particle physics at the turn of the century, with Goldstein discovering positively charged 'rays' in 1886 and Wien (1898) studying their electric and magnetic properties. In the early 1900s J.J. Thomson built his 'parabola mass spectrograph' to measure the charge to mass ratio (z/m) for several ionic species. In the expression z/m, z is the charge number, i.e. the total charge on an ion divided by the elementary charge (e), and m is the nucleon number, i.e. the sum of the total number of protons and neutrons in an atom, molecule or ion. In modern mass spectrometry, the parameter measured is m/z, rather than z/m: the unit of m/z was recently designated the thomson (Th).

Aston continued the work at Cambridge and built instruments that helped him to establish the presence of isotopes. He was subsequently able to measure the atomic mass of most elements with sufficient accuracy to be able to calculate the 'packing fraction' of their atomic nuclei. The packing fraction is the difference between the accurate atomic mass of the isotope and the nearest whole number divided by the mass number, also known as the mass defect. Aston also obtained accurate measurements of the ratios of the stable isotopes of many of the known elements.

At the end of this exciting period of development, Aston was convinced that much of the potential of mass spectrometry had been exploited. It was not until the 1940s that the technique was put to work in elucidating organic structures in the petroleum industry. Ionisation was effected by electron 'impact' [now called electron ionisation (EI)] for those molecules that could withstand vaporisation into the heated and evacuated ion source without decomposition. This limited the practical mass range to less than 1000 daltons (Da) but yielded useful fragmentations for structure elucidation (see Chapter 2). By choosing to work with 70 electron volt (eV) electrons many ions were formed with internal energies far in excess of the ionisation energy (IE)]. These ions decompose rapidly to produce lower mass (fragment) ions and neutral radicals or molecules.

During the 1950s, commercial instruments were being built and new applications discovered. One of the earliest of these was the identification of low molecular weight volatile food flavour compounds. Ten years later, the powerful combination of electron ionisation mass spectrometry (EIMS) with gas chromatography (GC/MS) led to an explosion of applications where mass spectrometry was used in qualitative and quantitative, chemical and biochemical studies. GC/MS instruments produced enormous amounts of data, which were best handled by computers, and data acquisition and processing methods were devised. In 1966, Munson and Field described chemical ionisation (CI). This technique increased the yield of ions representative of the molecular weight of volatile molecules through interactions with reagent gas ions (e.g. CH+5 ions from methane) with little excess energy. Other 'soft' ionisation techniques such as field desorption (FD) and particle desorption methods based upon ion generation by Cf-252 fast fission products [plasma desorption, (PDMS)] were introduced during the 1970s for in volatile compounds. At the same time (and in response to these developments) the instrumental mass range was increased to cope with the larger sample molecule ions now entering the gas phase. This process accelerated in the 1980s with the introduction of Fast Atom Bombardment (FAB) ionisation. FAB was the first ionisation technique to enable biologists and biochemists routinely to obtain molecular weight information on complex, labile biomolecules, including polypeptides and small proteins.

Ionisation from the liquid state, followed by evaporation/desolvation of charged droplets, includes techniques such as ion spray, thermospray (TSP) and electrospray ionisation (ESI). These methods differed mainly in the manner in which ionisation was initiated. Multiply charged molecular ions could be formed under ESI, facilitating the measurement of high molecular masses, even on conventional instruments (i.e. those with a mass range up to 2000 or 4000 Th). More efficient pumping systems were required to cope with the increased gas volumes generated by vaporising liquids.

Separation techniques such as liquid chromatography and capillary electrophoresis coupled to mass spectrometry (LC/MS and CE/MS respectively) have extended the advantages first associated with the analysis of volatile compounds by GC/MS to compounds of low volatility and high molecular weight. Tandem mass spectrometry (MS/MS) collision-induced dissociation (CID), focal-plane array detectors, ion traps and hybrid instruments are providing a high sensitivity structure elucidation facility for involatile compounds similar to that provided by EIMS of volatiles. Recently, Laser Desorption (LD), and especially Matrix Assisted Laser Desorption Ionisation (MALDI), combined with Time-of-Flight (ToF) mass analysis has extended the practical mass range to over 300 000 Da, producing mainly singly-charged molecular ions. Fourier transform mass spectrometry (FTMS), also known as Fourier transform ion cyclotron resonance (FTICR) mass spectrometry is only slowly entering the commercial area.

Only a brief introduction to the principles and practice of mass spectrometry is given herein. The reader is directed towards more general textbooks, e.g. Throck Watson or Rose and Johnstone, for a more detailed discussion of the principles and practice of mass spectrometry. Beynon and Brenton lucidly introduce the physical aspects of mass spectrometry and ionisation in the gas phase. Finally, the latest volume by McLafferty and Turecek is recommended for a very thorough introduction to the interpretation of organic mass spectra.

Mass Spectrometry – a definition

Mass spectrometry is the study of systems causing the formation of gaseous ions, with or without fragmentation, which are then characterised by their mass to charge ratios (m/z) and relative abundances.

Mass spectrometry is unlike most other forms of spectroscopy or spectrometry that are concerned with non-destructive interactions between molecules and electromagnetic radiation. This is because mass spectrometry is the study of the effect of ionising energy on molecules. It depends upon chemical reactions in the gas phase in which sample molecules are consumed during the formation of ionic and neutral species. Although sample is consumed destructively by the mass spectrometer the technique is very sensitive and only trace amounts of material are used in the analysis. A mass spectrometer converts sample molecules into ions in the gas phase, separates them according to their mass to charge ratio (m/z) and sequentially records the individual ion current intensities at each mass – the mass spectrum. If these ion current intensities are drawn in histogram form taking the most intense ion current as 100%, the values of m/z versus percentage relative intensity (%RI) is called a line diagram, e.g. Figure 1.1.

2 Ionisation of Molecules

Several types of mass spectrometry can generate ions representative of the mass of the sample molecule. These are described below and, because ionisation is often intimately linked to sample introduction techniques, this topic is also discussed.

Production of Molecular Ions

If a quantity of energy equivalent to the IE of the molecule is supplied under EI conditions, a molecular ion is formed that is a radical ion denoted as M+ -. There are several ways of forming molecular ions.

Electron Ionisation (EI)

70 eV electrons passing through the ionisation chamber from the filament to the trap, under the control of the magnetic field, interact with volatilised sample molecules, which enter their path (Figure 1.2).

The temperature of volatilisation can be varied from ambient to > 400 °C in a vacuum of about 10-3 Pa. Interaction of the sample molecule with an energetic electron removes one (and sometimes two or more) electrons from the valence orbitals. During this process, an excess of energy can be transferred to the newly formed positively charged 'radical ion'.

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1.1)

Since the IEs of most organic molecules lie in the range 8–12 eY, the excess energy can cause bond dissociation (fragmentation) within the ion. The distribution of energy among the ions formed, and the ensuing pattern of fragmentation produced from different chemical structures, depends on several factors. Two of these, chemical bond lability and fragment ion stability, form a basis for the qualitative interpretation of mass spectra and are discussed in Chapter 2. The efficiency of the ionisation process in relation to the electron beam energy used is shown in Figure 1.3.

Most organic molecules are ionised by 8–12 eY and an additional 1-6 eV will dissociate any cleavable bonds. The choice of 70 eV electrons for conventional El mass spectrometry (a) ensures efficient transfer of enough excess energy to induce structurally informative fragmentation and (b) is in the plateau region of the figure, where it is easier to generate reproducible mass spectra. Although 70 eV is the ionisation energy of choice in EI, it is important to note that not all this energy is transferred to molecules during ionisation.

In principle, a spectrum free from fragmentation can be obtained by lowering the energy of the bombarding electrons to values close to the IE. However, the ionisation efficiency is also lowered by a factor of 200–300, so that large quantities of sample would be required for successful analysis. Some molecules with little excess energy can dissipate this by stabilisation among their degrees of vibrational freedom. These ions will be detected as molecular ions in the mass spectrum. Other ions will apportion the energy to cleave bonds to produce well-stabilised fragment ions. Many organic compounds decompose at the elevated temperatures required for vaporisation into the conventional EI source and others do not produce stable molecular ions under normal EI conditions. In these cases, alternative ionisation methods are available.

'In-beam' (Desorption) Electron Ionisation

For the improved ionisation of relatively involatile but stable molecules, a special probe is used to place the sample very close to the electron beam. This technique, commonly known as desorption EI (DEI), has been successful with a limited range of previously intractable biomolecules, for example cyclic peptides and some glycosides.

Chemical Derivatisation

The preparation of chemical derivatives will increase the volatility and (sometimes) reduce the IE of polar compounds, allowing EI to generate stable molecular ions. Organic acids, fatty acids, etc. can be analysed by EIMS of their methyl esters. Many chemical classes can be rendered volatile through the preparation of a variety of silylated products that enable mass and structural information to be obtained. Special chemical derivatives can be made that yield characteristic fragmentation properties, thereby aiding analysis. An example is the preparation of t-butyldimethylsilyl derivatives of steroids. These yield an abundant [M - 57] + ion (generated by loss of C4H9) that is especially useful in quantitative measurements.

The Particle Beam Interface

The particle beam interface (Figure 1.4) is a sample introduction rather than an ionisation technique in which the incoming liquid sample, e.g. HPLC eluent, is nebulised with helium gas to form an aerosol of solvent droplets.

The stream of liquid droplets is allowed to desolvate at ambient temperature in a chamber at the reduced pressure provided by vacuum pump 1. The remaining molecules then expand into the second evacuated region (provided by vacuum pump 2) through skimmer 1, where a supersonic jet (molecular beam) is formed. Skimmer 2 allows the molecular beam, containing heavy sample particles, to pass through into the ionisation chamber for subsequent electron, or chemical ionisation. The lighter helium and solvent particles are skimmed off and removed by vacuum pump 2. Although particle beam systems appear to be an ideal method for ionising involatile compounds, the technique lacks sensitivity and is unsuitable for polar, thermally labile molecules.

Production of Protonated Molecules and Adduct Ions

Methods that generate ions representative of the molecular weight of less volatile and thermally labile molecules are generally based on two main principles. Firstly, reaction and thermal equilibration with a reagent gas. Secondly, direct desorption from a liquid or solid matrix during or prior to ionisation. Under these conditions, ion/molecule reactions are generally responsible for the ionisation process.

Chemical Ionisation (CI)

In CI, chemical reactions occur between thermally equilibrated reagent ions and sample molecules. The conditions required are: a large excess (104:1) of reagent gas (R) to sample molecules (M), necessitating the use of higher energy electrons (500 eV) at the resultant source pressures in the range of 10–150 Pa. Primary ions are formed in the dense gas:

R + e- [right arrow] R+· (1.2)

These primary ions react with more reagent gas to form stable, reactive secondary cations (even-electron species) by collisional hydrogen transfer:

R+· + R [right arrow] + RH + (R - H)· (1.3)

If methane is the reagent gas, CH+5 (a Lewis acid) is produced. This ion reacts strongly with organic molecules (at around 20 Pa pressure) by proton or hydride transfer reactions, generating stable, protonated molecules. Other reagent ions formed during methane CI include C2H+ and C2H+3 although C3H+5 and C2H+5 are the predominant species.

Protonation

Protonation occurs when the proton affinity of the sample molecules is higher than that of the reagent gas:

RH+ + M [right arrow] R + MH+ (1.4)

Hydride Ion Abstraction (Dissociative Proton Transfer)

This is common for samples with lower proton affinities than the reagent gas, e.g. alkanes:

RH+ + C10 H22 [right arrow] C10H+21 + R + H2 (1.5)

Charge Exchange

Monoatomic reactant gas ions have no vibrational degrees of freedom and therefore the ionisation energy, e.g. 15.755 eV for argon, is all transferred to the colliding molecule; a useful property for energetic and kinetic studies. Charge exchange can also be used selectively to ionise particular compound classes.

R+· + M [right arrow R + M+· (1.6)

Other popular CI reagent gases include isobutane, which yields mainly CH, and ammonia, which yields NH+4 ions. Depending on the acidity of the reactant secondary ion and the basicity of the sample molecule, adduct ions (electrophilic addition) can be formed

M + C2H+5 [right arrow] [M + C2H5]+ (1.7)

M + NH+4 [right arrow] [M + NH4]+ (1.8)

Negative Ion Chemical Ionisation

Various types of reaction can take place according to the nature of the reagent gas used, e.g. a mixture of hydrocarbon and water (95:5) can produce negatively-charged sample ions by the following reactions:

H2O + e- [right arrow] OH- + H (1.9)

AH + OH- [right arrow] A- + H2O (1.10)

M + HO- [right arrow] MOH- (1.11)

Other possible reagent gases include hydrocarbon/organic halide/oxygen (which produces Cl- and O- attachment ions) and fluorocarbons yielding negative ions by hydride abstraction and fluorine attachment. Negative ions can also be formed from suitable sample molecules by electron capture processes. In this case, the reagent gas acts as a moderator, generating thermal electrons that can then attach to molecules with high electron affinities forming negative radical ions.

M + e- (thermal) [right arrow] M-· (1.12)

Negative ion CI, especially electron capture ionisation, can be two to three orders of magnitude more sensitive than positive ion CI for electronegative molecules. It is therefore especially useful in the quantitative determination of trace substances that have, or that through the production of suitable derivatives are induced to have, electron-capturing properties.