More Ways to Make Ions

The ionization process in mass spectrometry (MS) creates ions from neutral molecules or extracts preformed ions from the sample. Additionally, the ionization source must facilitate the efficient transport of ions into the mass analyzer of the instrument for mass analysis. In the nearly 100 years of MS, and in 75 years of MS as a method of instrumental analysis, a great breadth and variety of ionization methods have been developed. The applications of these methods reflect the nature of the sample molecule (its volatility and thermal stability), the complexity of its environment (solid, liquid, or gas, and whether it is a mixture or purified), and the requisite flux and persistence of the ions created from the sample. The final four ionization methods in modern MS are electron ionization (EI) and chemical ionization (CI) for volatile samples, and electrospray ionization (ESI) and matrix-assisted laser desorption–ionization (MALDI) for nonvolatile samples. Of the many other distinct ionization methods available (as contrasted with commercial variants of the final four), some are no longer used, a few remain useful in specialized research, and several are used in successful commercial implementations of modern MS. In this installment, we describe the basics of a few "less often seen" ionization methods.

In the last two installments of this column, we discussed the basics of models that relate the extent of dissociations of isolated ions to their internal energies. The validity of these models is tested by comparison of the predicted results with experimental measurements. For example, an ionization method could create a molecular ion with different amounts of known internal energy, and then the extent of dissociation to fragment ions could be assessed as a function of the internal energy.

Charge-Exchange Ionization

One method used to create molecular ions of known internal energy is charge-exchange ionization. Tedder and Vidaud (1) provide an overview of this method. The basic charge-exchange reaction is

X⁺ + M → X + M⁺                                     [1]

where X represents the ionizing agent and M represents the sample molecule to be ionized. In the collision between these two species, the charge is transferred from the agent X to the molecule M. The collision should be at a low relative kinetic energy so that none of the kinetic energy is transformed into internal energy of the ion M⁺. The center-of-mass collision energy is usually held to less than 10 eV. The ionization potential of X is known. If it exceeds the ionization potential of M, then M⁺ is created with an internal energy that equals the difference between the ionization potentials.

Charge-exchange ionization can be used to create mass spectra that resemble electron ionization (EI) mass spectra (2) when the source conditions preclude the usual EI process. The selectivity inherent in charge exchange can be used as an advantage when only certain components of a mixture should be ionized. In these applications, the agent X is chosen so that some components are ionized and others are not, and therefore no separation is necessary. Praun and Villinger (3) describe such an application in the analysis of exhaled breath. An off-line analysis is most often used, but the system is simple enough that it also could be used on-line, although the cost of this dedicated analysis would be high. Because no separation is necessary, analysis is rapid, even when 114 different volatile compounds (distinguished by their mass) were monitored (samples included acetaldehyde, acetone, acetonitrile, ammonia, benzene, ethanol, methane, acetic acid, isoprene, and compounds directly related to administered drugs). Why would one want to monitor so many different compounds? Individual organ function in humans has been linked to approximately 60 different compounds found in varying concentrations in exhaled breath (4,5). The comprehensive and quantitative analysis provided by charge-exchange mass spectrometry (MS) may be useful for diagnostic and evaluation purposes.

Selection of the charge-exchange agent X provides the specificity described in the application described above. Theoretically, the choices are limitless, but in practice, simple species are used so there are not multiple ions that can react in a charge-exchange collision. Positive ions of the noble gases are often used, along with other diatomic and triatomic species. A few of these ionizing agents are listed in Table I along with their ionization potentials. If a given sample molecule M is ionized by a series of charge-exchange ionization agents, the appearance potentials of each of the fragment ions formed by dissociation of the molecule M can be determined. A plot of the intensities of these fragment ions as a function of the internal energy of M⁺ is called a breakdown curve.

Table I: Ionizing agents used in charge-exchange ionization and their ionization potentials. The agent will ionize sample molecules M with lower ionization potentials in low energy collisions; the difference in ionization potentials becomes the internal energy of M+. These data are drawn from reference 1, reprinting ionization potentials from a standard compilation.

Lindholm (6) described the creation of breakdown curves using charge-exchange MS and a two-stage mass spectrometer. The first mass spectrometer selects the ionization agent ion formed in the source, and is set to pass only this ion into a collision chamber where the charge-exchange ionization occurs. The second mass analyzer (set at 90° to the incident beam) measures the mass spectrum consisting of the fragment ions formed by dissociation of M⁺.

Chan and colleagues (7) recently described an ambient MS technique that they termed "desorption ionization by charge exchange." This method is related to desorption electrospray ionization (DESI) and several related methods that involve adding a dopant to the nebulized spray directed at the surface. Chan provides evidence that molecular ions formed in their procedure by a charge-exchange reaction between toluene ions (formed by oxidation that occurs in the electrospray itself, or present in the solution) and the neutral molecules M of the sample. The method was shown to produce ions successfully for samples that proved intractable with other ambient ionization methods. Research showed that charge exchange was probably only one of several ionization processes that were simultaneously occurring, and this is not surprising. Charge exchange should always be considered as a possibility in ion sources and elsewhere. We close this section with a quotation from a paper presented by Wade Fite (8) to the American Society of Mass Spectrometry in 1963:

Perhaps a little less obvious are the consequences of charge transfer research to more conventional mass spectrometry, however. It would seem clear that the analytical chemist should concern himself with charge transfer in the tube of his analyzer. Resonant charge transfer cross sections become large at typical mass spectrometer ion energies, and the danger of losing certain ions between the source and the detector is a very real one in some analytical mass spectrometers. Since the charge transfer cross section will depend strongly on the ion species in any given residual gas, the ion populations arriving at the detector may not be the same as those leaving the source, particularly if the vacuum in the analyzer is not the best.

Penning Ionization

Frans Michel Penning (1894–1953) was a Dutch physicist whose name is often encountered by mass spectrometrists. For example, the pressure inside the mass spectrometer may be measured with a type of cold cathode ionization gauge known as a Penning gauge (also a Phillips gauge). Later researchers used some of the concepts exemplified in the design of the gauge to develop the Penning trap, which uses electric potentials and a static magnetic field to trap ions in stable orbits for extended periods of time. Penning has a long list of U.S. patents, and his name is also attached to the "Penning mixture" used in neon signs. Penning's classic book Electrical Discharge in Gases was published in 1957 and showcased his expertise in the field. He was known as a talented and careful experimenter (9).

Penning ionization in MS often involves a metastable atom formed from a gas. For example, metastable gas atoms He*, Ne*, Ar*, and Kr* can be generated within the corona discharge at atmospheric pressure of these gases. As in charge exchange, a collision between the metastable ions and the gaseous sample molecule M results in ionization to form M^+.. The equation illustrates the ionization reaction for metastable He. The molecular ion formed is a radical cation, and the electron shown in the equation is that removed from the molecule; both masses and charges balance in the equation as shown.

He* + M → He + M^+. + electron                 [2]

The ionization process occurs as written because the ionization potential of the molecule M is lower than internal energy of the metastable ion. The rare gases are often used in Penning ionization for organic analysis because their excited states are of sufficient energy to ionize the sample molecules. Penning first described this process in 1927 (10), in a half-page journal article. In contrast to the precision with which molecular ions with various internal energies can be produced in charge-exchange ionization, Penning ionization can be a more complicated process energetically because there are many metastable states for the rare gas ions, and these states have different lifetimes. Additionally, different sources produce different populations of these various metastable states. Table II lists the most abundant metastable states for various gases commonly used in Penning ionization. Conveniently, the energy levels of these metastable species are grouped. Helium, neon, and argon should be able to ionize almost all organic molecules. The use of krypton or xenon may provide some selectivity. Depending on the metastable atom used, the M⁺ formed may have little to no internal energy, minimizing dissociation. However, the exact amount of internal energy in the molecular ion is not determined.

Table II: Energies of selected metastable ion states for some gases

We leave behind, for now, the wide breadth of Penning ionization processes in science, which includes surface analysis and modification, fundamental studies in spectroscopy, as well as molecular collision and dissociation dynamics (11). Instead, we describe how Penning ionization has provided yet another powerful method for the ionization of organic molecules. In 1993, metastable ion bombardment (MAB) was introduced (12) as a tunable source for organic analysis, designed with the intention of controlling the extent of ionization and the degree of dissociation of the ions formed by changing the identity of the metastable atom. Metastable ions could be created in a variety of sources; the first generation MAB source design uses a low-voltage corona discharge in a chamber filled with the chosen gas at a pressure of 10–100 mbar. Expansion through a skimmer nozzle creates a jet of ions, metastable atoms, and unreacted gas. Electrodes remove the charged species and what passes into the sampling region of the mass spectrometer are the gas and metastable atoms, with the latter involved in the Penning ionization process with gas-phase sample molecules M. The corona discharge produces a flux of metastable atoms of about 10¹⁴ –10¹⁵ atoms sr^-1 s^-1, depending on the exact source design (13). The metastable ions can be transported into other parts of a mass spectrometer source without regard to source potential or the presence of external magnetic fields.

Several intriguing applications of the MAB source have appeared in the literature. The MAB source has been fitted to a Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometer (14), and the tunability of the source demonstrated both with respect to selective ionization of one component of a mixture but not the other, and a variable degree of dissociation in the ionized samples. Dissociation of singly charged peptide ions stored in a radio frequency linear ion trap can be induced by collisions with metastable atoms (15); the neutral atoms are easily introduced into the trap because they are unaffected by the electric fields. Thermally vaporized aerosol particles have been analyzed by MS using metastable atom ionization, with the reduced dissociation thought to provide superior classification information for broad groups of compounds present in the aerosol (16). One of the challenging environmental applications of high resolution gas chromatography–electron ionization high resolution mass spectrometry (HRGC-EIHRMS) is the quantitative analysis of polychlorinated dioxins and furans (PCDDs and PCDFs), which are highly toxic contaminants usually found at trace levels. Common interferences in this analysis are the polychlorinated diphenyl ethers (PCDEs). Because the interference is caused by the isobaric fragment ions, MAB was used to produce only the molecular ions of the compounds, resulting in more-assured quantitative results (17).

In the examples provided above, Penning ionization is used to analyze samples that are in the gas phase at pressures concordant with the vacuum in the mass spectrometer. However, the samples can also be at atmospheric pressure. Samples and sample solutions are placed on a needle tip positioned near an aperture leading to the analyzer of the mass spectrometer. The needle is heated and a high voltage is applied to aid in the desorption of ions. The initial Penning ionization at the surface begins a cascade of subsequent ionization reactions, including protonation when the sample is dissolved in a matrix that can act as an acid. Tsuchiya and Kuwabara described this method in 1982 and developed it further over the next few years for the analysis of nonvolatile organic compounds (18,19). Recent developments use a similar source to create a sniffer to detect gases at atmospheric pressure (20).

In 2005, researchers described the direct analysis in real time (DART) ionization source (21). The basic source is well-known to scientists familiar with Penning ionization. Nitrogen or helium is introduced into a chamber, and a high voltage between cathode and anode creates ions and metastable atoms of the gas. The ions are removed from the effusive stream, and the metastable atoms are directed toward the mass spectrometer sampling orifice. Simplistically, sampling occurs when the liquid or solid sample is placed nearby, and the gas flow bounces off the surface into the mass spectrometer. The exact positioning of the sample is not critical; ions are collected from a large solid angle by virtue of the directed conductance through the aperture into the mass analyzer. The total distance from discharge chamber exit to mass spectrometer entrance is a few millimeters.

In a few situations, a classic Penning ionization mechanism is proposed to explain the ions formed in DART. Usually, a combination of ionization mechanisms is thought to occur, depending on the Penning gas used, and the presence of dopants and solvents. Positive ions include the radical molecular ions M^+., as formed in classic Penning ionization, as well as protonated molecules thought to form from the reactions of protonated water clusters with the sample molecule (the source of water is the air itself). Negative ions can be formed by electron attachment, with the population of thermal electrons produced by surface Penning ionization. All of these reactions are well-known and have been studied extensively. The key to the now extensive applications of DART has been in the ease of sampling, since materials can be analyzed directly on surfaces such as glass, paper, pills, skin, or foods, as well as liquids. It is, in retrospect, amazingly easy and direct to create the ions from such samples and capture at least some of them in the gas stream sucked into the mass spectrometer. The sensitivity, dynamic range, and accurate mass measurement capabilities of the instrument then support the identification of the sample (22). After 85 years, Penning ionization is alive and well, and it can be combined with other ionization methods to provide a flexible ionization probe for ambient MS.

References

(1) J.M. Tedder and P.H. Vidaud, J. Phys. D: Appl. Phys. 13, 1949–1956 (1980).

(2) E.D. Lee, S.-H. Hsu, and J.D. Henion, Anal. Chem. 60, 1990–1994 (1988).

(3) S. Praun and J. Villinger, Clinical Laboratory International, 23–25 (April 2002). http://www.alpha-mos.co.jp/separation/documents/airsense/vf_non-invasive.pdf.

(4) M. Phillips, Sci. Am. 267, 74–79 (1992).

(5) M. Phillips, J. Herrera, S. Krishnan, M. Zain, J. Greenberg, and R.C. Cataneo, J. Chromatogr. B 629, 75–88 (1999).

(6) E. Lindholm, in lon-Molecule Reactions, Volume 2, J.L. Franklin, Ed. (Butterworths, London, 1970), p. 457.

(7) C.-C. Chan, M.S. Bolgar, S.A. Miller, and A.B. Attygalle, J. Am. Soc. Mass Spectrom. 21, 1554–1560 (2010).

(8) W.L. Fite, "Charge Exchange Reactions," Presented at the 1963 Annual Conference of the American Society for Mass Spectrometry. http://www.dtic.mil/dtic/tr/fulltext/u2/409878.pdf.

(9) Penning's translated obituary is found here: http://www.arjenboogaard.nl/penning.html.

(10) F.M. Penning, Die Naturwissenschaften 15, 818 (1927).

(11) R.S. Berry, Radiat. Res. 59, 367–375 (1974).

(12) D. Faubert, J.G.C. Paul, J. Giroux, and M.J. Bertrand, Int. J. Mass Spectrom. Ion Phys. 124, 69–78 (1993).

(13) A.J. Palmer, M. Baker, and R.T. Sang, Rev. Scient. Instrum. 75, 5056–5058 (2004).

(14) C. Le Vot, C. Afonso, C. Beaugrand, and J.-C. Tabet, Int. J. Mass Spectrom. 306, 150–158 (2011).

(15) V.D. Berkout, Anal. Chem. 78, 3055–3061 (2006).

(16) C.B. Robinson, J.R. Kimmel, D.E. David, J.T. Jayne, A. Trimborn, D.R. Worsnop, and J.L. Jimenez, Int. J. Mass Spectrom. 303, 164–172 (2011).

(17) S. Moore, Chemosphere 49, 121–125 (2002).

(18) M. Tsuchiya and H. Kawabara, J. Mass Spectrom. Soc. Jpn. 30(4), 305–312 (1982).

(19) M. Tsuchiya and H. Kuwabara, Anal. Chem. 56, 14–19 (1984).

(20) T. Iwama, M. Hirose, I. Yazaw, H. Okoda, and K. Hiraoka, J. Mass Spectrom. Soc. Jpn. 54(4), 227–233 (2006).

(21) R.B. Cody, J.A. Laramée, and H.D. Durst, Anal. Chem. 77, 2297–2302 (2005).

(22) L. Nyadong, A.S. Galhena, and F.M. Fernández, Anal. Chem. 81, 7788–7794 (2009).

Kenneth L. Busch vaguely recalls something about "no new stories, but always new audiences." The reappearance of Penning ionization in MAB and DART applications is a reminder that maintaining a firm and broad foundation in science is fundamental to progress. We have forgotten more than we understand, and this is just as true for Sheldon as it is for Leonard. This column is the sole responsibility of the author, who can be reached at wyvernassoc@yahoo.com.

Kenneth L. Busch