Alternating current corona discharge ... - Wiley Online Library

Research Article Received: 16 July 2013

Revised: 20 September 2013

Accepted: 24 September 2013

Published online in Wiley Online Library

Rapid Commun. Mass Spectrom. 2013, 27, 2760–2766 (wileyonlinelibrary.com) DOI: 10.1002/rcm.6744

Alternating current corona discharge/atmospheric pressure chemical ionization for mass spectrometry Ahsan Habib1, Dilshadbek Usmanov1, Satoshi Ninomiya2, Lee Chuin Chen2 and Kenzo Hiraoka1* 1

Clean Energy Research Center, University of Yamanashi, 4-3-11 Takeda, Kofu, Yamanashi, 400-8511, Japan Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi, 4-3-11 Takeda, Kofu, Yamanashi, 400-8511, Japan

2

RATIONALE: Although alternating current (ac) corona discharge has been widely used in the fields of material science and technology, no reports have been published on its application to an atmospheric pressure chemical ionization (APCI) ion source. In this work, ac corona discharge for an APCI ion source has been examined for the first time. METHODS: The ambient atmospheric pressure ac corona discharge (15 kHz, 2.6 kVptp) was generated by using a stainless steel acupuncture needle. The generated ions were measured using an ion trap mass spectrometer. A comparative study on ac and direct current (dc) corona APCI ion sources was carried out using triacetone triperoxide and trinitrotoluene as test samples. RESULTS: The ac corona discharge gave ion signals as strong as dc corona discharge for both positive and negative ion modes. In addition, softer ionization was obtained with ac corona discharge than with dc corona discharge. The erosion of the needle tip induced by ac corona was less than that obtained with positive mode dc corona. CONCLUSIONS: A good ’yardstick’ for assessing ac corona is that it can be used for both positive and negative ion modes without changing the polarity of the high-voltage power supply. Thus, ac corona can be an alternative to conventional dc corona for APCI ion sources. Copyright © 2013 John Wiley & Sons, Ltd.

Atmospheric pressure point-to-plane direct current (dc) corona discharge has had a long history since its application to an atmospheric pressure chemical ionization (APCI) ion source for liquid chromatography/mass spectrometry (LC/MS) in 1974.[1,2] For example, Tsuchiya and Taira applied dc corona discharge to the ionization of less-volatile compounds in the liquid phase.[3,4] This was the first study that applied Penning ionization to atmospheric pressure mass spectrometry. In the mid-1980s, Sciex commercialized the explosive detection system that used temperature desorption coupled with a dc corona discharge APCI ion source.[5] Takada and coworkers[6–8] developed a point-to-plane dc corona APCI source equipped with counter-flow introduction for use in explosive detection systems in public transportation. This ion source dramatically improved the ionization efficiency for explosives compared with a normal-flow APCI source.[8] Singh et al.[9] used negative mode dc corona discharge for electron capture in APCI-MS and obtained attomole sensitivity for pentafluorobenzyl derivatives of biological samples. Because dc corona discharge is regarded as a ’point source’ for ion generation, it is suitable for application to microanalytical systems. One good example is the microchip-heated nebulizer for APCI-MS.[10] McEwen et al.[11,12] developed an atmospheric pressure solid analysis

2760

* Correspondence to: K. Hiraoka, Clean Energy Research Center, University of Yamanashi, 4-3-11 Takeda, Kofu, Yamanashi, 400-8511 Japan. E-mail: [email protected]

Rapid Commun. Mass Spectrom. 2013, 27, 2760–2766

probe (ASAP) to analyze volatile or semivolatile liquid or solid materials using a needle point source corona discharge. In 2005, Cody et al.[13] developed the direct analysis in real time (DART) method (patented in 2003) that uses a point-to-plane atmospheric pressure dc glow discharge. The introduction of DART triggered the use of ambient ionization mass spectrometry for solids, liquids, and gaseous samples.[14] Hiraoka et al.[15] studied the atmospheric pressure Penning ionization (APPeI) of aliphatic hydrocarbons with long-lived electronically excited rare gas (Rg) atoms (Rg*: He*, Ne*, Ar*, and Kr*). They found that the relative abundances of fragment ions increased in the order He* → Ne* → Ar* → Kr*, inverse to the internal energies of Rg*. Furthermore, they suggested that in He and Ar APPeI, the major factors for Penning ionization would be the rare gas excimers Rg2*.[16] Cooks and coworkers[17,18] developed desorption APCI (DAPCI) by inserting a metal needle into a plastic pipette tip. The ionized gas stream impinging on the sample surface desorbed/ionized the absorbed analytes deposited on the substrates. Galhena et al.[19] presented an ambient ion source, infrared laser ablation metastable-induced chemical ionization (IR-LAMICI). A portion of the ablated materials was ionized by DART. Ovchinnikova et al.[20,21] developed an atmospheric pressure proximal probe thermal desorption sampling method coupled with secondary ionization by electrospray or dc corona discharge. Sekimoto and Takayama measured the spatial distribution of OH–, NO–x and CO–x negative ions observed on point-to-plane electrode atmospheric pressure dc corona discharge.[22] They found that the NO–x and CO–x ions were dominant on the field lines arising

Copyright © 2013 John Wiley & Sons, Ltd.

Alternating current corona discharge/APCI-MS from the needle tip apex, whereas the field lines emanating from the tip peripheral regions with lower field strength resulted in the formation of the OH– ion. In contrast to the wealth of applications of dc corona discharge in APCI-MS as described above, we found no reports on alternating current (ac) corona discharge applied to the APCI ion source. On the other hand, ac corona discharge is a standard technique in material science and technology,[23,24] and this is due to the ease of operation, long lifetime and stability of the reactor. For example, low-current ac discharges between a point (or an edge) and a dielectriccoated plane is a standard practical technique for modifying dielectric surfaces.[25] The dc corona discharge is not applicable to such dielectric materials due to the build-up of surface static charging, resulting in discharge instability. In this report, an application of point-to-plane ac corona discharge to an APCI ion source is presented for the first time. The explosives triacetone triperoxide (TATP) and trinitrotoluene (TNT) were used as test samples for the positive and negative modes of operation, respectively. TNT is suitable for examining the softness of the ion source because the molecular ion M– readily decomposes to [M–H]– and [M–NO]– ions under severe experimental conditions. In the present work, ac corona was found to be as good as or even better than dc corona with respect to durability, reproducibility, and mild ionization. There was also no need to change the polarities of the high-voltage power supply for positive and negative modes of operation.

EXPERIMENTAL The ambient atmospheric pressure ac and dc corona discharges were generated by using a stainless steel acupuncture needle (outer diameter (o.d.): 0.12 mm, tip diameter: 700 nm, Seirin, Shizuoka, Japan). As shown in Fig. 1, the needle was positioned at 3 mm from the inlet of the LTQ XL ion trap mass spectrometer (Thermo Scientific, San Jose, CA, USA). The temperature of the ion transport tube was 120 °C. Auto gain



wileyonlinelibrary.com/journal/rcm

2761

Figure 1. The experimental setup of ac and dc corona discharge for APCI. An acupuncture needle (o.d.: 0.12 mm, tip diameter: 700 nm) was used for the needle electrode. The distance between the tip of the needle electrode and the mass spectrometer inlet was 3 mm: (a) open space system; (b) plastic (PFA) tube (length: 8 mm, i.d.: 2 mm, o.d.: 4 mm) system.

control was enabled with a maximum ion injection time of 2 ms, and the number of microscans was 1 (m/z range scanned: 50–500). Mass spectral data were acquired and processed by Xcalibur 2.1 software (Thermo Scientific) and the R open source statistical programming environment.[26] Figure 1 shows the two experimental approaches, the open space (Fig. 1(a)) and the plastic tube (Fig. 1(b)) system. The latter was used to investigate the effect of plasma confinement on ion intensity and fragmentation. We used a plastic perfluoroalkoxy (PFA) tube, 8 mm long, with an o.d. of 4 mm and an inner diameter (i.d.) of 2 mm. In both ac and dc corona discharge experiments, the applied voltages were adjusted to maintain a stable discharge with the values just above the threshold of the gas breakdown to avoid occurrence of arc or spark discharge. If the applied voltage becomes too high, transition from glow-like discharge to arclike discharge occurs with discharge currents higher than ~200 μA[27] where the needle tip and the counter electrode are bridged with an arc, i.e. a short circuit between the two electrodes. In the ac corona experiment, 15 kHz radiofrequency (RF) voltages with 2.6 kVptp (peak-to-peak) in the open system and 2.7 kVptp in the plastic tube system were applied to the needle. In the dc corona experiments, the discharge current was kept constant at 4 μA with applied voltages of +2.5 and – 1.5 kV in the open system, and +3.4 and –2.3 kV in the plastic tube system for the positive and negative modes of operation, respectively. To maintain the stable discharge, higher voltages were necessary for dc in the plastic tube system, whereas the applied ac voltages were nearly the same for ac discharge in the open and plastic tube systems. The needle was examined before and after a 20 h discharge experiment by scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDX) (FE-SEM (EDX), JSM-6500 F, JEOL, Akishima, Japan). TATP was synthesized in the laboratory employing the explosive handling procedures approved by the Japan Police Department. 2,4,6-TNT and 1,3,5-TNB as 1000 ppm solutions in acetonitrile were purchased from AccuStandard (New Haven, CT, USA). In the experiment with TATP, ammonia was first supplied by placing the vial containing 100 μL of 28% NH3 aqueous solution about 5 mm below the inlet of the mass spectrometer. A head-space gas formed from a few mg of solid TATP contained in a small vial was placed about 3 mm from the gap between the needle tip and the inlet of the mass spectrometer. Just before supplying the TATP vapor, the lid of the vial containing TATP was opened. This sample insertion method gave only qualitative information because the sample amounts introduced were not known and could vary for each operation. Acetonitrile of reagent grade was used without further purification. A 1 μL acetonitrile solution of TNT was supplied to the mass spectrometer inlet by using a gel-loading pipette tip (GELoader, Eppendorf, Germany) with o.d. and i.d. of 300 and 100 μm, respectively. The TNT solution (1 μL), loaded into the gel-loading tip by using a pipette, was placed on the peripheral surface of the mass spectrometer inlet. The expelled droplet vaporized instantly and ions produced by ac or dc discharge were sampled into the mass spectrometer. The air suction rate from the inlet to vacuum was about 1 L/min (i.e., 17 mL/s). With this suction rate, the sample containing acetonitrile vapor with a volume less than 1 mL should be effectively introduced into the mass spectrometer.

A. Habib et al. The head-space gas of solid TATP or acetonitrile solutions of TNT gave no memory effect because of the rather high vapor pressure (TATP: vapor pressure = ~6 Pa at 25 °C, melting point (m.p.) = 91 °C, boiling point (b.p.) = 97–160 °C; TNT: vapor pressure = 1.65 × 10–4 Pa at 25 °C, m.p. = 80.1 °C, b.p. = 240 °C). Thus, neither a TATP nor a TNT residual signal was detected. The vaporized sample was ionized either by electron attachment or by reagent ions generated by either ac or dc corona discharge.

RESULTS For the evaluation of the ac corona discharge as an APCI ion source, a comparative study of ac and dc corona discharges was conducted using TATP and TNT as test samples in the positive and negative mode of operation, respectively. In the open system (Fig. 1(a)), the plasma generated at the needle tip became diffusive along the line of radial electric force. In contrast, the plasmas in Fig. 1(b) were confined in the plastic tube system because of the charging of the inner wall of the insulator tube. The diffusive and confined plasmas gave different characteristics for the dc corona discharges. Positive mode APCI using ac and dc corona discharges Figure 2 shows the positive mode mass spectra for TATP measured in all four combinations of ac/dc and open/plastic tube systems. Although the mass spectra were contaminated with many background signals originating from laboratory air, ammoniated TATP, [TATP + NH4]+, was observed as the base peak in all cases, suggesting that ac corona (Figs. 1(b) and 1(d)) could be applied in the APCI ion source with performance similar to dc corona discharge (Figs. 1(a) and 1(c)).

Sigman et al.[28] analyzed TATP by gas chromatography (GC)/MS and GC/tandem mass spectrometry (MS/MS) by electron and chemical ionization and found that isolation of the [TATP + NH4]+ ion with subsequent collision-induced dissociation (CID) produced extremely low-abundance product ions with m/z values greater than 60, and the m/z 223 ion corresponding to [TATP + H]+ was not observed. They also calculated the binding energy of NH4/TATP to be 25 kcal/mol at the B3LYP/6-31G level, a result suggesting that CID of the rather weakly bound [TATP + NH4]+ ion would primarily dissociate the complex, forming TATP and the NH+4 ion. This agreed with the experimental finding that few product ions originating from TATP could be detected from the CID of the ion at m/z 240 [TATP + NH4]+. The strong appearance of the [TATP + NH4]+ ion in Figs. 2(a)–2(d) indicated that, in positive mode APCI, soft ionization takes place in all dc/ac and open/plastic systems. The static charging of the inner wall in the dc/plastic system causes the positive ions to be trapped in the central region of the plasma because the wall potential becomes higher (i.e. more positive) than the plasma potential. We expected that this ion trapping effect would increase the positive ion intensities. Contrary to our expectation, the signal intensity was found to be weaker in the dc/plastic than in the dc/open system, as shown in Figs. 2(a) and 2(c). The [TATP + NH4]+ ion in the dc/plastic system seems to have been decomposed in the plasma. A similar trend was observed in the negative mode of operation. Negative mode APCI using ac and dc corona discharges Figures 3(a) and 3(b) show the mass spectra measured in dc/open and ac/open systems for a 100 pg TNT sample in 1 μL acetonitrile solution. The strong [M NO] fragment ion at m/z 197 was the base peak; the [M H] fragment

2762

Figure 2. Positive mode mass spectra measured for head-space gas of solid TATP: (a) dc corona in open space (applied voltage: +2.5 kV dc); (b) ac corona in open space (applied voltage: 2.6 kVptp ac); (c) dc corona in plastic tube (applied voltage: +3.4 kV dc); and (d) ac corona in plastic tube (applied voltage: 2.7 kVptp ac).




Alternating current corona discharge/APCI-MS

Figure 3. Negative mode mass spectra measured for 100 pg TNT dissolved in 1 μL of acetonitrile: (a) dc corona in open space (dc applied voltage: 1.5 kV); (b) ac corona in open space (ac applied voltage: 2.6 kVptp); (c) dc corona in plastic tube (applied voltage: 2.3 kV dc); and (d) ac corona in plastic tube (applied voltage: 2.7 kVptp ac).


respectively. The standard deviation with five repeat measurements shown in the figure demonstrates the good reproducibility in the dc/plastic and ac/plastic systems.

DISCUSSION As described above, the ions undergo a more severe fragmentation in the dc/plastic than in the dc/open system for both positive and negative modes (see Figs. 2(a) and 2(c) and 3(a) and 3(c)). This must be caused by the static charging of the plastic tube as discussed by Akishev et al.[27] In the dc/ open system, the equipotential surface becomes convex at the needle tip, causing the plasma to be diffusive toward the radial direction. Due to the diffuseness, the primary ions generated in the plasma have less chance to undergo secondary decomposition in the plasma. In contrast, the equipotential surface at the needle tip in dc/plastic system becomes concave due to the wall static charging, resulting in the confinement of the plasma in a narrow space and the ions generated must thus undergo a more severe secondary fragmentation. Contrary to the dc corona, the ac corona plasma taking place in the dielectric tube must have some unique features compared with that in the open system. The insulating wall has a strong effect on the local field distortion caused by a surface charge accumulation on the dielectric.[27] The dielectric limits the discharge current due to charge accumulation on the dielectric surface. Thus, the initially formed plasma ceases before the transition from glow discharge to arc-like hot plasma. That is, the wall charging guarantees the occurrence of a nonequilibrium cold plasma. This may explain the better performance in the ac/plastic system shown in Fig. 3.



2763

ion at m/z 226 and the molecular ion M at m/z 227 were also observed in the dc/open system (Fig. 3(a)). In contrast, in the ac/open system, the molecular ion M was the base peak and the [M H] and [M NO] fragment ions are weaker (Fig. 3(b)), suggesting that the ac corona is a softer APCI ion source than dc corona APCI in the negative mode of operation. Figures 3(c) and 3(d) show the mass spectra in dc/plastic and ac/plastic systems, respectively. In the dc/plastic system, only the [M NO] fragment ion was observed, with no molecular ion M being recorded (Fig. 3(c)). In contrast, in the ac/plastic system the molecular ion M is the base peak with a much [M NO] weaker fragment ion (Fig. 3(d)). Thus, much softer ionization took place in the ac/plastic than in the dc/plastic system. To confirm the softer ionization for the ac/plastic than for the dc/plastic system, the mass spectra of 1,3,5trinitrobenzene (TNB) were also examined. As shown in Supplementary Fig. S1 (Supporting Information), a strong signal for [M–NO]– but no molecular ion M– was observed in the dc/plastic system, whereas M– was the major ion in the ac/plastic system. For the evaluation of the quantitation, the relationship between the sample amounts introduced and signal intensities was examined for TNT in dc/plastic and ac/plastic systems. Supplementary Figs. S2(a) and S2(b) (Supporting Information) show the calibration curves obtained for [M–NO]– and M– for the dc/plastic and ac/plastic systems, respectively. Good linearity in the range up to 2 ng is observed for both cases with a correlation coefficient R2 of ~0.99. The limits of detection (LODs) for TNT (signal-to-noise ratio: 3) were 50 and 30 pg in the dc/plastic and ac/plastic systems,

A. Habib et al.

Figure 4. SEM images for acupuncture needles: (a) before use; (b) ac corona discharge in open space for 20 h (applied voltage: 2.6 kVptp ac); (c) negative dc corona discharge for 20 h (applied voltage: 1.5 kV dc); and (d) positive dc corona discharge for 20 h (applied voltage: +2.5 kV dc at the start and +2.8 kV dc after 20 h).

2764

In all the dc mode experiments, the dc discharge current was kept constant at 4 μA. To maintain this current, the applied voltage for the negative mode ( 1.5 kV) was lower than that for the positive mode (+2.5 kV) in the open system. That is, a voltage lower than that of the positive mode is needed to maintain a stable plasma in the negative mode. This may be partly due to the supply of additional electrons by the photoelectron emission.[25] In the long-term negative mode experiment, a 4 μA current was maintained without changing the negative voltage ( 1.5 kV). On the other hand, in the positive dc mode, we had to increase the voltage to the needle tip from the initial value of +2.5 kV to higher voltages (up to ~ +2.8 kV) to maintain the constant discharge current for a long discharge time. After a continuous 20 h of operation for dc and ac discharge in the open system, the needle tip was examined for erosion using SEM and EDX. Figure 4(a) shows the SEM image of the needle before use. Figures 4(b)–4(d) show the SEM images for ac, negative and positive dc corona after 20 h of continuous discharge operation. In Figs. 4(b) and 4(c) some erosion is observed around the tip of the needle, but the original shape of the needle was reasonably maintained. In contrast, much severer erosion was observed in the positive dc corona, as shown in Fig. 4(d). The tip of the needle was eroded and some rugged product was formed at the tip. When a positive voltage was applied to a point electrode, the discharge was characterized by the streamer discharge, which extended to a much longer distance from the point towards the gap than for negative corona.[29] The streamer discharge exists in two modes: the


diffuse mode and the filamentary streamer mode. In the streamer mode, the bulk of the Joule energy is released to the surface of the electrode.[25] It is conceivable that the more severe erosion in the positive dc corona is due partly to the heat effect.[23] That is, for the positive dc corona, the streamer discharge leads to the melting of the needle tip at the filamentary discharge spot.[23] For the positive dc APCI ion source, the use of a thicker needle is recommended as is usually adopted in conventional dc corona APCI. Supplementary Fig. S3 (Supporting Information) shows the EDX image of the needle shown in Fig. 4(d). The signal intensity of the oxygen Kα peak of the eroded component is much stronger than that in the needle body. The formation of oxides at the needle tip is therefore obvious. Stainless steel is covered by a layer of Cr2O3 that prevents further oxidation of solid stainless steel in air, but the molten tip is readily exposed to further oxidation of both Fe and Cr in the alloy.

CONCLUSIONS In the present work, the point-to-plane ac corona discharge was evaluated for the APCI ion source using TATP and TNT as test compounds. For both the positive and negative modes the ac corona was found to give ion signals that were as strong as in conventional dc corona discharges. Corona discharges of both ac and dc type were examined with and without the plastic insulation tube. The static charging of



Alternating current corona discharge/APCI-MS the insulator wall in negative dc corona led to fragmentation of the primary product ions, but this effect was minor in ac corona. The ac/plastic system was found to be the best for the APCI ion source. The needle for the ac corona was less eroded than that for positive mode dc corona and was found to be suitable for long-term operation. Because the ac power source can be scaled down to a palm size, it would be suitable for coupling with a miniaturized mass spectrometer. The great merit of dc corona is that it can generate a basically monopolar positive or negative ion swarm. This makes it possible to apply dc corona, for example, to DAPCI, where charging of the insulation surfaces promotes ion desorption. This kind of application is not foreseen for ac corona because the plasma generated is bipolar. An investigation of the application of ac corona to APCI for a wide variety of less-volatile compounds including explosives such as 1,3,5-trinitro-1,3,5-triazacyclohexane (RDX), pentaerythritol tetranitrate (PETN), ammonium nitrate (AN) and hexamethylenetriperoxidediamine (HMTD) is now in progress.

[9]

[10]

[11]

[12]

[13]

[14]

Acknowledgements

[15]

The financial support for this work from the Japan Science and Technology Agency is gratefully acknowledged. We thank Dr Takada and Dr Sakairi for their illuminating discussions.

[16]

REFERENCES [17]


[18]

[19]

[20]

[21]

[22]

[23] [24]

[25]



2765

[1] D. I. Carroll, I. Dzidic, R. N. Stillwell, K. D. Haegele, E. C. Horning. Atmospheric pressure ionization-MS. Corona discharge ion source for use in a liquid chromatography-mass spectrometer-computer analytical system. Anal.Chem. 1974, 47, 2369. [2] I. Dzidic, R. N. Stillwell, D. I. Carroll, E. C. Horning. Comparison of positive ions formed in nickel-63 and corona discharge ion sources using nitrogen, argon, isobutene, ammonia and nitric oxide as reagents in atmospheric pressure ionization-MS. Anal. Chem. 1976, 48, 1763. [3] M. Tsuchiya, T. Taira. A new method of ionization for organic compounds in the liquid phase under atmospheric pressure. Int. J. Mass Spectrom. Ion Phys. 1980, 34, 351. [4] M. Tsuchiya, H. Kuwabara. Liquid ionization mass spectrometry of nonvolatile organic compounds. Anal. Chem. 1984, 56, 14. [5] G. J. Van Berkel, S. P. Pasilis, O. Ovchinnikova. Established and emerging atmospheric pressure surface sampling/ ionization techniques for mass spectrometry. J. Mass Spectrom. 2008, 43, 1161. [6] K. Kojima, M. Sakairi, Y. Takada, J. Nakamura. Vapor detection of TNT and RDX using atmospheric pressure chemical ionization mass spectrometry with counterflow introduction (CFI). J. Mass Spectrom. Soc. Jpn. 2000, 48, 360. [7] Y. Takada, H. Nagano, M. Suga, Y. Hashimoto, M. Yamada, M. Sakairi, K. Kusumoto, T. Ohta, J. Nakamura. Detection of military explosives by atmospheric pressure chemical ionization mass spectrometry with counter-flow introduction. Propellants, Explosives, Pyrotechniques 2002, 27, 224. [8] Y. Takada, H. Nagano、 Y. Suzuki, M. Sugiyama, E. Nakajima, Y. Hashimoto, M. Sakairi. High-throughput walkthrough detection portal for counter terrorism:

detection of triacetone triperoxide (TATP) vapor by atmospheric-pressure chemical ionization ion trap mass spectrometry. Rapid Commun. Mass Spectrom. 2011, 25, 2448. G. Singh, A. Gutierrez, K. Xu, I. B. Blair. Liquid chromatography/electron capture atmospheric pressure chemical ionization/mass spectrometry: Analysis of pentafluorobenzyl derivatives ob biomolecules and drugs in the attomole range. Anal. Chem. 2000, 72, 3007. P. Ostman, S. J. Marttila, T. Kotiaho, S. Franssila, R. Kostiainen. Microtip atmospheric pressure chemical ionization source for mass spectrometry. Anal. Chem. 2004, 76, 6659. C. N. McEwen, R. G. McKay, B. S. Larsen. Analysis of solids, liquids, and biological tissues using solid probe introduction at atmospheric pressure on commercial LC/MS instruments. Anal. Chem. 2005, 77, 7826. C. N. McEwen, R. G. McKay. A combination atmospheric pressure LC/MS:GC/MS ion source: advantages of dual AP-LC/MS:GC/MS instrumentation. J. Am. Soc. Mass Spectrom. 2005, 16, 1730. R. B. Cody, J. A. Laramee, H. D. Durst. Versatile new ion source for the analysis of materials in open air under ambient conditions. Anal. Chem. 2005, 77, 2297. G. A. Harris, A. S. Galhena, F. M. Fernandez. Ambient sampling/ionization mass spectrometry: applications and current trends. Anal. Chem. 2011, 83, 4508. K. Hiraoka, H. Furuya, S. Kambara, S. Suzuki, Y. Hashimoto, A. Takamizawa. Atmospheric-pressure Penning ionization of aliphatic hydrocarbons. Rapid Commun. Mass Spectrom. 2006, 20, 3213. H. Furuya, S. Kambara, K. Nishidate, S. Fujimaki, Y. Hashimoto, S. Suzuki, T. Iwama, K. Hiraoka. Quantitative aspects of atmospheric-pressure Penning ionization. J. Mass Spectrom. Soc. Jpn. 2010, 58, 211. Z. Takats, I. Cotte-Rodriguez, N. Talaty, H. Chen, R. G. Cooks. Direct, trace level detection of explosives on ambient surfaces by desorption electrospray ionization mass spectrometry. Chem. Commun. 2005, 1950. Y. Song, R. G. Cooks. Atmospheric pressure ion/molecule reactions for the selective detection of nitroaromatic explosives using acetonitrile and air as reagents. Rapid Commun. Mass Spectrom. 2006, 20, 3130. A. S. Galhena, G. A. Harris, L. Nyadong, K. K. Murray, F. M. Fernandez. Small molecule ambient mass spectrometry imaging by infrared laser ablation metastable-induced chemical ionization. Anal. Chem. 2010, 82, 2178. O. S. Ovchinnikova, G. J. Van Berkel. Thin-layer chromatography and mass spectrometry coupled using proximal probe thermal desorption with electrospray or atmospheric pressure chemical ionization. Raid Commun. Mass Spectrom. 2010, 24, 1721. O. S. Ovchinnikova, V. Kertesz, G. J. Van Berkel. Molecular surface sampling and chemical imaging using proximal probe thermal desorption/secondary ionization mass spectrometry. Anal. Chem. 2011, 83, 598. K. Sekimoto, M. Takayama. Negative ion formation and evolution in atmospheric pressure corona discharge between point-to-plane electrodes with arbitrary needle angle. Eur. Phys. J. D 2010, 60, 589. L. Bárdos, H. Baránková. Plasma processes at atmospheric and low pressures. Vacuum 2009, 83, 522. R. A. El-Koramy, A. Yehia, M. Omer. Development of ac corona discharge modes at atmospheric pressure. Phys. Plasmas 2011, 18, 023502. Y. S. Akishev, A. V. Dem’yanov, V. B. Karal’nik, A. E. Monich, N. I. Trushkin. Comparison of the ac barrier corona with dc positive and negative coronas and barrier discharge. Plasma Phys. Reports 2003, 29, 82.

A. Habib et al. [26] R. Ihaka, R. Gentleman. R: a language for data analysis and graphics. J. Comput. Graph. Stat. 1996, 5, 299. http://www.rproject.org/. [27] Y. S. Akishev, M. Grushin, I. Kochetov, V. Karal’nik, A. Napartovich, N. Trushkin. Negative corona, glow and spark discharges in ambient air and transitions between them. Plasma Sources Sci. Technol. 2005, 14, S18. [28] M. E. Sigman, C. D. Clark, R. Fidler, C. L. Geiger, C. Clausen. Analysis of triacetone triperoxide by gas chromatography/mass spectrometry and gas chromatography/tandem mass spectrometry by electron

and chemical ionization. Rapid Commun. Mass Spectrom. 2006, 20, 2851. [29] J. J. Lowke, R. Morrow. Theory of electric corona including the role of plasma chemistry. Pure Appl. Chem. 1994, 66, 1287.

SUPPORTING INFORMATION Additional supporting information may be found in the online version of this article at the publisher’s web site.

2766 wileyonlinelibrary.com/journal/rcm

