super-atmospheric pressure chemical ionization mass spectrometry

Research article

Received: 13 November 2012 Revised: 9 January 2013 Accepted: 17 January 2013 Published online in Wiley Online Library

(wileyonlinelibrary.com) DOI 10.1002/jms.3173

392

Super-atmospheric pressure chemicalionization mass spectrometryLee Chuin Chen,a* Md. Matiur Rahmanb and Kenzo Hiraokab

Super-atmospheric pressure chemical ionization (APCI) mass spectrometry was performed using a commercial mass spectrometerby pressurizing the ion source with compressed air up to 7 atm. Similar to typical APCI source, reactant ions in the experiment
were generated with corona discharge using a needle electrode. Although a higher needle potential was necessary to initiatethe corona discharge, discharge current and detected ion signal were stable at all tested pressures. A Roots booster pump withvariable pumping speed was installed between the evacuation port of the mass spectrometer and the original rough pumps tomaintain a same pressure in the first pumping stage of the mass spectrometer regardless of ion source pressure. Measurementof gaseous methamphetamine and research department explosive showed an increase in ion intensity with the ion sourcepressure until an optimum pressure at around 4–5 atm. Beyond 5 atm, the ion intensity decreased with further increase ofpressure, likely due to greater ion losses inside the ion transport capillary. For benzene, it was found that besides molecularion and protonated species, ion due to [M+2H]+ which was not so common in APCI, was also observed with high ion abundanceunder super-atmospheric pressure condition. Copyright © 2013 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this article.

Keywords: high pressure ion source; atmospheric pressure chemical ionization; corona discharge; gaseous ionization; benzene;explosive

* Correspondence to: L. C. Chen, Interdisciplinary Graduate School of Medicineand Engineering, University of Yamanashi, 4-3-11 Takeda, Kofu, Yamanashi,400–8511 Japan. E-mail: [email protected]

a Interdisciplinary Graduate School of Medicine and Engineering, University ofYamanashi, 4-3-11 Takeda, Kofu, Yamanashi, 400-8511, Japan

b Clean Energy Research Center, University of Yamanashi, 4-3-11 Takeda, Kofu,Yamanashi, 400-8511, Japan

Introduction

Atmospheric pressure chemical ionization (APCI) source is widelyused as a complementary ion source in most electrospray ionization(ESI)-based atmospheric pressure ionization (API)mass spectrometerfor the detection of non-polar and gaseous compounds.[1,2]

Although APCI is mostly used in liquid chromatography massspectrometry (MS), it had also been adopted in the MS-based airmonitoring and explosive detection system with near real-timeresponse.[3,4] As indicated by its name, the CI is done underatmospheric pressure, and the gas pressure is several orders highercompared to that of CI in vacuum.[5] CI under atmospheric pressurewas softer than typical CI, and the reactant ions such as hydroniumion H3O

+ can be supplied by the water or solvent vapors containingambient gas. Besides corona discharge, reactant ions can also beproduced by other discharge sources such as atmospheric pressureglow discharge,[6] dielectric barrier discharge[7] or by Penningionization with metastable helium atoms.[8–12]

A key factor that promotes the widespread use of APCI is theemergence of efficient differential vacuum pumping system andhigh-performance ion guide, which are developed originally forESI for the transportation of ions from the atmospheric pressureregion to the high vacuumone. The designs of the ion transmissionsystem in most API mass spectrometer are based on the upstreampressure of 1 atm and the down stream pressure of 1–2 Torr inthe first vacuum stage.[13] Using modified ion transport capillaries,we have reported a series of super-atmospheric pressure ionsources implemented with conventional electrospray,[14,15] probeelectrospray[16] and field desorption (FD).[17] In those ion sources,the high pressure environment where the ion emitter was locatedwas primarily used to prevent the corona or arc discharge thatcould disrupt the proper operation of ESI and FD, and partially forthe enhancement of ion desolvation.

J. Mass Spectrom. 2013, 48, 392–398

Owing to higher gas density and higher collision frequency forreactant ions and analytes, the performance of gaseous ionizationtechnique such as APCI can potentially be enhanced by increasingthe ion source pressure. Although there has been little report onthe super-atmospheric pressure CI-MS, there do exist some earlierreports that suggest an enhancement of gas phase ionizationefficiency under higher operation pressure. For example in the highpressure ion mobility spectrometry (IMS) where the high pressurewas used to enhance the resolution of IMS, the ion signal forthe analytes ionized by 63Ni source was found to be maximum atan optimum pressure above 1 atm.[18] Also, with ion mobilityspectrometer, rate constants for certain gas phase ion moleculereactions were found to be higher with increased buffer gaspressure.[19,20] A recent work with femto-second laser ionizationsource for gaseous compound also showed a better performancewith helium carrier gas pressurized to> 1 atm.[21] Some theoreticaltreatment has been worked out by Juan Fernandez de la Moraon the gaseous ionization efficiency of secondary electrospray(which is also applicable to unipolar corona discharge), and it wasalso proposed that ‘super-atmospheric operation would be morepreferable in space-charge-limited situations’.[22]

In this paper, we performed the super-atmospheric pressureCI-MS using a typical corona discharge. It may sound contrary

Copyright © 2013 John Wiley & Sons, Ltd.

Super-atmospheric pressure CI-MS

to our previous work in which the high pressure gas was used toquench the corona discharge, but if sufficient high potential wasapplied to the discharge needle, it was still possible to obtain astable corona discharge. High pressure corona discharge sourcewas not new, and it had been tested to >8 bar[23] and used inthe molecular beam experiment for beam excitation.[24]

Experimental

Super-atmospheric pressure chemical ion source

The experiment was performed using commercial linear ion trapmass spectrometers (LTQ-Velos, Thermo Fisher Scientific, San Jose,CA). The schematic showing the coupling of the high pressure CI tothe mass spectrometer is shown in Fig. 1. Photographs of theexperimental setup are shown in Supplementary Figure S1. Theoriginal ion transport capillary (with 0.55 mm i.d.) of the LTQ-Veloswas replaced with a custom-made stainless steel capillary with0.5 mm i.d. The total length of the modified capillary was slightlylonger than the original one, and it protruded into the ionizationchamber to sample the gaseous ions. The distance betweenthe needle electrode and the inlet of the ion transport capillarywas ~4 mm.

The high pressure ionization chamber was made of aluminum,and it had two threaded ports functioned as inlet and outlet forthe sample gas. The body of the ion source and gas tubing werewrapped with heater during the operation. Temperatures for theion source body and the ion transport capillary were set at 150�C. The original corona discharge needle used in the LTQ-VelosAPCI source was mounted to an insulating flange made of

Heater

LTQ-Velos

1.5 Torr

Needle Valve

To Exhaust

Sample Gas

PEEK

Corona DischargeNeedle

HV

Ion TransportCapillary (0.5 mm i.d.)

1~7 atm

R

Figure 1. Schematic of the super-atmospheric pressure chemical ionizationsource. The distance between the needle electrode and the ion samplinginlet was ~ 4 mm. The ion source chamber was pressurized to> 1 atmwith compressed air, and the mass flow rate of the gas was maintainedat 23.2 g/min (which was equivalent to 20 l/min under 1 atm at roomtemperature) with a needle valve. The pressure in the first pumping stageof the LTQ-Velos was maintained at 1.5 Torr regardless of ion sourcepressure by varying the pumping speed of the additional booster pump.The resistance R is 50 MΩ.

J. Mass Spectrom. 2013, 48, 392–398 Copyright © 2013 John W

polyether ether ketone. The corona discharge was operated at aconstant current mode of 4 mA, and the high potential appliedto the needle electrode in positive mode was ~3kV at 1 atmand ~6.3 kV at 6 atm. A 50 MΩ resistor was connected betweenthe needle electrode and the output of the high voltage powersupply (HCZE, Matsusada, Shiga, Japan) for the safety of the ionsource operator. This resistor also functioned as a ballast toprevent the occurrence of harsh arc discharge.

The ion source chamber was pressurized with dry air suppliedfrom an air compressor (oil-free scroll compressor, Anest Iwata,Yokohama, Japan). The air was dried with membrane air dryerand filtered by odor filter (activated carbon filter). The relativehumidity of the dry air under atmospheric pressure was less than10%. The pressure of the ion source was monitored by an analogpressure gauge, and the pressure and gas flow rate wereadjusted using a pressure regulator and a needle valve. The flowrate of the sample gas measured by mass flow meter (Azbil,Tokyo, Japan) was 23.2 g/min (which was equivalent to 20 l/minunder 1 atm at room temperature of 27 �C). Carrier gas containinggaseous analyte was flowed towards the ion sampling inlet toreact with the plasma, and the gas that was not be sampled bythe mass spectrometer was directed out to the exhaust.

In order not to affect the ion transmission in the first andsubsequent vacuum stage, the pressure in the first pumpingstage of the mass spectrometer was maintained at 1.5 Torr whenthe ion source pressure was varied. It was achieved by having anadditional Roots booster pump (PMB-001C, Ulvac, Kanagawa,Japan) with variable pumping speed installed in between theevacuation port of the LTQ and the original rotary pumps. Withthis pumping system, it was possible to operate the ion sourceat pressure up to 7 atm using the ion transport capillary with0.5 mm i.d., which was about the same size with the originalion transport capillary. The three-phase AC motor of the boosterpump was controlled by a variable frequency controller (VF-AS1,Toshiba Schneider Inverter, Japan), and the pumping speed inthe first pumping stage of the mass spectrometer could beadjusted from 500 to ~3160 l/min. Supplementary Figures S2a,S2b and S2c show the pumping speed of the Roots pump, intakerate of the gas sampled by the mass spectrometer and theestimation of Reynolds number for the gas flow inside the iontransport capillary.

Sample preparation

All solvents were of reagent grade. Research department explosive(RDX) was purchased from Ultra Scientific (North Kingstown, RI),and themethamphetamine hydrochloride salt was fromDainipponSumitomo Pharma (Osaka, Japan). RDX and methamphetaminewere prepared in acetonitrile and methanol, respectively. Thesample solution in the concentration of 0.01 mg/ml was dispersedinto a heated tube which was connected to the ion source using ahigh linear force syringe pump (PHD 4400, Harvard Apparatus, MA)at a solution flow rate of 10 ml/min. The liquidwas rapidly evaporatedby the heated air flowed at 23.2 g/min (~0.8 mol/min), and thegas phase concentrations for RDX and methamphetamine were0.56 ppb and 0.84 ppb, respectively, in molar fraction.

393

Results and discussion

Under super-atmospheric pressure condition, the high potentialapplied to the needle was higher compared to that under 1

iley & Sons, Ltd. wileyonlinelibrary.com/journal/jms

L. C. Chen, M. M. Rahman and K. Hiraoka

394

atm to maintain a same discharge current. However, the coronadischarge was still in the unipolar regime (i.e. only positive ioncan be detected if the corona needle was positively biased andvice versa), and the ionization process was likely the same as thatof standard APCI. In the case of positive mode, hydronium ions orsolvent related ions produced by a series of reaction in thecorona plasma are responsible for the protonation of analytesand other background impurities which have proton affinityhigher than that of water or solvent molecule.Figure 2 shows the changes of background ion intensities with

the increase of ion source pressure in positive and negative ionmodes. During this measurement, the parameters for the LTQmass spectrometer (especially the ion guide potential) were setfor the detection of small m/z ions. In the positive mode, thedominant peak was due to the protonated acetone, and the peakintensity decreased with the rise of ion source pressure. Thesource of acetone was from the ambient air of the laboratorywhere organic solvents were kept. The ion signal due to thewater dimer was much weaker than that of acetone, but theintensity was about the same at all pressures. Larger waterclusters were not observed even when the ion source waspressurized to 7 atm, owing to sufficient heating of the iontransport capillary.In the negative ion mode, the most abundant ion signal was

originated from CO3�, and this observation was quite common

a)

b)

Figure 2. Intensities for background ions detected in a) positive and b)negative ion modes using corona discharge chemical ionization sourceoperated under different air pressure. The mass flow rate of the air waskept at 23.2 g/min at all pressures.

wileyonlinelibrary.com/journal/jms Copyright © 2013 Jo

in negative corona ionization using dry air.[25] The peak intensityof CO3

�first increased with pressure to 2 atm, and after that it

decayed gradually with further increase of pressure. On the otherhand, the intensity for NO3

� was much lower at 1 atm but it rosesignificantly with pressure. NO3

� could be formed by the electrontransfer from O2

� to the NOx radicals that were generated from aseries of reactions between oxygen and nitrogen-relatedmolecules,[26] and it was expected that the occurrence of suchgas phase reactions were more frequent under higher ion sourcepressure. In some cases, the presence of NOx was not desirable asthey could compete with the analyte for electron, and someattempt like reverse-flow corona discharge ion source have beenused to reduce their abundance.[27] Such reverse-flow techniquewas also found to work in the current source design by reversingthe gas flow direction, and the abundance of NO3

� was found tobe reduced by ~ 1/5. Nevertheless, all measurements shown inthe following section employed only the standard forwardflowing method. The presence of NO3

� could, in certain casessuch as RDX, be useful for the formation of [M+NO3]

� adducts.The analytical performance of the present ion source was first

evaluated with methamphetamine, an illicit drug which is widelyabused worldwide, in the positive ion mode. The detection ofthis compound is well studied with mass spectrometric method,and typical APCI-MS produced protonated species and itscharacteristic fragments at m/z 91 and 119.[28] In our measurement,themethanolic solution containingmethamphetamine hydrochloricsalt was vaporized inside a heated tube, and the gaseous samplewas carried by the compressed air to the ionization chamber toreact with the corona plasma. To have a fair comparison forthe detection sensitivities under different pressures, the massflow rate of the compressed air, the gaseous concentration ofmethamphetamine and the pressure in the first pumping stageof the mass spectrometer were kept constant when the gaspressure of the ion source was varied.

Figure 3 show the detection of methamphetamine (0.8 ppb)from the analyte containing gas under different ion sourcepressures. Figure 3a shows the plot of total ion current (TIC)versus pressures. The change of ion intensities for the protonatedmethamphetamine and the fragment at m/z 91 versus ion sourcepressure is shown in Fig. 3b. The ion intensities for protonatedmethamphetamine and its fragment first increased steadily withthe pressure until an optimum value at 4 atm, and after that, itdropped with the further increase of pressure. The CI massspectra taken at 1 atm and 4 atm are shown in Figs. 3c and 3d,respectively. In Fig. 3c, peaks at m/z 59 and 65 were due to theprotonated acetone and methanol dimer. Other peaks like thoseat m/z 198, 225 and 251 were originated from the nylon tubingand the air compressor. The change of these peaks with the ionsource pressures is depicted in Supplementary Figure S3, andthe background mass spectra for the blank sample takenunder 1 atm and 4 atm are shown in Supplementary Figure S4.Methamphetamine has a high proton affinity, and the protonatedspecies and its fragment could be easily detected with APCI withgood S/N at ppb level under 1 atm. Nevertheless, an enhancementof more than five times in ion signal could be observed when theion source was optimized at 4 atm.

Figure 4 shows the result of RDX (1,3,5-Trinitroperhydro-1,3,5-triazine), a military explosive in the negative ion mode. RDXappeared in the mass spectra as [RDX+NO2]

� and [RDX+NO3]�,

which are typical characteristic peaks for RDX in APCI-MS. Thechange of TIC with ion source pressure is depicted in Fig. 4a,and the plot of ion intensities for [RDX+NO2]

� and [RDX+NO3]�

hn Wiley & Sons, Ltd. J. Mass Spectrom. 2013, 48, 392–398

a)

b)

d) 4 atm

c) 1 atm

Figure 3. Detection of methamphetamine by atmospheric pressure andsuper-atmospheric pressure chemical ionization under different gaspressures. a) Plot of total ion current (TIC) versus pressures. b) Normalizedion intensities for the protonated methamphetamine ([M+H]+) and afragment at m/z 91 versus gas pressures. c) and d): Mass spectraacquired at 1 atm (~ 1 bar) and 4 atm. The gas phase concentration ofmethamphetamine was 0.84 ppb.

a) TIC

b)

d)

c)

Figure 4. Detection of RDX by atmospheric pressure and super-atmospheric pressure chemical ionization under different gas pressures. a)Total ion current (TIC) versus pressures. b) Normalized ion intensities for[M+NO3]

� and [M+NO2]� versus gas pressure. c) and d): Mass spectra

of RDX acquired at 1 atm and 5 atm. The gaseous concentration ofthe RDX in the flowing gas was 0.56 ppb. Asterisk denotes the locationof m/z 284 for [M+NO3]

�.


395

versus pressure is shown in Fig. 4b. The change of otherbackground peaks (e.g. CO3

�, NO3� and those from impurities)

with ion source pressure is shown in Supplementary Figure S5.Similar to that of methamphetamine measurement, the ionintensities associated with RDX first increased with the pressureuntil ~5 atm and then decreased with further increase of gaspressure. The corresponding mass spectra acquired at 1 atmand 5 atm ion source pressures are shown, respectively, in Figs. 4cand 4d. Background peaks such as those at m/z 205 and 221 areoriginated from the compressed air. The gas phase concentrationof RDX supplied to the ion source for all pressures was 0.56 ppb.At this concentration, the RDX was barely detected from theflowing gas at 1 atm, but the peak of [RDX+NO3]

� could be wellobserved at 5 atm with more than 10� signal enhancement. Theion abundance for [RDX+NO2]

� at m/z 268 was lower comparedto that of [RDX+NO3]

�, but it had the same response with theion source pressure. Besides the background ions, the source ofNO2 and NO3 for [RDX+NO2]

� and [RDX+NO3]� could also be

originated from the RDX itself.[29]

The detection of benzene under different ion source pressuresusing compressed air as working gas is shown in Fig. 5. Theionization potential for benzene is 9.2 eV, and it can be ionized


by active species such as N2•+, and O2

•+, produced from the coronaplasma. The protonation of benzene by hydronium ion is alsopossible since the proton affinity of benzene is higher than thatof water. Although the detection of protonated and adduct ionswas found to be more preferable under higher ion sourcepressure, for the aromatic compound like benzene, theenhancement was not as profound for the detection of molecularion, M+. The abundance of benzene molecular ions even becamelower with the appearance of complex ion species when the airpressure was increased. At 1 atm, the APCI spectrum in Fig. 5ashows the molecular ions, M+ as the major peak, together withprotonated species, [M+H]+ which was of lower abundance.A peak at m/z 94, which is due to [M+O]+, was also observed.When the air pressure was increased, e.g. to 3 atm, owing tothe increase of collision frequency between benzene moleculesand the water and solvent related ions, the protonated speciesbecome more abundant than the molecular ions (Fig. 5b). Inter-estingly, besides the protonated benzene, an adjacent peak of[M+2H]+ also appeared at 3 atm, and its intensity also increasedwith gas pressure and overtook the protonated benzene tobecome the major peak when the ion source pressure reached6 atm (Fig. 5c). Similar trend was also found for toluene. Other


a) 1 atm

b) 3 atm

c) 4 atm

d) 6 atm

Figure 5. Chemical ionization mass spectra of benzene acquired by thepresent corona discharge ion source at air pressure of a) 1 atm, b) 3 atm,c) 4 atm and d) 6 atm.


396

peaks that appeared under super-atmospheric pressure were m/z97 and 110, which were likely due to C6H9O

+ and C6H6O2+. The

observation of [M+ 2H]+ for benzene has been reported by Naet al. using dielectric discharge barrier source with helium asthe carrier gas[30] and that species was attributed to the di-hydrogenation of benzene. In their experiment, the surface ofthe insulating discharge tube was negatively charged due tothe diffusion of electron from the plasma,[31] and they hadproposed that these low binding energy electrons and the presenceof Si–OH function group on the surface were responsible for thereduction of benzene. In our case, however, the appearance of[M+2H]+ was more dependent to the presence of gas phase waterin the air. For example, when the experiment was repeated with99.995% nitrogen with< 10 ppm water as working gas, thespectrum for benzene still shows M+ as themajor peak even whenthe ion source pressures was increased to 6 atm (SupplementaryFigures S7a and S7b). The ion signal for protonated benzeneand [M+ 2H]+ became stronger when additional pure water wasinjected to the heated nitrogen stream (Supplementary FiguresS7c and S7d). Although we do not exclude the possibility ofsurface reaction in our ion source, the primary source of hydrogenfor [M+ 2H]+ was likely from the gas phase water.Compared to compressed air, a better performance for the

detection of benzene or similar aromatic compound like toluenecould be achieved by using highly dried air or pure nitrogen. Acomparison of TIC and ion intensities for the benzene-relatedion acquired with compressed air and 99.995% nitrogen is shown


in Fig. 6. For compressed air, the intensity for M+ droppeddrastically with pressure, and there was no increase in TIC. Inthe case of nitrogen, signal enhancement for M+ could beobserved by optimizing the ion source pressure at around 2atm. The enhancement was, however, not as great as those ofmethamphetamine and RDX.

In sum, for the ionization process that involved protonation oradduct ion formation, the detected ion abundance was found toincrease with ion source pressure until it reached a plateaus ataround 4–5 atm. The origins of enhancement could be due tothe following factors.

i. The increased gas phase density of analyte and reactant ionsled to a higher collision frequency among them, and itpromoted the ion/molecules reaction for the CI.

ii. Higher gas intake rate to the vacuum with higher pressuredifference introduced more ions into the mass spectrometerwithin certain period of time.

iii. Owing to smaller ion mobility and diffusion coefficient underhigher gas pressure, the spread out of ions in the ion sourcebecame slower, and that was beneficial for the ion sampling.

However, increasing the ion source pressure also introducedadverse effect to the ion transmission at the ion transportcapillary. This loss was due to the annihilation of charge whenions collided with the inner surface of the metallic capillary.Effective ion transportation relied on the laminar flow, which ischaracterized by a small Reynolds number inside the capillary.[32]

When the ion source pressure was increased above ~2 atm, thelaminar flow was unlikely to be maintained when the Reynoldsnumber rose beyond 2300 (Supplementary Figure S2c), and higherion transmission loss was expected. However, below 4 atm, we stillobserved a net gain in ion intensities with the increase of ion sourcepressure because the ion loss was compensated by higher ionabundance in the ion source and higher ion intake rate. At muchelevated pressures like those above 4–5 atm, the ion loss due toturbulent flow became much greater, and the detected ionintensities began to drop with pressure. Innovation on the iontransmission technique is certainly necessary in the future forfurther development of high pressure ion source.

Conclusion

Super-atmospheric pressure CI-MS has been performed on a com-mercial mass spectrometer with operating air pressure up to 7 atm.The increase of ion intensities for gaseous drug and explosive withthe ion source pressure suggests that pressurizing the ion sourcecould be a useful strategy to further improve the sensitivity ofAPCI-MS, particularly for the application that involves high flow rateof sample gas. Besides APCI, the abundance of ions from thegaseous compound that was intentionally added to our previoushigh pressure electrospray ion source was also found to be higherunder super-atmospheric pressure. This suggests that the presentwork can be extended to other gaseous ionization techniques likesecondary ESI,[29] and atmospheric pressure photo-ionization.

Acknowledgements

This work was supported by the Grants-in-Aid for ScientificResearch (Kakenhi, Grant no. 12015369) from JSPS, and theProgram to Disseminate Tenure Tracking System from theMinistry of Education, Culture, Sports, Science and Technologyof the Japanese government.


a) TIC b) TIC

c) d)

Figure 6. Ion intensities versus ion source pressure for the chemical ionization of benzene using compressed air (a and c), and 99.995% nitrogen (b and d)as working gases. a) and b) are total ion current (TIC) and c) and d) are ion intensities for M+ (solid circle), [M+H]+ (open circle), [M+ 2H]+ (solid square)and [M+O]+ (open triangle).


397

Supporting information

Supporting information may be found in the online version ofthis article.

Reference[1] E. C. Horning, D. I. Carroll, I. Dzidic, K. D. Haegele, M. G. Horning,

R. N. Stillwell. Liquid chromatograph—mass spectrometer—computeranalytical systems: A continuous-flow system based on atmosphericpressure ionization mass spectrometry. J. Chromatogr. A 1974, 99,13–21.

[2] D. I. Carroll, I. Dzidic, R. N. Stillwell, K. D. Haegele, E. C. Horning.Atmospheric pressure ionizationmass spectrometry. Corona dischargeion source for use in a liquid chromatograph-mass spectrometer-computer analytical system. Anal. Chem. 1975, 47, 2369–2373.

[3] D. A. Lane, B. A. Thomson. Monitoring a Chlorine Spill from a TrainDerailment. J. Air Pollut. Control Assoc. 1981, 31, 122–127.

[4] Y. Takada, H. Nagano, Y. Suzuki, M. Sugiyama, E. Nakajima, Y. Hashimoto,M. Sakairi. High-throughput walkthrough detection portal for counterterrorism: detection of triacetone triperoxide (TATP) vapor byatmospheric-pressure chemical ionization ion trap mass spectrometry.Rapid Commun. Mass Spectrom. 2011, 25, 2448–2452.

[5] M. S. B. Munson, F. H. Field. Chemical Ionization Mass Spectrometry.I. General Introduction. J. Am. Chem. Soc. 1966, 88, 2621–2630.

[6] F. J. Andrade, J. T. Shelley, W. C. Wetzel, M. R. Webb, G. Gamez,S. J. Ray, G. M. Hieftje. Atmospheric Pressure Chemical IonizationSource. 1. Ionization of Compounds in the Gas Phase. Anal. Chem.2008, 80, 2646–2653.

[7] N. Na, M. Zhao, S. Zhang, C. Yang, X. Zhang. Development of aDielectric Barrier Discharge Ion Source for AmbientMass Spectrometry.J. Am. Soc. Mass Spectrom. 2007, 18, 1859–1862.

[8] K. Hiraoka, S. Fujimaki, S. Kambara, H. Furuya, S. Okazaki.Atmospheric-pressure Penning ionization mass spectrometry. RapidCommun. Mass Spectrom. 2004, 18, 2323–2330.

[9] R. B. Cody, J. A. Laramee, H. D. Durst. Versatile New Ion Source for theAnalysis of Materials in Open Air under Ambient Conditions. Anal.Chem. 2005, 77, 2297–2302.


[10] L. C. Chen, Y. Hashimoto, H. Furuya, K. Takekawa, T. Kubota, K. Hiraoka.Rapid detection of drugs in biofluids using atmospheric pressurechemi/chemical ionization mass spectrometry. Rapid Commun. MassSpectrom. 2009, 23, 333–339.

[11] L. C. Chen, Z. Yu, H. Furuya, Y. Hashimoto, K. Takekawa, H. Suzuki,O. Ariyada, K. Hiraoka. Development of ambient sampling chemi/chemical ion source with dielectric barrier discharge. J. Mass Spectrom.2010, 45, 861–869.

[12] L. C. Chen, Z. Yu, K. Hiraoka. Vapor phase detection of hydrogenperoxide with ambient sampling chemi/chemical ionization massspectrometry. Anal. Methods. 2010, 2, 897–900.

[13] T. R. Covey, B. A. Thomson, B. B. Schneider. Atmospheric pressure ionsources. Mass Spectrom. Rev. 2009, 28, 870–897.

[14] L. C. Chen, M. K. Mandal, K. Hiraoka. High Pressure (>1 atm) ElectrosprayIonization Mass Spectrometry. J. Am. Soc. Mass Spectrom. 2011, 22,539–544.

[15] L. C. Chen, M. K. Mandal, K. Hiraoka. Super-Atmospheric PressureElectrospray Ion Source: Applied to Aqueous Solution. J. Am. Soc.Mass Spectrom. 2011, 22, 2108–2114.

[16] M. M. Rahman, L. C. Chen, K. Hiraoka. Development of high-pressureprobe electrospray ionization for aqueous solution. Rapid Commun.Mass Spectrom. 2013, 27, 68–74.

[17] L. C. Chen, M. M. Rahman, K. Hiraoka. Non-vacuum field desorptionion source implemented under super-atmospheric pressure. J. MassSpectrom. 2012, 47, 1083–1089.

[18] E. J. Davis, P. Dwivedi, M. Tam, W. F. Siems, H. H. Hill. High-PressureIon Mobility Spectrometry. Anal. Chem. 2009, 81, 3270–3275.

[19] N. G. Adams, L. M. Babcock. Advances in Gas Phase Ion Chemistry. JAIPress, Greenwich, Connecticut. 1996.

[20] K. E. Sahlstrom,W. B. Knighton, E. P. Grimsrud. Rate Constants for the GasPhase Reaction of Chloride Ion with Methyl Bromide over the PressureRange 300 to 1100 Torr. J. Phys. Chem. A 1997, 101, 5543–5546.

[21] J. Peng, N. Puskas, P. B. Corkum, D. M. Rayner, A. V. Loboda. High-Pressure Gas Phase Femtosecond Laser Ionization Mass Spectrometry.Anal. Chem. 2012, 84, 5633–5640.

[22] J. Fernandez de la Mora. Ionization of vapor molecules by anelectrospray cloud. Int. J. Mass. Spectrom. 2011, 300, 182–193.

[23] M. Haidara, A. Denat, P. Atten. Corona discharges in high pressureair. J. Electrostat. 1997, 40–41, 61–66.



398

[24] A. T. Droege, P. C. Engelking. Supersonic expansion cooling ofelectronically excited OH radicals. Chem. Phys. Lett. 1983, 96, 316–318.

[25] J. Skalný, G. Hortváth, N. J. Mason. Spectra of Ions Produced byCorona Discharges. AIP Conf. Proc. 2006, 876, 284–293.

[26] K. Sekimoto, M. Sakai, M. Takayama. Specific Interaction BetweenNegative Atmospheric Ions and Organic Compounds in AtmosphericPressure Corona Discharge Ionization Mass Spectrometry. J. Am. Soc.Mass Spectrom. 2012, 23, 1109–1119.

[27] Y. Takada, H. Nagano, M. Suga, Y. Hashimoto, M. Yamada, M. Sakairi,K. Kusumoto, T. Ota, J. Nakamura. Detection of Military Explosives byAtmospheric Pressure Chemical Ionization Mass Spectrometry withCounter-Flow Introduction. Propellants Explos. Pyrotech.2002, 27, 224–228.

[28] M. J. Bogusz, K. D. Krüger, R. D. Maier. Analysis of underivatized amphet-amines and related phenethylamines with high-performance liquid


chromatography-atmospheric pressure chemical ionization massspectrometry. J. Anal. Toxicol. 2000, 24, 77–84.

[29] M. Tam, H. H. Hill. Secondary Electrospray Ionization-Ion MobilitySpectrometry for Explosive Vapor Detection. Anal. Chem. 2004,76, 2741–2747.

[30] N. Na, Y. Xia, Z. Zhu, X. Zhang, R. G. Cooks. Birch Reduction ofBenzene in a Low-Temperature Plasma. Angew. Chem. Int. Ed.2009, 121, 2051–2053.

[31] Y. B. Golubovskii, V. A. Maiorov, J. Behnke, J. F. Behnke. Influence ofinteraction between charged particles and dielectric surface over ahomogeneous barrier discharge in nitrogen. J. Phys. D: Appl. Phys.2002, 35, 751–761.

[32] B. Lin, J. Sunner. Ion transport by viscous gas flow throughcapillaries. J. Am. Soc. Mass Spectrom. 1994, 5, 873–885.


super-atmospheric pressure chemical ionization mass spectrometry

Documents