Gas Analysis by Electron Ionization Combined with ... - ACS Publications

May 20, 2018 - Gas introduction via controlled gas pulses, electron ionization, ion–molecule reactions, ion selection, and detection are all perform...
0 downloads 0 Views 1MB Size
Subscriber access provided by Warwick University Library

Article

Gas Analysis by Electron Ionization Combined with Chemical Ionization in a Compact FT-ICR Mass Spectrometer Michel Heninger, Hélène Mestdagh, Essyllt Louarn, Gérard Mauclaire, Pierre Boissel, Julien Leprovost, Elsa Bauchard, Sebastien Thomas, and Joel Lemaire Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b01107 • Publication Date (Web): 20 May 2018 Downloaded from http://pubs.acs.org on May 20, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Gas Analysis by Electron Ionization Combined with Chemical Ionization in a Compact FT-ICR Mass Spectrometer Michel Heningera*, Hélène Mestdagha, Essyllt Louarna, Gérard Mauclaireb, Pierre Boisselb, Julien Leprovostb, Elsa Bauchardb, Sebastien Thomasa, Joël Lemairea a Laboratoire de Chimie Physique, CNRS UMR 8000, Université Paris-Sud, Université Paris Saclay, 91405 Orsay, France. b AlyXan, centre Hoche, 3 rue Condorcet, 91260 Juvisy sur Orge, France * Corresponding author: Tel.: +33 (0)1 6915 3458, [email protected]

1 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 26

ABSTRACT:

In this paper, a compact Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometer based on a permanent magnet is presented. This instrument has been developed for real-time analysis of gas emissions. The instrument is well suited to industrial applications or analysis of toxic and complex samples where the concentrations can vary rapidly on a wide range. The novelty of this instrument is the ability to use either electron ionization (EI) or chemical ionization (CI) individually or both of them alternatively. Also in CI mode, different precursor ions can be used alternatively. Volatile organic compounds (VOCs) from the ppb level to very high concentrations (% level) can be detected by CI or EI. The magnet is composed of three Halbach arrays, and the nominal field achieved is 1.5 T. The ICR cell is a 3 cm side length cubic cell. . The mass range is 12 to 200 u with a broad band detection. The mass accuracy of 0.005 u and the resolving power allow the separation of isobaric ions such as C3H7+ and CO2+. Gas introduction via controlled gas pulses, electron ionization, ion-molecule reactions, ion selection and detection are all performed in the ICR cell. The potential of the instrument will be illustrated by an analysis of a gas mixture containing trace components at ppm level (VOCs) and components in the 0.5-100% range (N2, alkanes, CO2).

2 ACS Paragon Plus Environment

Page 3 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Real-time gas analysis is mandatory in many areas such as atmospheric control in working areas or industrial process control1,2. This method implies a fast time response from a few seconds to one minute and the ability to follow a broad variety of compounds in a wide concentration range.

Volatile organic compound (VOC) analysis is usually performed by an offline laboratory analysis of samples collected in containers, bags or with sorbent materials. Gas chromatography (GC) coupled with flame ionization detection (FID) for quantification and mass spectrometry (MS) for identification is the reference technique for VOC analysis. However, this technique is still laboratory based3,4 and does not enable real-time measurements. Portable and field GC-MS, potentially associated with a preconcentrator, can be used only when the required time response is longer than 10 minutes5.

For faster measurements, several broadband techniques have been applied to VOC monitoring in real time (e.g., laser6 or ion mobility7 techniques), among which chemical ionization mass spectrometry (CIMS) is the reference8. In proton transfer reaction mass spectrometry (PTRMS)9 or selected ion flow tube mass spectrometry (SIFT-MS) instruments10, H3O+ is the most often used precursor ion. The proton transfer reaction of H3O+ with most VOCs is fast with little fragmentation, whereas H3O+ ions do not react with the major components of air, namely, N2, O2, Ar, and CO2. The analysis is quantitative if the reaction rate coefficient is known either from experimentation, which is the case for many VOCs, or from calculations based on molecular properties11. PTR can be utilized with other precursor ions such as C6H5F2+, which provides less fragmentation for large oxygenated molecules12. For molecules that do not react with H3O+, other positive (NO+, O2+, CF3+)13,14 or negative (OH–, O–)15–17 precursor ions can be used. Specific precursor ions such as N2H5+ can also be used for carbonyl compounds18,19 for example. It is also possible to alternate between several precursor ions to increase the selectivity and distinguish between isobaric compounds20,21.

Because the compounds are identified by their mass measurements, the mass resolution is important. When dealing with a complex mixture, an exact mass measurement may be a crucial piece of information to unequivocally identify the molecules. The first generation of PTR-MS and SIFT instruments was based on quadrupole mass spectrometers with a unit mass resolution. A new generation of instruments based on a high-resolution time of flight mass

3 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 26

spectrometer has been developed by Ionicon Analytik GmbH, Innsbruck Austria22 and Tofwerk AG, Thun, Switzerland23.

Fourier transform ion cyclotron resonance (FT-ICR) spectrometry is one of the most powerful techniques concerning mass accuracy and resolution. Therefore, developments regarding transportable FT-ICR mass spectrometers were undertaken as soon as permanent magnets were available. The main advantages of the FT-ICR technique include high-precision mass measurements for better compound identification, high mass resolution for quasi-isobaric separation and broadband detection for a better compound screening.

The first generation of permanent magnet FT-ICR mass spectrometers developed by Campana et al. and Extrel FTMS, US

24

, and the Nikkiso Co. Ltd., Japan25 was limited by the use of

continuous gas sampling. FT-ICR detection needs a good vacuum to obtain high resolution. The presence of gas during detection destroyed the coherent motion of the ions in the cell and degraded the resolution, which limited the samples to low pressure undiluted gases, excluding complex mixtures. The former instrument24 was mainly used for ion/molecule reactions, and measurement of the exact mass of ions. The performance obtained with the latter instrument25 for small molecule separation such as He/D2 and N2/CO/C2H4 was good. However, the low nominal magnetic field (0.646 T) was a drawback for the detection of larger molecules.

Instruments based on magnetic structures such as Halbach magnetic cylinders obtained much better results26. The MICRA instrument27 developed in our laboratory in 2000 was mainly used for spectroscopic28 and kinetic studies29. This magnetic configuration can very efficiently produce an intense and homogenous field (close to 1 T) inside the magnet bore with a relatively low external stray field. Additionally, a pulsed gas sampling method was designed for reproducible gas pulse generation associated with low background pressure during ion detection. Using two gas inlet lines, chemical ionization such as PTR-MS could be performed, which produces H3O+ precursor ions from the first H2O pulse, and the second line was used to introduce the sample pulse.

During the same period, Siemens Energy & Automation Inc. launched the commercialization of a compact and low-cost FT-ICR instrument for industrial applications30. The vacuum chamber was evacuated by an ionic pump ensuring very low pressure during detection, and 4 ACS Paragon Plus Environment

Page 5 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

the gases were sampled by pulse valves designed for the instrument31. The main objective of this project was to bring the power of FT-ICR mass spectrometry to online measurements such as for monitoring of gas mixtures in fermenters, electronic applications (detection of NF3, SF6 or perfluorocarbons), and catalyst characterization.

More recently, two instruments with external ion sources have been presented32,33. These developments are based on a magnet configuration producing a magnetic field that is coaxial with the magnetic cylinder’s main axis. The strength of the field is 1 T.

Among the above-mentioned FT-ICR instruments, only MICRA allows controlled CI, and none of them was designed for CI analysis of complex mixtures. In FT-ICR mass spectrometry, the ion-molecule reactions occur at near thermal kinetic energy. Consequently, the technique is suitable for chemical ionization. The ion trap can be used successively for the production of precursor ions, for CI reaction with the sample and for detection34. The gases and the different steps of the sequence must be pulsed. Producing a large variety of precursor ions is easy, and switching from positive to negative mode implies only a change in the polarity of the trapping plates.

Recent high-resolution CI-MS instruments22,23 are very powerful tools for trace-level VOC analysis in air since they combine high mass resolution and high sensitivity. However, chemical ionization is not necessarily appropriate for monitoring all types of gas mixtures. This is the case if the analytes are not reactive with the usual CI precursor ions (e.g., small alkanes)35 or if a major component of the matrix reacts with the products of the CI reaction. Both small inorganic gases or light alkane mixtures may be more conveniently analyzed using electron ionization (EI) associated with high-resolution mass detection.

This paper presents a new compact and transportable FT-ICR instrument based on a permanent magnet, which is optimized for analysis of complex VOC mixtures and is more efficient than MICRA in terms of mass measurement accuracy. This instrument, referred to as BT4, can follow at high mass resolution the VOCs present at trace concentration using CI mode and the other gases present at 0.1-100% concentration levels using EI mode.

5 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 26

Compared to PTR-MS or SIFT instruments, the present mass spectrometer has a lower sensitivity but works at very high concentrations (%) without dilution. Switching between EI and CI is easy, and it is also possible to alternate negative or positive precursor ions36. Aqueous solutions samples using a membrane inlet can also be analyzed in this instrument37. Thus the present instrument has been designed to follow the composition of complex mixtures where concentrations vary rapidly on a wide range, possibly containing unknown analytes and isobaric compounds. EXPERIMENTAL SECTION

Specifically designed for real-time monitoring of VOC traces at atmospheric pressure close to their occurrences, the described FT-ICR spectrometer is compact (l×w×h: 65×72×104 cm). The key parts are the permanent magnet and the matching ICR cell.

Magnetic field The highest field value with a high homogeneity compatible with a reasonable size and weight was reached for a 60 kg FeNdB magnet with a usable bore of 52 mm and overall dimensions of a 22 cm length and a 25 cm external diameter. The field strength was 1.5 T with a homogeneity better than 0.5% over the central 27 cm3 occupied by the ICR cell.

As illustrated in Figure 1, the magnet consisted of the juxtaposition of three Halbach cylinders. Each cylinder was composed of sixteen segments.

6 ACS Paragon Plus Environment

Page 7 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 1: Schematic of the magnet used for this study, composed of three Halbach cylinders. The small red arrows indicate the magnetization of each segment. The resulting magnetic field inside the bore is indicated by the thick yellow arrow. In this type of magnet, the magnetic field direction is not co-linear with the bore axis but is perpendicular to it. This precludes the use of the common three sections of the open-ended cell often associated with supraconducting magnets. A cell derived from those developed for resistive magnets38 was designed and built for the present instrument, and it is described in the next section.

ICR cell

The ICR cell developed for this instrument has an open configuration geometry designed to ensure gas pulses with a fast rise time and a fast enough decay. The decay time depends on the pumping efficiency through the 50 mm diameter tube that contains the cell. Therefore, the cell can be “open ended” by suppressing the two electrodes perpendicular to the bore axis, normally used for the excitation of the cyclotron motion. For the ions, the electrostatic field in the cell is equivalent to that of a 3×3×3 cm3 cell. Only two plates successively cause the excitation and then the detection, which uses a rapid optoelectronic switching device38,39.

To prevent ions from escaping, a zero potential must be maintained on the open sides; this is done on both sides by a pile of six grounded thin plates.

Halogen lamps are placed near the cell on both sides and are used to heat the cell and its surroundings. The temperature of the cell is measured using a PT100 sensor probe and is regulated through software.

The filament assembly consists of a ceramic holder and two stainless steel bars on which a rhenium yttria-coated filament ribbon is welded. A small stainless steel cover, connected to 7 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 26

the more negative side, acts as a repeller for the electrons. An intermediate diaphragm electrode is present between the filament and the trapping plate, and the applied voltage allows both pulsed electrons through the cell and electron current regulation during the pulse. The electron beam goes through 1.2 mm diameter holes in the trapping plates, and a collector electrode placed after the exit hole of the beam is used to monitor the electron current admitted through the cell.

The intensity of the ionizing electron beam is controlled - via the potential applied on the diaphragm electrode - by a circuit comparing the current collected during the pulse with the current value set through software by the operator.

Such a fast and efficient regulation of the electron beam intensity is necessary in our experiments involving a rapid burst of gases. The filament emissivity is strongly dependent on its temperature and, therefore, on the variable cooling associated with the duration, intensity and frequency of gas pulses occurring during the measurement sequence.

Figure 2: ICR cell: (1) filament, (2) trapping plate, (3) excitation/detection plate, (4) pile of six grounded thin plates, (5) halogen lamps and (6) diaphragm electrode.

Data acquisition system

The different voltages, currents, pulses involved are generated by three specially designed electronic cards, and the relevant parameters are preset by the operator through the associated software. During the measurements, the pulse sequence, which is predefined by the operator, 8 ACS Paragon Plus Environment

Page 9 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

is delivered autonomously by a custom digital signal processor card with a time accuracy better than 10 ns.

As close as possible to the cell, an electronic “detection head” comprises a two-phase excitation amplifier, a broadband and low noise differential preamplifier for the induced image current during detection, as well as a fast but low capacitance commutation device between the excitation/detection circuit and the pair of excitation-detection plates.

An Acquitek card (CH-3150) with a 12-bit resolution and a maximum sampling rate of 20 MHz is used successively as an arbitrary waveform generator during excitation and as a digitizer during detection.

Sampling

To obtain good reproducibility of the sample gas pulses, we have adopted a configuration similar to that used in earlier versions of the instruments12,34,40. The sample flow is continuous and directed either in the vacuum chamber housing the ICR cell (during the gas pulse) or to another vacuum chamber, also fitted with a 70 L/s turbomolecular pump using a three-way valve (Parker miniature valve, series 9). Only a minute amount of gas is introduced into the mass spectrometer.

For atmospheric pressure sampling, the response time and time resolution of the measurements is crucial for real-time analysis since the instrument must be able to follow a concentration change in the sample. A special line was designed to sample gases at atmospheric pressure (Figure 3). To obtain a time resolution close to one second, the pressure drop is produced in two steps: the first drop is from atmospheric pressure down to a pressure of a few millibars using a needle valve and a membrane primary pump, and the second drop uses a stainless steel capillary (internal diameter, 130 µm; length, 50 mm) placed before the three-way valve.

Because the different steps of the experiment occur in the ICR cell, the different gases used must be pulsed. The sample gas quantity introduced into the instrument can be adjusted by changing the pulse width. This adjustment is essential when conducting CI since the reaction conditions must be adapted to the VOC concentration present in the sample (from a few tens 9 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 26

ppb to the % range). Furthermore, the sample amount must be high enough to maximize the sensitivity but low enough to minimize secondary reactions that distort the quantification (CI reaction progress below 25%).

When chemical ionization is used, a first gas pulse is introduced into the ICR cell for the formation of the precursor ions, followed by a pulse of the sample gas with a typical pressure of 3 10-5 mbar and a valve opening time of 300 ms. For electron ionization, there is only one pulse and an even lower pressure that is typically a few 10-7 mbar for an opening time of 5 ms. After the sample pulse, sufficient time must be allowed for the pressure to come back to the base level (a few 10-8 mbar), so that the detection will not be appreciably perturbed by collisions.

In PTR-MS mode, the gas quantity in the cell must be known precisely to quantify the concentration of the different VOCs detected. The gas pressure in the ICR cell is measured using a Bayard-Alpert ionization gauge (MICRO-ION 354 Granville Phillips, MKS). The gauge is shielded to avoid perturbation from the magnetic field. The temporal pressure profile in the cell given by the analog output of the Bayard-Alpert gauge is recorded and integrated over the reaction time to provide the gas quantity introduced into the cell.

Figure 3: Schematic of the interface for sampling gases at atmospheric pressure

Sequence 10 ACS Paragon Plus Environment

Page 11 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Electron Ionization: In this mode, the sample is introduced into the cell where it is ionized by a 70 eV electron beam. When using the electron ionization mode, it is necessary to work at a low sample pressure and to reduce the delay between the ionization and the detection to minimize secondary reactions. A typical sample gas pulse length is 5 ms for a stationary state pressure of 10-7 mbar. The delay between the gas pulse and the electron pulse is 80 ms for a 2 ms electron pulse. Finally, the delay between the end of the electron pulse and the detection is 18 ms. The overall sequence duration is less than 500 ms.

Chemical Ionization: To produce the ionic precursors used for chemical ionization, the first gas pulse is introduced in the instrument. Then, the precursor ion is isolated by FT-ICR selective ejection techniques, and the sample is introduced in the second gas pulse. To produce H3O+, a 5 ms gas pulse of H2O is introduced into the cell and ionized by electron impact, resulting in the formation of H2O+, OH+, O+, H2+, and H+. The electron beam is placed sufficiently early (30 ms after the beginning of the gas pulse) to ensure that all these ions react with the H2O neutrals, finally leading to formation of a nearly pure H3O+ ion packet in the ICR cell. A 450 ms delay follows for the complete formation of H3O+ and for H2O gas pumping. Because a few background ions can remain, an excitation signal is applied to eject all unwanted ions before introducing the gas sample for 300 ms at 3 10-5 mbar. After a 500 ms delay for cell pumping, all the ions present are detected. The overall sequence duration is 4 s.

Gases

Three gas cylinders were used for these experiments: one cylinder containing methane as the main component (89%) with butane (3%), propane (5%) and carbon dioxide (3%) (cylinder alkanes/CO2), a cylinder of N2 with five organic compounds (ethanol, benzene, toluene, cyclohexanone and xylene) at 5 ppm concentration (cylinder VOC/N2) and pure N2 (Alphagaz 1, Air Liquide) for gas dilution. A GasmixTM Zephyr diluter (AlyTech, Juvisy sur Orge, France) was used to produce a mixture from the two gas cylinders with known concentrations.

RESULTS AND DISCUSSION 11 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 26

Electron ionization and chemical ionization can be used in the instrument to analyze a gas mixture. It is also possible to alternate these methods or use successively different precursor ions in either the positive or negative mode.

Pressure measurements

Figure 4 shows the temporal pressure profile in the cell given by the analog output of the Granville Phillips gauge for different gas pulse durations. The various curves begin with an exponential rise with a time constant of ~80 ms and then an exponential decrease with a time constant of 300 ms. If the pulse duration is long enough, i.e. >100 ms, the pressure reaches a plateau (stationary state pressure).

The area under the pressure curve is proportional to the number of molecules introduced in the instrument. This area is automatically integrated to provide a quantity denoted Pxt, which is expressed in pressure x time units. In Figure 4b), the Pxt quantity depends linearly on the valve opening time. Finally, the Pxt quantity obtained from the gauge measurements will be directly used to quantify the VOC concentrations. The possibility to adjust both the opening time of the valve and the pressure using the flow control valve (figure 3) enables wide-range control of the sample gas amount introduced in the instrument.

a)

b)

12 ACS Paragon Plus Environment

Page 13 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 4: a) Pressure profiles of the gas pulses measured in the cell vacuum chamber for different opening times of the three-way valve. b) Integrated pressure of the sample gas pulse as a function of the three wayvalve opening time.

Electron ionization.

Figure 5 shows the mass spectrum obtained for a 70 eV electron ionization on a mixture produced with 60% of the gas flow from the alkane/CO2 cylinder and 40% of the gas from the N2 cylinder. This spectrum is compared with the theoretical spectrum calculated from the 70 eV spectrum and the ionization cross sections present in the National Institute of Standards and Technology (NIST) databases.

The main features of experimental (red) and calculated (green) spectra are the same. Some differences may come from the presence of secondary reactions. Indeed, the fragmentation of alkanes may lead to EI spectra that are different from the NIST spectra since the FT-ICR instrument is an ion trap, and the ion residence time is longer than in a magnetic sector instrument used for the NIST database, which favors secondary reaction product formation. However, the amount of CH5+ ions (m/z 17) produced by the fast reaction CH4+ + CH4 is small, confirming that the secondary reactions are limited.

Figure 5: Comparison of experimental (red) and calculated (green) spectra under electronic ionization conditions. Peaks observed at mass 44 are magnified in the upper part. Two isobaric compounds (CO2+ and C3H8+) are clearly differentiated. One of the advantages of FT-ICR mass spectrometry compared to a quadrupole analyzer is the high resolution that allows the identification and separation of isobaric compounds. The 13 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 26

main masses detected are reported in Table 1. The experimental masses are compared to the exact mass calculations, and then a chemical formula is rendered. The agreement between experimental and exact masses is excellent as the differences between detected and measured masses have a standard deviation of 0.0012 u. Thus, a well-separated doublet is observed for mass 44, which corresponds to the quasi-isobaric ions CO2+ and C3H8+, magnified in Figure 5. The relative mass accuracy obtained is much better than the 0.5% value of the local field homogeneity. In the FT process the relevant factor is the mean value of the magnetic field experienced by ions along a cyclotron orbit, and the averaging induced by the ion trajectories is particularly efficient with this transverse magnet configuration. A few small peaks were unexpected in the EI spectrum of the sample: m/z 18 (H2O+), 19 (H3O+) and 32 (O2+). The originating species are easily identified, namely, H2O, which is the main background gas in the cell (~10-8 mbar), and O2. The presence of the latter is probably due to a small air leak in the sampling device. This finding is confirmed by the spectral pattern at m/z 28. When accumulating the signal, three quasi-isobaric peaks are distinguished, which correspond to CO+ from the dissociative ionization of CO2, C2H4+ from C3H8 and C4H10, and N2+ from N2 in the air leak. The flows from the two gas cylinders were adjusted with a GasmixTM diluter to produce a progressive dilution of the alkanes/CO2 with the N2/VOCs while maintaining the total flow at a constant value of 100 sccm, starting with 100% alkanes/CO2 and decreasing the proportion to 80%, 60%, 50%, 40%, 20% and 0%.

Nominal mass (u)

Detected mass (u)

Exact mass (u)

Ion molecular formula

Main original neutral(s)

∆m (EI)

58

58.0749

58.0783

C4H10+

Butane

-0.0034

44.0638

44.0626

C3H8+

Propane

0.0012

43.9924

43.9898

CO2+

CO2

0.0026

43

43.0553

43.0548

C3H7+

Butane, Propane

0.0005

41

41.0402

41.0391

C3H5+

Butane, Propane

0.0011

29.0401

29.0391

C2H5+

Butane, Propane

0.0010

29.0140

+

N2 /Alkanes

0.0024

+

Butane, Propane

-0.0047

44

29 28

29.0164 28.0266

28.0313

N2H

C2H4

+

14 ACS Paragon Plus Environment

Page 15 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

28.0073

N2+

28.0061

N2

0.0012

Butane, Propane

0.0005

H2O

0.0011

Methane

0.0007

CH4

Methane

0.0009

+

27

27.0240

27.0235

C2H3

19

19.0195

19.0184

H3O+

17.0391

+

17.0398 17

16 15 14

CH5 13

+

17.0355

17.0346

17.0035

17.0027

OH+

H2O

0.0008

16.0313

CH4

+

Methane

0.0002

+

Alkanes

-0.0008

Alkanes

-0.0011

N2

-0.0004

16.0311 15.0227

15.0235

CH3

14.0145

14.0156

CH2+

14.0027

14.0031

+

N

Table 1: Main masses detected with their attribution and the deviation from the exact mass for one scan. The peaks related to N2 (N2+, N2H+ and N+) are observed to increase regularly with N2 ratio, whereas the various peaks originating from the alkanes/CO2 cylinder decrease (figure 6). A close to linear variation is observed for the sum of N2+, N+ and N2H+ ion intensities and for the sum of the main peaks that originate from the alkanes/CO2. The limits of detection (LODs) are determined from the calibration curves and a blank from the N2 bottle. This blank and each point of the calibration curves are recorded for at least 80 scans. For a given analyte, a representative ion or the sum of the representative ions is chosen: e.g., m/z 16 for CH4, 58 for butane, 43.99 for CO2, and 44.06 for propane. The LOD is calculated from the calibration curve as the concentration for which the signal would be + 3 σblank (the mean value plus three times the standard deviation). Under these conditions, the LODs are between 0.3% and 0.7%. The use a 1.5

T magnet and

a larger ICR

cell with an

improved

design enables

efficient

separation

quasi-isobaric

ions.

of

15 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 26

Figure 6: Variation in the sum of ions from alkanes/CO2 (CxHy+, CO2+; red dots) and in the sum of ions from N2 (N2+, N+; blue dots) as a function of the dilution.

Chemical ionization

In the past, FT-ICR mass spectrometry was widely used for ion-molecule rate coefficient and branching ratio measurements41,42. As explained above, the kinetic energy of the ions is nearly thermal. Moreover, in gas trace analysis conditions, the collisions with unreactive O2 and N2 species from the matrix gas efficiently thermalize the precursor ions. Consequently, this technique is suitable for a chemical ionization technique such as PTR-MS.

The reaction conditions are such that less than 25% of the precursor ions react to produce protonated VOCs so that secondary reactions are minimized; thus, H3O+ is the major component in the mass spectra. The intensity of a product ion is: I(MH+) = kM CM Pxt I(H3O+)

(1)

where kM is the proton transfer rate coefficient and CM the mixing ratio of M. Because the 5 VOCs present in the sample have the same concentration, their intensity is proportional to the bimolecular rate coefficient of their reaction with H3O+. As for electron ionization, the agreement between the experimental and exact masses demonstrates a standard deviation of 0.0016 u (Table 2). 16 ACS Paragon Plus Environment

Page 17 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Nominal mass (u)

Detected mass (u)

Exact mass (u)

Ion molecular formula

Main original neutral(s)

∆m

107

107.0878

107.0861

C8H11+

Xylene

0.0017

99

99.0830

99.0810

C6H11O+

Cyclohexanone

0.0020

93

93.0750

93.0704

C7H9+

Toluene

0.0046

79

79.0568

79.0548

C6H7+

Benzene

0.0020

45

45.0373

45.0340

C2H5O+

Ethanal

0.0033

19

19.0184

19.0184

H3 O+

Water

0.0000

Table 2: Main masses detected with their attribution and the deviation from the exact mass. The protonated analytes are the only products of the CI reaction. Therefore, MH+ is clearly the representative ion for the LOD determination. The LOD for these 5 VOCs and the slope of the calibration curve were determined by diluting the gas of the 5 ppm VOC gas cylinder in pure N2 to produce VOC concentrations in the range from 0 to 5 ppm. Figure 7 shows the measured relative concentrations as a function of injected relative concentrations. The measured concentrations are determined from the ion intensities, Pxt values and protonation rate coefficients for the 5 compounds. These rate coefficients are approximated to the corresponding capture rate coefficients taken from Latappy et al.21.

The observed response for each the VOC is linear, with a slope close to one except for that of cyclohexanone. In this concentration range, the LOD obtained for one scan (in 1 to 5 sec) is 250 ppb for acetaldehyde, 500 ppb for the benzene, toluene and xylene (BTX) compounds and one ppm for cyclohexanone. The concentration measured for cyclohexanone is half the concentration injected. This discrepancy may be due to losses caused by adsorption in the sampling system or on the vacuum chamber walls since cyclohexanone is a polar molecule.

17 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 26

Figure 7: a) Measured concentrations calculated using the capture rate coefficients as a function of injected concentrations. b) VOC mass spectra obtained for a 2 ppm injection. Averaging the transient signal over n sequences lowers the LODs by a factor √n with a degradation of the temporal response to a few minutes: 20 ppb for 100 accumulations in ca. 3 min. Online preconcentrators devices like MIMS (Membrane Inlet Mass Spectrometry) cans also be used to improve the LODs. Similarly to the EI mode, the innovative design of the cell, with switching between excitation and detection on the same pair of electrodes, leads to an open geometry which allows efficient pumping of the vacuum chamber even at the relatively high pressure used under CI conditions: the response time remains short for a given performance in terms of mass resolution and sensitivity. The possibility to adjust the Pxt factor (figure 4) allows the detection of high concentration in % level without dilution.

Switching between electron ionization and chemical ionization

Alternation between different ionization modes is fast and easy when using an FT-ICR mass spectrometer. Alternating between EI and CI implies a switch between three different inlet lines: one line is for the CI precursor; one line for sample introduction in the CI mode and another sample line for the EI mode that is operated at a much lower pressure.

18 ACS Paragon Plus Environment

Page 19 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

The residual pressure in the cell must remain as low as possible, especially during the EI sequence, which is performed entirely at low pressure. The CI sequence includes water introduction, and water requires a relatively long pumping time. For this reason, a pumping delay of 2 s was added at the beginning of the EI sequence making the sequence longer than in the pure EI mode.

The software can be programmed to allow an alternation between the two sequences, as previously described. Figure 8 shows the results obtained when programming the diluter to produce a 50%-50% mixture of the gases that originate from the alkanes/CO2 and VOC/N2 gas cylinders.

a)

b)

Figure 8: Alternation of the PTR-MS (a) and EI (b) ionization modes. For EI (b), only the major ions are plotted. The isobaric ions for m/z=44 are plotted in green, and the ions for m/z 19 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 26

29 are in blue. In EI mode, the major ions detected are N2+, CH4+ and CH3+. H2O+ and H3O+, which originate from the ionization of water present in the background gas. As expected, C3H7+, C3H8+, CO2+ and C4H10+ are also detected in very small quantities.

In the PTR-MS mode, H3O+ is the main ion detected. Five VOCs from the VOC/N2 cylinder are detected. Except for cyclohexanone, their intensities increase in the expected order of increasing rate coefficients. Two other ions are detected at m/z 57.075 and m/z 58.081, which are attributed to C4H9+ (57,070 u) and its 13C isotope (58.074 u). This attribution is confirmed by the 58/57 intensity ratio, and the mean experimental value of I(58)/I(57) is 0.055+/- 0.006. The theoretical value 0.044 based on the isotopic abundance is slightly lower.

In EI and CI, the ion intensities are roughly constant with time. The spectra observed in each mode are similar to those observed in the individual EI or CI ionization mode experiment. This experiment demonstrates the ability of our instrument to switch between two ionization modes. Then, we can take advantage of the two modes. For instance, the presence of butane as a significant component of our test mixture is supported by detection of C4H10+ ion in the EI spectrum. The CI spectrum of the same sample at the same time confirms this hypothesis. In the present work, these modes allow a satisfactory evaluation of the hydride abstraction rate constant under FT-ICR conditions so that in future work, this rate constant might be used for butane quantification at the % level.

Rate coefficient evaluation: Since m/z 57 is not detected in the CI spectrum of VOCs/N2 cylinder, the C4H9+ ions must come from the alkanes/CO2 cylinder and therefore from the butane compound. The only neutral component capable of producing C4H9+ by reaction with H3O+ is butane, through the following CI reaction: H3O+ + C4H10  C4H9+ + H2O +H2

(2)

This reaction is known to be very slow with reported rate coefficients varying from 0 to 3 10– 12

cm3 s–1 with 0.30 branching ratio43,44, i.e., 0.9 10–12 cm3 s–1 for C4H9+ formation45 under

SIFT-MS conditions. 20 ACS Paragon Plus Environment

Page 21 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Under our conditions, this rate coefficient kC4H10 can be estimated from the results shown in Figure 8b and the equation (1). Therefore, for any of the 5 VOCs present in the CI spectrum: I(C4H9+) / I(VOCH+) = kC4H10 CC4H10 / (kVOC CVOC)

(3)

The average temporal values of I(C4H9+) / I(VOCH+) obtained for each VOC lead to an estimation of kC4H10. These estimations are reasonably consistent, varying from 3.5 10–12 to 6.3 10–12 cm3 s-1. Averaging kC4H10 on the 5 VOCs and estimating the uncertainty as twice the standard deviation of kC4H10 finally gives: kC4H10 = 5.1 ± 2 10–12 cm3 s-1, which is slightly larger but of the same order of magnitude as the reported values. Studies of the H3O+ reactivity on a series of alkanes have shown that the rate coefficient of this hydride abstraction reaction increases with alkane size, which makes it potentially useful for larger alkane analysis.46,47

If more accuracy is needed, the rate constant can be measured using a sequence increasing the Pxt values by changing the opening time of the valve.

CONCLUSIONS

A new compact FT-ICR mass spectrometer was built to improve gas analysis in different gaseous matrixes (air, methane, etc.). This instrument can also be used for kinetic studies or spectroscopic characterization of the ions. The use of a 1.5 T permanent magnet enhanced the detection limit down to 200 ppb without accumulation. Two ionization modes were implemented: CI and EI. The CI mode, applied using H3O+ precursor ions, is well suited for VOC detection with a detection limit of a few tens ppb when accumulating the spectra. EI provides complementary information on other gases that are present at much higher concentration (%) and not detectable by CI. In both cases, the FT-ICR mass resolution enables to attribute a molecular formula to the ions observed and separate the quasi-isobaric ions. In our instrument, it is easy to alternate between the two modes and the number of sample lines is not limited. Therefore, switching between ionization modes, CI anionic or cationic precursors and between different samples facilitates the real-time analysis of complex gas mixtures. 21 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 26

ACKNOWLEDGMENTS

The authors wish to thank Laurent Clochard for the magnet design, David Robin for the mechanical design and assembly and Philippe Gremillet for the software development.

22 ACS Paragon Plus Environment

Page 23 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

BIBLIOGRAPHY (1) Smith, J. N.; Keil, A.; Likens, J.; Noll, R. J.; Cooks, R. G. Facility Monitoring of Toxic Industrial Compounds in Air Using an Automated, Fieldable, Miniature Mass Spectrometer. The Analyst 2010, 135 (5), 994. (2) Duarte, K.; Justino, C. I. L.; Freitas, A. C.; Duarte, A. C.; Rocha-Santos, T. A. P. DirectReading Methods for Analysis of Volatile Organic Compounds and Nanoparticles in Workplace Air. TrAC Trends Anal. Chem. 2014, 53, 21–32. (3) Dewulf, J.; Van Langenhove, H.; Wittmann, G. Analysis of Volatile Organic Compounds Using Gas Chromatography. TrAC Trends Anal. Chem. 2002, 21 (9), 637– 646. (4) Qi, M.; Ge, X.; Liang, M.; Fu, R. Flash Gas Chromatography for Analysis of Volatile Compounds from Houttuynia Cordata Thunb. Anal. Chim. Acta 2004, 527 (1), 69–72. (5) Beck, J. J.; Porter, N.; Cook, D.; Gee, W. S.; Griffith, C. M.; Rands, A. D.; Truong, T. V.; Smith, L.; San Román, I. In-Field Volatile Analysis Employing a Hand-Held Portable GC-MS: Emission Profiles Differentiate Damaged and Undamaged Yellow Starthistle Flower Heads: Discrimination of Damaged and Undamaged Flower Heads by Portable GC-MS. Phytochem. Anal. 2015, 26 (6), 395–403. (6) Tuzson, B.; Jagerska, J.; Looser, H.; Graf, M.; Felder, F.; Fill, M.; Tappy, L.; Emrnenegger, L. Highly Selective Volatile Organic Compounds Breath Analysis Using a Broadly-Tunable Vertical-External-Cavity Surface-Emitting Laser. Anal. Chem. 2017, 89 (12), 6377–6383. (7) Zhang, X.; Krechmer, J. E.; Groessl, M.; Xu, W.; Graf, S.; Cubison, M.; Jayne, J. T.; Jimenez, J. L.; Worsnop, D. R.; Canagaratna, M. R. A Novel Framework for Molecular Characterization of Atmospherically Relevant Organic Compounds Based on Collision Cross Section and Mass-to-Charge Ratio. Atmospheric Chem. Phys. 2016, 16 (20), 12945–12959. (8) Biasioli, F.; Yeretzian, C.; Märk, T. D.; Dewulf, J.; Van Langenhove, H. Direct-Injection Mass Spectrometry Adds the Time Dimension to (B)VOC Analysis. TrAC Trends Anal. Chem. 2011, 30 (7), 1003–1017. (9) Ellis, A. M.; Mayhew, C. A. Proton Transfer Reaction Mass Spectrometry; John Wiley & Sons, Ltd, 2014. (10) Smith, D.; Španěl, P. The SIFT and FALP Techniques; Applications to Ionic and Electronic Reactions Studies and Their Evolution to the SIFT-MS and FA-MS Analytical Methods. Int. J. Mass Spectrom. 2015, 377, 467–478. (11) Sekimoto, K.; Li, S.-M.; Yuan, B.; Koss, A.; Coggon, M.; Warneke, C.; de Gouw, J. Calculation of the Sensitivity of Proton-Transfer-Reaction Mass Spectrometry (PTRMS) for Organic Trace Gases Using Molecular Properties. Int. J. Mass Spectrom. 2017, 421, 71–94. (12) Latappy, H.; Lemaire, J.; Heninger, M.; Louarn, E.; Bauchard, E.; Mestdagh, H. Protonated 1,4-Difluorobenzene C6H5F2+: A Promising Precursor for Proton-Transfer Chemical Ionization. Int. J. Mass Spectrom. 2016, 405, 13–23. (13) Smith, D.; Sovová, K.; Španěl, P. A Selected Ion Flow Tube Study of the Reactions of H3O+, NO+ and O2+• with Seven Isomers of Hexanol in Support of SIFT-MS. Int. J. Mass Spectrom. 2012, 319–320, 25–30. (14) Dehon, C.; Lemaire, J.; Heninger, M.; Chaput, A.; Mestdagh, H. Chemical Ionization Using CF3+: Efficient Detection of Small Alkanes and Fluorocarbons. Int. J. Mass Spectrom. 2011, 299 (2–3), 113–119. 23 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 26

(15) Mayhew, C. A.; Peverall, R.; Timperley, C. M.; Watts, P. A Selected Ion Flow Tube Study of the Reactions of OH− with a Number of Fully and Partially Halogenated Methanes. Int. J. Mass Spectrom. 2004, 233 (1–3), 155–163. (16) Hera, D.; Langford, V.; McEwan, M.; McKellar, T.; Milligan, D. Negative Reagent Ions for Real Time Detection Using SIFT-MS. Environments 2017, 4 (1), 16. (17) Le Vot, C.; Lemaire, J.; Pernot, P.; Heninger, M.; Mestdagh, H.; Louarn, E. Oxygen Anion (O−) and Hydroxide Anion (HO−) Reactivity with a Series of Old and New Refrigerants. J. Mass Spectrom. 2018, 53 (4), 336–352. (18) Morrison, G. C.; Howard, C. J. Selective Detection of Gas-Phase Aldehydes and Ketones Using Protonated Hydrazine. Int. J. Mass Spectrom. 2001, 210, 503–509. (19) Custer, T. G.; Kato, S.; Bierbaum, V. M.; Howard, C. J.; Morrison, G. C. Gas-Phase Kinetics and Mechanism of the Reactions of Protonated Hydrazine with Carbonyl Compounds. Gas-Phase Hydrazone Formation: Kinetics and Mechanism. J. Am. Chem. Soc. 2004, 126 (9), 2744–2754. (20) Sulzer, P.; Agarwal, B.; Jürschik, S.; Lanza, M.; Jordan, A.; Hartungen, E.; Hanel, G.; Märk, L.; Märk, T. D.; González-Méndez, R.; et al. Applications of Switching Reagent Ions in Proton Transfer Reaction Mass Spectrometric Instruments for the Improved Selectivity of Explosive Compounds. Int. J. Mass Spectrom. 2013, 354–355, 123–128. (21) Lanza, M.; Acton, W. J.; Sulzer, P.; Breiev, K.; Juerschik, S.; Jordan, A.; Hartungen, E.; Hanel, G.; Maerk, L.; Maerk, T. D.; et al. Selective Reagent Ionisation-Time of FlightMass Spectrometry: A Rapid Technology for the Novel Analysis of Blends of New Psychoactive Substances. J. Mass Spectrom. 2015, 50 (2), 427–431. (22) Müller, M.; Mikoviny, T.; Feil, S.; Haidacher, S.; Hanel, G.; Hartungen, E.; Jordan, A.; Märk, L.; Mutschlechner, P.; Schottkowsky, R.; et al. A Compact PTR-ToF-MS Instrument for Airborne Measurements of Volatile Organic Compounds at High Spatiotemporal Resolution. Atmospheric Meas. Tech. 2014, 7 (11), 3763–3772. (23) Murschell, T.; Fulgham, S. R.; Farmer, D. K. Gas-Phase Pesticide Measurement Using Iodide Ionization Time-of-Flight Mass Spectrometry. Atmospheric Meas. Tech. 2017, 10 (6), 2117–2127. (24) Zeller, L. C.; Kennady, J. M.; Campana, J. E.; Kenttamaa, H. Characterization of a Small FTICR Mass Spectrometer Based on a Permanent Magnet. Anal. Chem. 1993, 65 (15), 2116–2118. (25) Yamazaki, H.; Tsuneda, M.; Takakuwa, Y.; Honda, Y.; Nukanobu, A.; Sakurai, H.; Nakagawa, K. A New Compact Size, Precision Mass Spectrometer by Ion Cyclotron Resonance. J. Mass Spectrom. Soc. Jpn. 1994, 42 (2), 105–115. (26) Halbach, K. Design of Permanent Multipole Magnets with Oriented Rare Earth Cobalt Material. Nucl. Instrum. Methods 1980, 169 (1), 1–10. (27) Mauclaire, G.; Lemaire, J.; Boissel, P.; Bellec, G.; Heninger, M. MICRA: A Compact Permanent Magnet Fourier Transform Ion Cyclotron Resonance Mass Spectrometer. Eur. J. Mass Spectrom. 2004, 10 (2), 155–162. (28) Maitre, P.; Le Caer, S.; Simon, A.; Jones, W.; Lemaire, J.; Mestdagh, H. N.; Heninger, M.; Mauclaire, G.; Boissel, P.; Prazeres, R.; et al. Ultrasensitive Spectroscopy of Ionic Reactive Intermediates in the Gas Phase Performed with the First Coupling of an IR FEL with an FTICR-MS. Nucl. Instrum. Methods Phys. Res. Sect. -Accel. Spectrometers Detect. Assoc. Equip. 2003, 507 (1–2), 541–546. (29) Romanzin, C.; Louarn, E.; Lemaire, J.; Žabka, J.; Polášek, M.; Guillemin, J.-C.; Alcaraz, C. An Experimental Study of the Reactivity of CN− and C3N− Anions with Cyanoacetylene (HC3N). Icarus 2016, 268, 242–252. (30) Rimkus, W. V.; Davis, D. V.; Gallaher, K. Learn FTMS Fundamentals and Quantitation on an Inexpensive Rugged Machine—Save Your High Powered FTMS for the 24 ACS Paragon Plus Environment

Page 25 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

(31) (32)

(33)

(34)

(35)

(36)

(37)

(38)

(39)

(40)

(41) (42) (43)

(44)

(45) (46)

Experienced User. In Proceedings of the 53rd ASMS Conference on Mass Spectrometry and Allied Topics; San Antonio, TX, 2005. Littlejohn, D. P.; Schubert, W. P. Method for Forming a Seat in a Pulsed Sampling Valve. US6355929 B1, March 12, 2002. Vilkov, A. N.; Gamage, C. M.; Misharin, A. S.; Doroshenko, V. M.; Tolmachev, D. A.; Tarasova, I. A.; Kharybin, O. N.; Novoselov, K. P.; Gorshkov, M. V. Atmospheric Pressure Ionization Permanent Magnet Fourier Transform Ion Cyclotron Resonance Mass Spectrometry. J. Am. Soc. Mass Spectrom. 2007, 18 (8), 1552–1558. Le Vot, C.; Bouaziz, M.; Heninger, M.; Boissel, P.; Mauclaire, G.; Mestdagh, H.; Da Costa, F.; Lemaire, J. Development of a Transportable FT-ICR MS Associated with a Glow Discharge Ionization Source. Int. J. Mass Spectrom. 2016, 407, 106–112. Sarrabi, S.; Colin, X.; Tcharkhtchi, A.; Heninger, M.; Leprovost, J.; Mestdagh, H. Real Time Analysis of Volatile Organic Compounds from Polypropylene Thermal Oxidation Using Chemical Ionization Fourier Transform Ion Cyclotron Resonance Mass Spectrometry. Anal. Chem. 2009, 81 (15), 6013–6020. Francis, G. J.; Wilson, P. F.; Milligan, D. B.; Langford, V. S.; McEwan, M. J. GeoVOC: A SIFT-MS Method for the Analysis of Small Linear Hydrocarbons of Relevance to Oil Exploration. Int. J. Mass Spectrom. 2007, 268 (1), 38–46. Lemaire, J.; Thomas, S.; Lopes, A.; Louarn, E.; Mestdagh, H.; Latappy, H.; Leprovost, J.; Heninger, M. Compact FTICR Mass Spectrometry for Real Time Monitoring of Volatile Organic Compounds. 2018, 15. Louarn, E.; Hamrouni, A.; Colbeau-Justin, C.; Bruschi, L.; Lemaire, J.; Heninger, M.; Mestdagh, H. Characterization of a Membrane Inlet Interfaced with a Compact Chemical Ionization FT-ICR for Real-Time and Quantitative VOC Analysis in Water. Int. J. Mass Spectrom. 2013, 353, 26–35. Grosshans, P. B.; Chen, R.; Limbach, P. A.; Marshall, A. G. Linear Excitation and Detection in Fourier Transform Ion Cyclotron Resonance Mass Spectrometry. Int. J. Mass Spectrom. Ion Process. 1994, 139, 169–189. Mauclaire, G.; Lemaire, J.; Heninger, M. Sharing a Pair of Electrodes between Excitation and Detection in a Fourier Transform Ion Cyclotron Resonance Mass Spectrometer. Rapid Commun. Mass Spectrom. 1996, 10 (5), 591–593. Chiper, A. S.; Blin-Simiand, N.; Heninger, M.; Mestdagh, H.; Boissel, P.; Jorand, F.; Lemaire, J.; Leprovost, J.; Pasquiers, S.; Popa, G.; et al. Detailed Characterization of 2Heptanone Conversion by Dielectric Barrier Discharge in N 2 and N 2 /O 2 Mixtures. J. Phys. Chem. A 2010, 114 (1), 397–407. Nibbering, N. M. Gas-Phase Ion/Molecule Reactions as Studied by Fourier Transform Ion Cyclotron Resonance. Acc. Chem. Res. 1990, 23 (9), 279–285. Heninger, M.; Pernot, P.; Mestdagh, H. Influence of the Internal Excitation of FeCO on Its Reactivity with O. Chem. Phys. Lett. 2001, 333 (358), 364. Španěl, P.; Smith, D. Selected Ion Flow Tube Studies of the Reactions of H 3 O+, NO+, and O 2+ with Several Aromatic and Aliphatic Hydrocarbons. Int. J. Mass Spectrom. 1998, 181 (1), 1–10. Francis, G. J.; Wilson, P. F.; Milligan, D. B.; Langford, V. S.; McEwan, M. J. GeoVOC: A SIFT-MS Method for the Analysis of Small Linear Hydrocarbons of Relevance to Oil Exploration. Int. J. Mass Spectrom. 2007, 268 (1), 38–46. Wilson, P. F.; Freeman, C. G.; McEwan, M. J. Reactions of Small Hydrocarbons with H3O+, O2+ and NO+ Ions. Int. J. Mass Spectrom. 2003, 229 (3), 143–149. Arnold, S. T.; Viggiano, A. A.; Morris, R. A. Rate Constants and Product Branching Fractions for the Reactions of H 3 O + and NO + with C 2 −C 12 Alkanes. J. Phys. Chem. A 1998, 102 (45), 8881–8887. 25 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 26

(47) Amador-Muñoz, O.; Misztal, P. K.; Weber, R.; Worton, D. R.; Zhang, H.; Drozd, G.; Goldstein, A. H. Sensitive Detection of N-Alkanes Using a Mixed Ionization Mode Proton-Transfer-Reaction Mass Spectrometer. Atmospheric Meas. Tech. 2016, 9 (11), 5315–5329.

26 ACS Paragon Plus Environment