Nonuniform Electric Field-Enhanced In-Source Declustering in High-Pressure Photoionization/Photoionization-Induced Chemical Ionization Mass Spectrometry for Operando Catalytic Reaction Monitoring

Article abstract of DOI:10.1021/acs.analchem.0c04081

Photoionization mass spectrometry (PI-MS) is a powerful and highly sensitive analytical technique for online monitoring of volatile organic compounds (VOCs). However, due to the large difference of PI cross sections for different compounds and the limitation of photon energy, the ability of lamp-based PI-MS for detection of compounds with low PI cross sections and high ionization energies (IEs) is insufficient. Although the ion production rate can be improved by elevating the ion source pressure, the problem of generating plenty of cluster ions, such as [MH]+·(H2O)n (n = 1 and 2) and [M2]+, needs be solved. In this work, we developed a new nonuniform electric field high-pressure photoionization/photoionization-induced chemical ionization (NEF-HPPI/PICI) source with the abilities of both HPPI and PICI, which was accomplished through ion-molecule reactions with high-intensity H3O+ reactant ions generated by photoelectron ionization (PEI) of water molecules. By establishing a nonuniform electric field in a three-zone ionization region to enhance in-source declustering and using 99.999% helium as the carrier gas, not only the formation of cluster ions was significantly diminished, but the ion transmission efficiency was also improved. Consequently, the main characteristic ion for each analyte both in HPPI and PICI occupied more than 80%, especially [HCOOH·H]+ with a yield ratio of 99.2% for formic acid. The analytical capacity of this system was demonstrated by operando monitoring the hydrocarbons and oxygenated VOC products during the methanol-to-olefins and methane conversion catalytic reaction processes, exhibiting wide potential applications in process monitoring, reaction mechanism research, and online quality control.

Full text of DOI:10.1021/acs.analchem.0c04081

pubs.acs.org/ac
Article
Nonuniform Electric Field-Enhanced In-Source Declustering in High-
Pressure Photoionization/Photoionization-Induced Chemical
Ionization Mass Spectrometry for Operando Catalytic Reaction
Monitoring
Ningbo Wan, Jichun Jiang, Fan Hu, Ping Chen, Kaixin Zhu, Dehui Deng, Yuanyuan Xie, Chenxin Wu,
Lei Hua,* and Haiyang Li*
Cite This: Anal. Chem. 2021, 93, 2207−2214
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Photoionization mass spectrometry (PI-MS) is a
powerful and highly sensitive analytical technique for online
monitoring of volatile organic compounds (VOCs). However, due
to the large difference of PI cross sections for different compounds
and the limitation of photon energy, the ability of lamp-based PI-
MS for detection of compounds with low PI cross sections and
high ionization energies (IEs) is insufficient. Although the ion
production rate can be improved by elevating the ion source
pressure, the problem of generating plenty of cluster ions, such as
[MH]+·(H2O)n (n = 1 and 2) and [M2]+, needs be solved. In this
work, we developed a new nonuniform electric field high-pressure
photoionization/photoionization-induced chemical ionization (NEF-HPPI/PICI) source with the abilities of both HPPI and PICI,
which was accomplished through ion−molecule reactions with high-intensity H3O+ reactant ions generated by photoelectron
ionization (PEI) of water molecules. By establishing a nonuniform electric field in a three-zone ionization region to enhance in-
source declustering and using 99.999% helium as the carrier gas, not only the formation of cluster ions was significantly diminished,
but the ion transmission efficiency was also improved. Consequently, the main characteristic ion for each analyte both in HPPI and
PICI occupied more than 80%, especially [HCOOH·H]+ with a yield ratio of 99.2% for formic acid. The analytical capacity of this
system was demonstrated by operando monitoring the hydrocarbons and oxygenated VOC products during the methanol-to-olefins
and methane conversion catalytic reaction processes, exhibiting wide potential applications in process monitoring, reaction
mechanism research, and online quality control.
of the ionization process for a certain photon energy and is
hoton ionization mass spectrometry (PI-MS) is a powerful
Pand highly sensitive technique for online and real-time usually measured in megabarn (Mb), whereby 1 Mb equals
10−18 cm2. The σPI value of VOCs of different classes varies
trace gas analysis of volatile organic compounds (VOCs). The
largely, which differs by more than 2 orders of magnitude.15
applications of PI-MS have been found in various fields, such
as real-time process monitoring,1−5 environment research,6,7
Unsaturated hydrocarbons usually exhibit relatively larger σPI,
food science,8,9 and breath analysis10−12 during the past few
such as 22 Mb for benzene and 24 Mb for propyne at a photon
energy of 10.49 eV. Meanwhile, the σPI values are rather small
for some oxygenated VOCs (OVOCs), especially for those
lower-mass molecules, such as 5 Mb for propanol at a photon
energy of 10.49 eV and no more than 0.2 Mb for methanol
when the photon energy is less than 10.8 eV.16 Thus, the
ionization efficiency and detection sensitivity of these
decades. The photon energies of the commonly used vacuum
ultraviolet krypton (VUV-Kr) discharge lamp in PI-MS are
10.0 and 10.6 eV, which are just a little higher than or equal to
the ionization energies (IEs) of most VOCs,13 leading to a
high yield of molecular ions and very limited degree of
fragmentation.14 The rather soft ionization process is advanta-
geous for online and direct MS analysis of complex mixtures
owing to the easily interpreted mass spectra without excessive
overlap of fragment peaks.
The ion production rate RPI (ions/s) of any compound
produced in a PI process is proportional to the PI cross section
σPI of the molecule, the photon energy, and light intensity of
the VUV light source.10 The σPI is a measure of the probability
Received: September 28, 2020
Accepted: December 17, 2020
Published: January 7, 2021
https://dx.doi.org/10.1021/acs.analchem.0c04081
© 2021 American Chemical Society
Anal. Chem. 2021, 93, 2207−2214
2207

Analytical Chemistry
pubs.acs.org/ac
Article
compounds by PI-MS were not satisfactory. In addition,
analytes with IEs higher than the photon energy (10.6 eV for a
Kr lamp), e.g., methanol (IE = 10.84 eV) and formic acid (IE =
11.33 eV),17 could not be effectively ionized and detected,
which obviously limits the practical application of PI-MS.
Efforts have been taken to develop novel PI sources to
improve the ability of PI-MS for detection of compounds with
low σPI and high IEs. Zimmermann et al.18,19 have designed a
combined ion source by coupling a conventional 70 eV
electron ionization (EI) module to a single-photon ionization
(SPI) source in which both SPI and EI mass spectra of the
analytes were obtained, and analyte molecules with IEs higher
than the photon energy could be effectively ionized and
detected through 70 eV EI. However, the ion source pressure
must be kept low enough (below 10−2 Pa) to protect the
heated filament, which resulted in low molecular density of
analytes in the ionization region and consequently limited
further improvement of the sensitivity. Zenobi et al.20 and Li et
al.12,21,22 presented a photoelectron ionization (PEI) source
based on a VUV-Kr lamp. VUV light-induced photoelectron
emission was used as an alternative source of electrons for EI
without the limitations of a heated filament. The PEI source
could work at relatively high pressure.
hydrated cluster ions H3O+·(H2O)n (n = 1 and 2) when
dealing with high-humidity samples.11,23,27 This will not only
increase the complexity of the mass spectra but also affect the
detection sensitivity of the analytes as the reaction capacities
for the reactions of H3O+·(H2O)n (n = 0−2) with analyte
molecules decreased as the number of water molecules (n)
increased. Bohme et al.30 studied the gas-phase kinetics of
reactions with H3O+·(H2O)n (n = 0−2) by the flowing
afterglow experiments at 300 K, which indicated that the rate
coefficients for the reactions of HCOOH (for example) with
H3O+, H3O+·H2O, and H3O+·(H2O)2 were 2.7 0.8, 2.4
0.7, and 1.5
0.5 (10−9 cm3 molecule−1 s−1), respectively.
Collision-induced dissociation (CID) of a cluster ion with
neutral molecules by increasing the kinetic energy of the ion is
a direct and effective method to get its core ion. In HPPI-MS,
efforts have been taken to realize in-source CID to dissociate
some cluster ions by elevating the ion source voltage to form a
high-intensity uniform electric field throughout the ionization
region.31−33 In PTR-MS, the degree of cluster ion formation
was usually regulated by choosing an appropriate value of E/
N,34,35 where E was the electric field intensity in the drift tube
and N was the density of gas molecules in the drift tube.
However, a higher value of E/N led to a reduced ion−molecule
reaction time, which could decrease the sensitivity of PTR-MS.
Additionally, the ion transmission efficiency might degrade
under excessive intensity of the uniform electric field. Even so,
in the analysis of high-humidity samples,10,36,37 such as exhaled
breath,10 cluster ions [MH]+·(H2O)n (n = 1 and 2) still existed
in the mass spectra at relatively high values of the electric field
intensity.
In this study, a nonuniform electric field high-pressure
photoionization/photoionization-induced chemical ionization
(NEF-HPPI/PICI) source with both HPPI and PICI
capabilities was developed for the TOFMS. HPPI was achieved
by using a VUV-Kr lamp, while PICI was accomplished
through ion−molecule reactions with high-intensity H3O+
reactant ions generated by PEI of water molecules. To
eliminate the effect of cluster ions, a nonuniform electric
field in a three-zone ion source was designed. The electric field
intensities of three zones as well as the type of carrier gas were
investigated. The performance of this NEF-HPPI/PICI source
in terms of declustering and ion beam focusing was also
evaluated. Finally, applications have been demonstrated by
operando monitoring the products during methane conversion
and methanol-to-olefins catalytic reaction processes.
Increasing the number density of analyte molecules in the
ionization region by elevating the ion source pressure is a
feasible way to effectively increase the ion production rate and
detection sensitivity. Li and co-workers10 have developed a
high-pressure photon ionization (HPPI) source for time-of-
flight mass spectrometry (TOFMS). By elevating the ion
source pressure to about 700 Pa and employing a radio
frequency (RF)-only quadrupole as the ion guide system, the
limits of detection (LODs) down to lower-ppbv (parts per
billion by volume) for some OVOCs, e.g., acetone and
isoprene, were achieved. As the elevation of ion source
pressure, ion−molecule reactions occur in the ion source,
which can be used as another channel for ionization of analyte
molecules. Li et al. developed a photoionization-generated
dibromomethane cation chemical ionization (PDCI) source
+
with an ion source pressure of 500 Pa, which utilized CH2Br2
as the reactant ions generated by using a VUV-Kr lamp to
achieve efficient ionization of volatile sulfur compounds
(VSCs) with high IEs.6 In 2018, they used PEI to generate
+
plenty of O2 ions as the reactant ion for chemical ionization
(CI) at kPa. Small n-alkanes with IEs higher than 10.6 eV
could be effectively ionized by O2+ and directly detected at the
sub-ppbv level.12 For OVOCs, the proton affinities (PAs) are
much higher compared with those of unsaturated hydro-
carbons. Thus, the proton transfer reaction (PTR) based on
the H3O+ (PA = 691 kJ/mol)17 reactant ion is suitable for
efficient ionization of OVOCs.23−25 Shu et al.11,26−29 reported
a VUV-excited and CH2Cl2/H2O-amplified ionization process
by introducing a gas mixture of CH2Cl2/H2O into the VUV
ion source to generate large amounts of H3O+ ions. Then, the
PTR between the H3O+ reactant ions and analyte molecules
occurred to produce analyte ions [MH]+ and [MH]+·H2O with
the ion source pressure in the range of 500−1300 Pa. The
signal intensities of some OVOCs in the PTR could be
amplified by more than 3 orders of magnitude in contrast to
those obtained with VUV-PI alone.
EXPERIMENTAL SECTION
■
VUV-Lamp-Based NEF-HPPI/PICI Source. The NEF-
HPPI/PICI source was composed of a VUV lamp and a three-
zone ionization region, as shown in Figure 1. A 10.6 eV VUV-
Kr lamp (PKS106, Cathodeon Ltd., Cambridge, U.K.) with a
photon flux of about 1 × 1012 photons/s was installed on the
top of the ionization region, which consisted of six concentric
stainless steel electrodes (SE1−SE6): a photoelectron
acceleration electrode (SE1: 6 mm i.d., 1 mm thick), a
photoelectron emission electrode (SE2: 3 mm i.d., 1 mm
thick), two identical transmission electrodes (SE3 and SE4: 12
mm i.d., 1 mm thick), an ion dissociation and focusing
electrode (SE5: 12 mm i.d., 1 mm thick), and a Skimmer
electrode (SE6: 1 mm i.d., 4 mm thick). Four 5 mm-thick
polyether−ether−ketone (PEEK) insulation annular washers
(16 mm i.d.) were used to separate the electrodes, except the
space between the SE5 and SE6 electrodes for excess neutral
However, as the elevation of ion source pressure for higher
sensitivity, cluster ions formed by ion−molecule reactions both
in the HPPI and CI sources were inevitable, such as [MH]+·
(H2O)n (n = 1 and 2) and [M2]+, especially the formation of
2208
https://dx.doi.org/10.1021/acs.analchem.0c04081
Anal. Chem. 2021, 93, 2207−2214

Analytical Chemistry
pubs.acs.org/ac
Article
center orifice of the Sampler electrode into a home-built radio
frequency (RF)-only quadrupole TOFMS for mass analysis.
The NEF-HPPI/PICI source could work at two different
operation modes, the HPPI mode and PICI mode, by simply
adjusting the two dc voltages, V1 and V2, respectively, and
simultaneously switching SC1 to corresponding gases. In the
HPPI mode, both SC1 and SC2 were used to introduce gas-
phase analytes, and the electric field intensity E1 between the
SE1 and SE2 electrodes was 8 V/cm. Thus, the photoelectrons
emitted from the surface of the SE2 electrode could not obtain
enough energy to ionize analyte molecules, and the ionization
process was dominated by PI. Meanwhile, in the PICI mode,
humidified He was selected by a switching valve to flow
through SC1 into the reactant ion production zone, and an
electric field intensity E1 of 250 V/cm allowed the photo-
electrons emitted from the surface of the SE2 electrode to get
sufficiently high energy to ionize He and H2O molecules.
Plenty of H3O+ reactant ions were produced in the reactant ion
production zone after a series of ion−molecule reactions and
transferred into the analyte ionization zone to react with the
analyte molecules by the PTR. The intensity of the H3O+
reactant ion could be regulated by adjusting the electric field
intensity E1. Compared with the previously reported HPPI ion
source, one of the major differences is the design of the ion
dissociation and focusing zone in the NEF-HPPI/PICI source,
which was employed to dissociate cluster ions and focus the
ion beam by elevating the electric field intensity E3 between
the SE5 and SE6 electrodes. The switching time between the
two modes was less than 1 s.
Figure 1. Cutaway diagram of the VUV-lamp-based nonuniform
electric field high-pressure photoionization/photoionization-induced
chemical ionization (NEF-HPPI/PICI) source (black line: PICI
mode; blue line: HPPI mode; MFC: mass flow controller).
exhaust. The SE2 and SE5 electrodes partitioned the ionization
region into three zones: a reactant ion production zone E1, an
analyte ionization zone E2, and an ion dissociation and
focusing zone E3. Gas-phase analytes and humidified helium
(He) were directly introduced into the ionization region
through two 250 μm i.d., 1 m long PEEK sampling capillaries
(SC1 and SC2) through which the sampling flow rate was
about 11 mL/min for each capillary. SC1 was used to
introduce humidified He or gas-phase analytes into the
reactant ion production zone, while SC2 was employed to
introduce only the gas-phase analytes into the analyte
ionization zone. Humidified He was obtained by bubbling of
11 mL/min high-purity He (99.999%) through purified water.
The electrodes from SE2 to SE4 were electrically connected by
using a 1 MΩ resistor string, and additionally, the SE5 was
connected by another 1 MΩ resistor to the earth. Four positive
dc voltages were individually applied to the SE1 (V1), SE2
(V2), SE5 (V3), and SE6 (V4) electrodes. The ions generated
in the ionization region were transferred through the 1 mm i.d.
RF-Only Quadrupole TOFMS. The mass spectrometer
used in the experiment was a home-built RF-only quadrupole
orthogonal acceleration reflectron TOFMS (see the Support-
ing Information, S1). A mass resolving power of 5000 (full
width at half-maximum, fwhm) was achieved with a 0.5 m
field-free drift tube. All the mass spectra were accumulated for
10 s at a repetition rate of 25 kHz, and all data were obtained
by averaging results from six parallel measurements.
Sample Preparation. The introduction of the sample
preparation method is shown in the Supporting Information,
S2.
Figure 2. Mass spectra of benzene (0.5 ppmv), p-xylene (0.5 ppmv), tetrachloroethylene (0.5 ppmv), hexachloro-1,3-butadiene (0.5 ppmv),
methanol (0.75 ppmv), and formic acid (6 ppmv) in the HPPI mode (a) and PICI mode (b), respectively. (c) Yield ratios of cluster ions [MH]+·
H2O at different ion source pressures in the PICI mode.
2209
https://dx.doi.org/10.1021/acs.analchem.0c04081
Anal. Chem. 2021, 93, 2207−2214

Analytical Chemistry
pubs.acs.org/ac
Article
Figure 3. Mass spectra of 0.25 ppmv methanol and 5 ppmv formic acid in the carrier gas of 99.999% N2 (a) and 99.999% He (b), respectively. (c)
Yield ratios of [MH]+·(H2O)n (n = 0 and 1) of methanol and formic acid in the carrier gas of 99.999% N2 and 99.999% He, respectively (star:
reactant ions H3O+·(H2O)n (n = 0, 1, and 2)).
carrier gas and optimizing the electric field in the ionization
region were investigated in the following experiments.
RESULTS AND DISCUSSION
■
Ionization of VOCs in the HPPI and PICI Modes. Figure
Influence of Different Carrier Gases in the PICI Mode.
To investigate the influence of different carrier gases on the
product ion distribution in the PICI mode, the mass spectra of
0.25 ppmv methanol and 5 ppmv formic acid in the carrier gas
of 99.999% N2 and 99.999% He with the ion source pressure of
both 300 Pa were acquired, respectively, as shown in Figure 3.
On the one hand, there were less cluster ions formed with
carrier gas He. Since the cluster formation was an exothermic
reaction, more collisions between the ions and neutral
molecules were required by using the lower-mass molecules
(He) as the collision gas to release the excess energy from the
cluster formation process.38 In addition, the uptake of the
collision energy into the N2 internal energy might also
contribute to the cluster formation in the N2 carrier gas.
Thus, the formation of cluster ions was more difficult in the He
atmosphere than in N2. On the other hand, the signal
intensities of the analytes in He gas were almost 1 order of
magnitude higher than those in N2 gas. This can be attributed
to the higher ion transmission efficiency in the ion source and
the RF-only quadrupole region with carrier gas He, especially
for the analyte ions with relatively low m/z, such as [CH3OH·
H]+ (m/z 33) and [HCOOH·H]+ (m/z 47). According to the
result of an ion trajectory simulation by SIMION 8.1, the ion
transmission efficiency with He gas in the ion source was about
1.5 times higher than that with N2 gas, while more than 4 times
was achieved in the RF-only quadrupole region (see the
Supporting Information, S3). Therefore, high-purity He was
employed as the carrier gas of the analytes in this work.
Declustering and Focusing Effect of Nonuniform
Electric Field. To achieve in-source CID by elevating the
electric field intensity uniformly in the whole ionization region
would result in the decrease of ionization efficiency (see the
Supporting Information, S4). Therefore, we made a first
attempt to integrate an ion dissociation and focusing zone into
the NEF-HPPI/PICI source to solve this issue. An ion
trajectory calculation was carried out by SIMION 8.1 software
to investigate the effect of electric field uniformity in the
analyte ionization zone E2 and the ion dissociation and
focusing zone E3. Five hundred ions with m/z 33 ([CH3OH·
H]+) and 500 ions with m/z 47 ([HCOOH·H]+) were defined
in a circular plane (Φ = 2 mm) right below the central hole of
the photoelectron emission electrode in the analyte ionization
2 illustrates the mass spectra of a mixed gas sample of benzene
(0.5 ppmv), p-xylene (0.5 ppmv), tetrachloroethylene (0.5
ppmv), hexachloro-1,3-butadiene (0.5 ppmv), methanol (0.75
ppmv), and formic acid (6 ppmv) in 99.999% N2 gas in the
HPPI and PICI modes, respectively. The electric field
intensities of E1, E2, and E3 were 250, 10, and 45 V/cm,
respectively. The signal intensities of data with m/z greater
than 75, as shown in Figure 2a,b, were magnified by 10 times.
As shown in Figure 2a, only the molecular ion peaks of
benzene, p-xylene, tetrachloroethylene, and hexachloro-1,3-
butadiene appeared in the HPPI mass spectrum, besides the
+
fragment ion peak of p-xylene at m/z 91 (C7H7 ). As the IEs of
methanol (IE = 10.84 eV) and formic acid (IE = 11.33 eV)17
were higher than the photon energy (10.6 eV), they were hard
to be effectively ionized by HPPI (only 4 counts for methanol
and 11 counts for formic acid). In the PICI mode, the OVOCs,
methanol, and formic acid could be efficiently ionized by the
ion−molecule reaction with the reactant ions H3O+·(H2O)n (n
= 0, 1, and 2), producing significant mass peaks of analyte ions
[MH]+·(H2O)n (n = 0, 1, and 2) in the PICI mass spectrum.
The characteristic ions for the hydrocarbons in Figure 2b are
molecular ions M+ or protonated molecular ions [MH]+.
However, the signal intensities of benzene, p-xylene,
tetrachloroethylene, and hexachloro-1,3-butadiene in the
PICI mode were much lower than those in the HPPI mode,
which indicated that these two modes were suitable for
different classes of compounds. The HPPI mode was powerful
in ionization of hydrocarbons, while higher sensitivity for
OVOCs can be achieved in the PICI mode. Therefore, the
coupling of HPPI and PICI modes in the ion source could be a
versatile ionization technique for efficient soft ionization of
multiple VOC species.
A major problem for the NEF-HPPI/PICI source was that
the excess hydrated cluster ions, H3O+·(H2O)n (n = 1 and 2)
and [MH]+·(H2O)n (n = 1 and 2), formed in the PICI mode,
which not only increased the complexity of the mass spectra
but also reduced the ionization efficiency of analytes with the
reactant ion (H3O+·(H2O)n (n = 1 and 2)).30 What is more,
the degree of cluster ion formation increased as the elevation
of the ion source pressure for higher sensitivity, as shown in
Figure 2c. Thus, the methods for declustering by changing the
2210
https://dx.doi.org/10.1021/acs.analchem.0c04081
Anal. Chem. 2021, 93, 2207−2214

Analytical Chemistry
pubs.acs.org/ac
Article
Figure 4. Ion trajectories calculated by SIMION 8.1 software in the elevated uniform electric field (a) and the E3-enhanced nonuniform electric
field (b). Comparison of the total ion intensities (c) and the ratios of [MH]+ (d) for methanol and formic acid in the elevated uniform electric field
and E3-enhanced nonuniform electric field.
Figure 5. (a) Signal intensities of H318O+ reactant ions and the ratio of [MH]+/H318O+ for methanol and formic acid at different electric field
intensities of E1. (b) Peak area of H318O+ and [CH3OH·H]+ ions as a function of monitoring time in the PICI mode. (c−f) Time course of relative
intensities of four OVOC products during the methane conversion process. (g) Possible reaction path for methane oxidation.
zone. In the calculation, He was selected as the collision gas,
and the ion source pressure was set at 300 Pa. Figure 4a and
Figure 4b illustrate the ion trajectories and electric field
distributions in the ionization zones with an elevated uniform
electric field and E3-enhanced nonuniform electric field,
respectively. To realize in-source CID, the electric field
intensities of both E2 and E3 were elevated to 50 V/cm in
the elevated uniform electric filed, while only E3 was enhanced
to 50 V/cm and E2 remained at a relatively low intensity of 10
V/cm in the E3-enhanced nonuniform electric field. The drift
time of ions transmitting through the E2 and E3 zones in the
E3-enhanced nonuniform electrical field was about 2.5 times as
long as that in the elevated uniform electrical field, which
means a longer ion−molecule reaction time and higher PICI
ionization efficiency in the E3-enhanced nonuniform electrical
field. Furthermore, it is obvious that the ion trajectories
exhibited a focusing trend in Figure 4b, and the ion
transmission efficiency was calculated to be about 12% in the
E3-enhanced nonuniform electrical field, which was almost 2
times as much as that in the elevated uniform electrical field
(about 7%). A mixed gas sample of 0.25 ppmv methanol and 5
ppmv formic acid gas in 99.999% He was taken as an example
to explore the actual difference between the configurations of
the elevated uniform electric field and E3-enhanced nonuni-
form electric field. Similar to the calculation results by
SIMION 8.1, the total ion intensities of methanol and formic
acid in the E3-enhanced nonuniform electric field were 2.8 and
9.4 times as much as those in the elevated uniform electric
field, respectively (Figure 4c). Although only E3 was enhanced,
the degree of cluster ion formation in the E3-enhanced
nonuniform electric field was even lower than that in the
elevated uniform electric field, as shown in Figure 4d. More
details about the influence of the electric field intensity E3 on
detection sensitivity and cluster ion dissociation are shown in
2211
https://dx.doi.org/10.1021/acs.analchem.0c04081
Anal. Chem. 2021, 93, 2207−2214

Analytical Chemistry
pubs.acs.org/ac
Article
Figure 6. (a−c) Mass spectra of 890 ppm n-pentene at different electric field intensities of E3 in the HPPI mode. (d) Time course of reaction
products in the outflow gas during the initial stage of the methanol-to-olefins reaction. (e) Mass spectrum of the light olefin products at the reaction
time of the ninth second.
the Supporting Information, S5 and S6. The degree of cluster
ion dissociation and the detection sensitivity can be regulated
flexibly by adjusting the value of E3.
H316O+ was saturated) and the electric field intensity of E1,
while the almost unchanged blue curves represent the variation
trends of intensity ratios of [MH]+ to H318O+ for 0.25 ppmv
methanol and 5 ppmv formic acid as a function of the electric
field intensity of E1. Obviously, the signal intensities of
[CH3OH·H]+ and [HCOOH·H]+ were proportional to those
of the H3O+ reactant ions. Thus, it is an effective solution to
improve the stability of the instrument by adjusting the electric
field intensity of E1 to eliminate the fluctuation of the signal
intensities of the H3O+ reactant ion and analyte ions in the
PICI mode. Therefore, a self-adjustment algorithm was used to
stabilize the intensity of the H3O+ reactant ions by automati-
cally adjusting the value of E1. Figure 5b displays the peak area
variations of H318O+ ions and [CH3OH·H]+ ions of 0.25 ppmv
methanol as a function of monitoring time after using the
algorithm. As a result, the relative standard deviations (RSDs)
of H318O+ and [CH3OH·H]+ were only 2.4 and 3.5%,
respectively.
Performance of NEF-HPPI/PICI TOFMS. The analytical
performance of the NEF-HPPI/PICI TOFMS for several
hydrocarbons and OVOCs is summarized in the Supporting
Information, S7. In the HPPI mode, the enhancement factors
(ratio of response for one ionization mode over another) of
HPPI over PICI toward benzene, p-xylene, tetrachloro-
ethylene, and hexachloro-1,2-butadiene were determined to
be 6-, 9-, 7-, and 10-fold, respectively. Meanwhile, in the PICI
mode, the enhancement factors of PICI over HPPI reveal the
amplification of signal intensities from 3-fold to more than 3
orders of magnitude (1885-fold) for methanol, formic acid,
butanone, and diethyl ether. The LODs (S/N = 3) ranged
from 0.17 to 13 ppbv (see Table S1). The linear ranges of 2−3
orders of magnitude were achieved, and the correlation
coefficients were in the range of 0.9897−0.9993 (see Figure
S6).
Details of the operation conditions for the methane
conversion reaction are described in the Supporting
Information, S8. The OVOC products from the reaction
solution carried by 99.999% He were detected in the PICI
mode. The electric field intensity of E3 was elevated to a
higher level (133 V/cm) in this experiment to eliminate the
cluster ion [HCOOH·H]+·H3O of the product HCOOH,
whose peak completely coincided with the protonated ion
[HOCH2OOH·H]+ of product HOCH2OOH in the mass
spectra. The catalytic reaction was continuously monitored in
operando for approximately 4 h, and the results are shown in
Figure 5c−f. During the reaction, the intensities of CH3OH
and CH3OOH increased gradually over time, and the
increasing rate of CH3OH gradually decreased after 150 min,
while the increasing rate of HOCH2OOH and HCOOH
increased significantly in the last 90 min. This variation
tendency was consistent with another similar catalyst,2 which
suggested that CH4 was first oxidized to CH3OH and
CH3OOH, and then, CH3OH was further oxidized to
HOCH2OOH and HCOOH.2 The possible reaction path of
methane conversion, as shown in Figure 6g, was further
supported by a 13C nuclear magnetic resonance (NMR)
Applications on Operando Monitoring of Products in
Two Microcatalytic Reaction Processes. The selective
activation and orientable conversion of methane to high-value-
added chemicals are a promising work and are considered the
“holy grail” in catalysis. An efficient nonprecious catalyst to
directly convert methane to C1 oxygenated products under
mild conditions based on graphene-confined single Fe atoms
has been developed in the Dalian Institute of Chemical
Physics.2 In-depth understanding of the catalytic reaction
mechanism requires the information of the product variation
during the reaction. Therefore, the NEF-HPPI/PICI TOFMS
was used to monitor the products in operando during the
reaction process.
As the VUV photon flux and the photoelectron emission
efficiency of the electrode may decrease due to contamination
of the window of the VUV lamp and oxidation of the electrode,
especially for the analytes with relatively high concentrations,
the stability will be affected during long-term measurement. As
shown in Figure 5a, the black curve displays a positive
correlation between the peak intensity of the H318O+ reactant
ions (the isotope peak of H316O+ as the signal intensity of
2212
https://dx.doi.org/10.1021/acs.analchem.0c04081
Anal. Chem. 2021, 93, 2207−2214

Analytical Chemistry
pubs.acs.org/ac
Article
experiment and density functional theory (DFT) calculations
in the previous work.2
processes exhibited an excellent performance and wide
potential applications of the NEF-HPPI/PICI TOFMS system
in process monitoring, reaction mechanism research, and
online quality control.
The methanol-to-olefins reaction is of strategic significance
for the production of light olefins.39,40 In this work, the
hydrocarbon products during the initial stage of the methanol-
to-olefins reaction were real-time monitored in the HPPI mode
with a high temporal resolution of 1 s. Details of the operation
conditions for this reaction are described in the Supporting
Information, S9. To investigate the influence of the electric
field intensity of E3 on the ion dissociation of light olefins, 890
ppm n-pentene was taken as an example to optimize the
electric field intensity of E3 for effectively avoiding the
formation of cluster ions and suppressing the production of
fragment ions. The ion source pressure was decreased to 50 Pa
to further reduce the formation of olefin cluster ions, and the
acquisition time of mass spectra in Figure 6a−c is 10 s. As
shown in Figure 6a, when the value of E3 was initially set to 0
V/cm, only the molecular ion peak of n-pentene was presented
in the mass spectrum without any cluster ions and fragment
ions. However, the signal intensity of n-pentene was very weak,
which was attributed to the poor ion transmission. As the
electric field intensity of E3 increased to 17 V/cm (Figure 6b),
the intensity of the molecular ion peak of n-pentene increased
sharply to a maximum value with very little fragment ions and
cluster ions. When the value of E3 was further elevated to 33
V/cm (Figure 6c), the molecular ion of n-pentene fragmented
dramatically and the dominant ions were the fragments [M-
CH3]+ and [M-C2H4]+, the latter of which was identical to the
molecular ion of propylene [C3H6]+. Therefore, to avoid
excessive fragments, the electric field intensity of E3 was set to
17 V/cm during the methanol-to-olefins reaction process.
Figure 6d illustrates the time course of the light olefin
products, propylene, butene, pentene, and hexene with a
temporal resolution of 1 s during the initial stage of the
methanol-to-olefins reaction, which indicates that the olefins
with a higher carbon number were produced later. The mass
spectrum acquired at the reaction time of the ninth second in
Figure 6e shows that propylene and butene were the main
products at this time, besides the reactant methanol and
intermediate methyl ether.
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.analchem.0c04081.
RF-only quadrupole TOFMS, sample preparation, ion
transmission efficiency simulation in the ion source and
RF quadrupole region with different carrier gases,
influence of electric field intensities of E2 and E3 on
the ion dissociation and sensitivity in the elevated
uniform electric field, signal intensities of methanol and
formic acid as a function of the electric field intensity E3
in the E3-enhanced nonuniform electric field, influence
of the electric field intensity E3 on characteristic ion
distributions, performance of the NEF-HPPI/PICI
TOFMS, and operation of the methane conversion
reaction and methanol-to-olefins catalytic reaction
(PDF)
AUTHOR INFORMATION
■
Corresponding Authors
Lei Hua − CAS Key Laboratory of Separation Science for
Analytical Chemistry, Dalian Institute of Chemical Physics,
Chinese Academy of Sciences, Dalian, Liaoning 116023,
People’s Republic of China; Email: lhua@dicp.ac.cn;
Fax: +86-411-84379517
Haiyang Li − CAS Key Laboratory of Separation Science for
Analytical Chemistry, Dalian Institute of Chemical Physics,
Chinese Academy of Sciences, Dalian, Liaoning 116023,
People’s Republic of China; orcid.org/0000-0002-6658-
4745; Email: hli@dicp.ac.cn
Authors
Ningbo Wan − CAS Key Laboratory of Separation Science for
Analytical Chemistry, Dalian Institute of Chemical Physics,
Chinese Academy of Sciences, Dalian, Liaoning 116023,
People’s Republic of China; University of Chinese Academy of
Sciences, Beijing 100049, People’s Republic of China
Jichun Jiang − CAS Key Laboratory of Separation Science for
Analytical Chemistry, Dalian Institute of Chemical Physics,
Chinese Academy of Sciences, Dalian, Liaoning 116023,
People’s Republic of China
Fan Hu − Henan Medical Instruments Testing Institute,
Zhengzhou 450018, People’s Republic of China
Ping Chen − CAS Key Laboratory of Separation Science for
Analytical Chemistry, Dalian Institute of Chemical Physics,
Chinese Academy of Sciences, Dalian, Liaoning 116023,
People’s Republic of China
Kaixin Zhu − State Key Laboratory of Catalysis, iChEM,
Dalian Institute of Chemical Physics, Chinese Academy of
Sciences, Dalian 116023, China
Dehui Deng − State Key Laboratory of Catalysis, iChEM,
Dalian Institute of Chemical Physics, Chinese Academy of
Sciences, Dalian 116023, China; orcid.org/0000-0001-
9702-0531
CONCLUSIONS
■
It has been demonstrated that the NEF-HPPI/PICI source
with a three-zone ionization region is a soft and versatile
ionization technique for online MS in real-time monitoring of
complex VOC mixtures. High-intensity H3O+ reactant ions
generated by PEI of water molecules could be used to
effectively ionize compounds with high IEs and low photo-
ionization cross sections but relatively high PAs through PICI,
such as OVOCs. A nonuniform electric field in the three-zone
ionization region was established to enhance in-source
declustering by elevating the electric field intensity of the ion
dissociation and focusing zone E3 while keeping the electric
field intensity in the analyte ionization zone E2 at a relatively
low value. The design of the nonuniform electric field as well as
the use of high-purity helium as the carrier gas not only
diminished the formation of cluster ions but also improved the
ion transmission efficiency. As a result, the main characteristic
ion of molecule ion M+ for HPPI and protonated molecule ion
[MH]+ for PICI occupied more than 80% of the total
characteristic ions. The achieved LODs were down to lower-
ppbv to sub-ppbv for some hydrocarbons and OVOCs. The
applications on operando monitoring of two catalytic reaction
Yuanyuan Xie − CAS Key Laboratory of Separation Science
for Analytical Chemistry, Dalian Institute of Chemical
2213
https://dx.doi.org/10.1021/acs.analchem.0c04081
Anal. Chem. 2021, 93, 2207−2214

Analytical Chemistry
pubs.acs.org/ac
Article
(19) Muhlberger, F.; Saraji-bozorgzad, M.; Gonin, M.; Fuhrer, K.;
̈
Physics, Chinese Academy of Sciences, Dalian, Liaoning
116023, People’s Republic of China
Chenxin Wu − CAS Key Laboratory of Separation Science for
Analytical Chemistry, Dalian Institute of Chemical Physics,
Chinese Academy of Sciences, Dalian, Liaoning 116023,
People’s Republic of China; University of Chinese Academy of
Sciences, Beijing 100049, People’s Republic of China
Zimmermann, R. Anal. Chem. 2007, 79, 8118−8124.
(20) Gamez, G.; Zhu, L.; Schmitz, T. A.; Zenobi, R. Anal. Chem.
2008, 80, 6791−6795.
(21) Hua, L.; Wu, Q.; Hou, K.; Cui, H.; Chen, P.; Wang, W.; Li, J.;
Li, H. Anal. Chem. 2011, 83, 5309−5316.
(22) Xie, Y.; Hua, L.; Hou, K.; Chen, P.; Zhao, W.; Chen, W.; Ju, B.;
Li, H. Anal. Chem. 2014, 86, 7681−7687.
̌
̌
(23) Vadhwana, B.; Belluomo, I.; Boshier, P. R.; Pavlou, C.; Spanel,
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.analchem.0c04081
P.; Hanna, G. B. Rapid Commun. Mass Spectrom. 2020, 34, No. e8706.
̌
̌
(24) Lacko, M.; Piel, F.; Mauracher, A.; Spanel, P. Phys. Chem.
Chem. Phys. 2020, 22, 10170−10178.
(25) Pedrotti, M.; Khomenko, I.; Fontana, M.; Somenzi, M.;
Falchero, L.; Arveda, M.; Cappellin, L.; Fogliano, V.; Biasioli, F. Int.
Dairy J. 2020, 109, 104729.
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
(26) Yang, B.; Zhang, H.; Shu, J.; Ma, P.; Zhang, P.; Huang, J.; Li,
Z.; Xu, C. Anal. Chem. 2018, 90, 1301−1308.
(27) Zhang, H.; Ma, P.; Shu, J.; Yang, B.; Huang, J. Anal. Chim. Acta
2018, 1035, 119−128.
(28) Shu, J.; Zou, Y.; Xu, C.; Li, Z.; Sun, W.; Yang, B.; Zhang, H.;
Zhang, P.; Ma, P. Sci. Rep. 2016, 6, 36820.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(29) Huang, J.; Yang, B.; Shu, J.; Zhang, Z.; Li, Z.; Jiang, K. Anal.
Chem. 2019, 91, 5605−5612.
Financial support from the National Natural Science
Foundation of China (nos. 21705148, 21904124, and
21927804), the LiaoNing Revitalization Talents Program
(XLYC1808022), the Natural Science Foundation of Liaoning
Province (2020-MS-017), and the Dalian Institute of Chemical
Physics (grants DICP ZZBS201701 and DICP ZZBS201801)
is gratefully acknowledged.
(30) Bohme, D. K.; Mackay, G. I.; Tanner, S. D. J. Am. Chem. Soc.
1979, 101, 3724−3730.
(31) Racine, A. H.; Payne, A. H.; Remes, P. M.; Glish, G. L. Anal.
Chem. 2006, 78, 4609−4614.
(32) Hua, L.; Hou, K.; Chen, P.; Xie, Y.; Jiang, J.; Wang, Y.; Wang,
W.; Li, H. Anal. Chem. 2015, 87, 2427−2433.
(33) Snyder, D. T.; Fedick, P. W.; Cooks, R. G. Anal. Chem. 2016,
88, 9572−9581.
REFERENCES
■
(34) Koss, A. R.; Warneke, C.; Yuan, B.; Coggon, M. M.; Veres, P.
R.; de Gouw, J. A. Atmos. Meas. Tech. 2016, 9, 2909−2925.
(35) Hewitt, C. N.; Hayward, S.; Tani, A. J. Environ. Monit. 2003, 5,
1−7.
(1) Li, Q.; Hua, L.; Xie, Y.; Jiang, J.; Li, H.; Hou, K.; Tian, D.; Li, H.
Analyst 2019, 144, 1104−1109.
(2) Cui, X.; Li, H.; Wang, Y.; Hu, Y.; Hua, L.; Li, H.; Han, X.; Liu,
Q.; Yang, F.; He, L.; Chen, X.; Li, Q.; Xiao, J.; Deng, D.; Bao, X. Chem
2018, 4, 1902−1910.
̌
̌
̌
(36) Spanel, P.; Zabka, J.; Zymak, I.; Smith, D. Rapid Commun. Mass
Spectrom. 2017, 31, 437−446.
(3) Xie, Y.; Chen, P.; Hua, L.; Hou, K.; Wang, Y.; Wang, H.; Li, H. J.
Am. Soc. Mass Spectrom. 2016, 27, 144−152.
̌
̌
̌
̌
(37) Shestivska, V.; Kolivoska, V.; Kubista, J.; Smith, D.; Spanel, P.
Rapid Commun. Mass Spectrom. 2020, 34, No. e8744.
(38) Lammert, S. A.; Wells, J. M. Rapid Commun. Mass Spectrom.
1996, 10, 361−371.
(39) Jiao, F.; Li, J.; Pan, X.; Xiao, J.; Li, H.; Ma, H.; Wei, M.; Pan, Y.;
Zhou, Z.; Li, M.; Miao, S.; Zhu, Y.; Xiao, D.; He, T.; Yang, J.; Qi, F.;
Fu, Q.; Bao, X. Science 2016, 351, 1065−1068.
(40) Wang, C.; Chu, Y.; Xu, J.; Wang, Q.; Qi, G.; Gao, P.; Zhou, X.;
Deng, F. Angew. Chem., Int. Ed. 2018, 57, 10197−10201.
(4) Jia, L.; Brech, Y. L.; Mauviel, G.; Qi, F.; Frowein, M. B.-v.;
Ehlert, S.; Zimmermann, R.; Dufour, A. Energy Fuels 2016, 30, 1555−
1563.
(5) Fischer, M.; Wohlfahrt, S.; Varga, J.; Saraji-bozorgzad, M.;
Matuschek, G.; Denner, T.; Zimmermann, R. J. Therm. Anal. 2014,
116, 1461−1469.
(6) Jiang, J.; Wang, Y.; Hou, K.; Hua, L.; Chen, P.; Liu, W.; Xie, Y.;
Li, H. Anal. Chem. 2016, 88, 5028−5032.
(7) Zhang, P.; Ma, P.; Zhang, H.; Shu, J.; Yang, B.; Li, Z. Atmos.
Environ. 2017, 164, 250−258.
(8) Heide, J.; Czech, H.; Ehlert, S.; Koziorowski, T.; Zimmermann,
R. J. Agric. Food Chem. 2020, 68, 4752−4759.
(9) Toniolo, R.; Pizzariello, A.; Dossi, N.; Lorenzon, S.; Abollino,
O.; Bontempelli, G. Anal. Chem. 2013, 85, 7241−7247.
(10) Wang, Y.; Jiang, J.; Hua, L.; Hou, K.; Xie, Y.; Chen, P.; Liu, W.;
Li, Q.; Wang, S.; Li, H. Anal. Chem. 2016, 88, 9047−9055.
(11) Li, Z.; Xu, C.; Shu, J.; Yang, B.; Zou, Y. Talanta 2017, 165, 98−
106.
(12) Wang, Y.; Hua, L.; Li, Q.; Jiang, J.; Hou, K.; Wu, C.; Li, H.
Anal. Chem. 2018, 90, 5398−5404.
(13) Hanley, L.; Zimmermann, R. Anal. Chem. 2009, 81, 4174−
4182.
(14) Bhardwaj, C.; Hanley, L. Nat. Prod. Rep. 2014, 31, 756−767.
(15) Adam, T.; Zimmermann, R. Anal. Bioanal. Chem. 2007, 389,
1941−1951.
(16) Cool, T. A.; Wang, J.; Nakajima, K.; Taatjes, C. A.; Mcllroy, A.
Int. J. Mass Spectrom. 2005, 247, 18−27.
(17) NIST. National Institute of Standards and Technology (NIST);
http://webbook.nist.gov/chemistry/ (accessed October 16, 2020).
̈
(18) Eschner, M. S.; Groger, T. M.; Horvath, T.; Gonin, M.;
Zimmermann, R. Anal. Chem. 2011, 83, 3865−3872.
2214
https://dx.doi.org/10.1021/acs.analchem.0c04081
Anal. Chem. 2021, 93, 2207−2214