Formation of positive product ions from substances with low proton affinity in high kinetic energy ion mobility spectrometry

Rationale: Ion mobility spectrometry (IMS) instruments are typically equipped with atmospheric pressure chemical ionization (APCI) sources operated at ambient pressure. However, classical APCI-IMS suffers from a limited ionization yield for nonpolar substances with low proton affinity (PA). This is mainly due to ion clustering processes, especially those that involve water molecules, inhibiting the ionization of these substances. Methods: High Kinetic Energy (HiKE)-IMS instruments are operated at decreased pressures and high reduced electric field strengths. As most clustering reactions are inhibited under these conditions, the ionization yield for nonpolar substances with low PA in HiKE-IMS should differ from that in classical APCI-IMS. To gain first insights into the ionization capabilities and limitations of HiKE-IMS, we investigated the ionization of four model substances with low PA in HiKE-IMS using HiKE-IMS-MS as a function of the reduced electric field strength. Results: The four model substances all have proton affinities between those of H 2 O and (H 2 O) 2 but exhibit different ionization energies, dipole moments, and polarizabilities. As expected, the results show that the ionization yield for these substances differs considerably at low reduced electric field strengths due to ion cluster formation. In contrast, at high reduced electric field strengths, all substances can be ionized via charge and/or proton transfer in HiKE-IMS. Conclusions: Considering the detection of polar substances with high PAs,


| INTRODUCTION
Due to their high sensitivity, fast response times, and compact design, ion mobility spectrometers are commonly used in safety and security applications such as the detection of chemical warfare agents, 1,2 toxic industrial chemicals, 3,4 drugs, 5,6 and explosives. [7][8][9] Basically, ion mobility spectrometry (IMS) instruments can be divided by their principle of ion separation. In this work, a drift tube (DT) ion mobility spectrometer is used. In DT-IMS, ions are separated by their motion along the axis of a drift tube driven by a homogeneous static electric field. To initiate the measurement, an ion packet is injected into the drift tube. During their motion, the ions are separated based on the absolute value of their ion mobility in the present drift gas. At the end of the drift tube, the ions are captured by a detector that converts and amplifies the ion current into a measurable voltage. By plotting the measured voltage over the drift time, that is, the time that the ions need to reach the detector, an ion mobility spectrum is obtained.
Typically, ion mobility spectrometers are equipped with atmospheric pressure chemical ionization (APCI) sources operated at ambient pressure. Here, the ionization proceeds in two steps. Initially, stable reactant ions are generated by ionizing the main constituents of the sample gas. In a second step, ionization of analyte molecules follows through reactions with these reactant ions. Due to the large number of collisions at ambient pressure, IMS with APCI sources can achieve limits of detection in the low ppt v (parts-per-trillion by volume) range in measurement times of less than a second for a broad range of substances. 10,11 Although this renders APCI-IMS at ambient pressure a very sensitive and versatile detection method with considerably low instrumental effort, it suffers from a low linear range, strong matrix effects, and a limited ionization yield for a number of compound classes. The ionization yield strongly depends on the generated reactant ion species. Typically, the positive reactant ions To overcome these limitations, we introduced High Kinetic Energy IMS (HiKE-IMS). 19,20 As in the classical DT-IMS method operated at ambient pressure, in HiKE-IMS, ions are generated in a reaction region by a reverse-flow continuous corona discharge ionization source before ions are separated in a drift region. In contrast to classical IMS, HiKE-IMS instruments are operated at a decreased pressure between 10 and 40 mbar to reach high reduced electric field strengths in both the reaction region and the drift region.
As known from PTR-MS, 16  The aim of this work is to gain first insights into the ionization processes in HiKE-IMS leading to the formation of positive product ions from substances with low PA. The mechanisms underlying the formation of negative product ions will be investigated in a separate paper, as a description of all ionization processes would exceed the scope of this paper. In this work, first, a theoretical overview of the possible ionization pathways resulting in positive product ions in APCI sources is given. Subsequently, the HiKE-IMS ionization of the four exemplary model substances, acetonitrile, methanol, phosphine, and benzene, is investigated as a function of the reduced electric field strength and the humidity in the reaction region. The occurring product ion species are identified using HiKE-IMS-MS. 21 2 | EXPERIMENTAL

| HiKE-IMS
A detailed description of the HiKE-IMS setup was previously provided by Kirk et al. 24 The operating parameters used in this work are presented in Table 1. To identify individual ion species associated with signals in the ion mobility spectrum, the HiKE-IMS-MS coupling described in a previous work 21 44 However, it is noteworthy that a spontaneous process may proceed quickly or slowly, as spontaneity is not related to kinetics or reaction rate.
According to Equation 1, the change in the Gibbs energy ΔG 0 T at constant temperature T can be calculated from the change in the enthalpy ΔH 0 T and the change in the entropy ΔS 0 T : where T is the absolute temperature and the superscript 0 refers to the standard state. In the ionization reactions considered in this work, the entropy change ΔS 0 T is typically negligibly small, as a proton or a charge is a simple entity whose transfer from one molecule to another does not significantly alter the net system entropy. Thus, in almost all practical cases, an ionization reaction occurs spontaneously if the change in the reaction enthalpy ΔH is negative. 16 1.
According to previous studies, 49 These ligand-switching reactions may occur if the switching process is sufficiently exothermic. 18 This may be achieved when the hydration energy of the analyte molecules is comparable with that of H 2 O. Thus, the permanent dipole moment and the polarizability of the analyte molecule seem to be key parameters to predict whether ligand-switching reactions occur. 10,17,18,52 However, as shown in several studies, 17,18,52-54 the rate constants of the ligand-switching reactions typically decrease with increasing water cluster size n. Therefore, increasing the sample gas humidity and thus the humidity in the reaction region results in a decrease in the sensitivity. In particular, IMS instruments operated at ambient pressure thus suffer from a sample gas humidity-dependent detection of analytes. 10,55 To minimize the influence of the sample gas humidity on the sensitivity, the HiKE-IMS instrument is operated at high reduced electric field strengths. In this context, the reduced field is a measure of the average ion kinetic energy. On increasing the reduced electric field strength, collision-induced dissociation of water clusters occurs.
However, due to the higher IE of O 2 , there is a high amount of excess energy from charge transfer with O 2 + that will be deposited in the product ion and may result in its fragmentation. 45 For the sake of completeness, it should be mentioned that ionmolecule association reactions that include a third body A (reaction 7) have also been reported for NO + if the IE of the analyte molecule is similar to that of NO. 25,31 However, it is to be expected that these association channels will be inhibited at elevated reduced electric field strengths in HiKE-IMS:

| RESULTS AND DISCUSSION
The ionization pathways occurring in HiKE-IMS significantly depend on the reduced electric field strength in the reaction region, affecting both the ion's residence time in the reaction region and its kinetic energy. Thus, in the presented measurements, the dependence of the product ion population on this parameter is investigated.

| Positive reactant ion population inside the reaction region
As the ionization pathways in HiKE-IMS are determined by the prevailing reactant ion population, the reactant ion spectrum is considered first. Figure 1A shows the recorded reactant ion spectra for different reduced electric field strengths in the reaction region at a constant reduced electric field strength of 110 Td in the drift region.
In these measurements, the water concentration in the drift gas is When interpreting the HiKE-IMS spectra presented in Figure 1, it is important to note that the reduced electric field strength in the drift region was set to a high value of 110 Td. Furthermore, the water concentration in the drift gas was kept constant at a low value of 50 ppm v . Under these conditions, conversion reactions inside the drift region are inhibited, as well as cluster association or dissociation reactions. 22 Thus, the HiKE-IMS spectra shown in Figure 1 Table 1 converted into H 3 O + (H 2 O) n while traversing the reaction region. This is in accordance with the simulated results in Figure 2

| Formation of positive product ions from substances with low PA
Here, the occurring ionization pathways in HiKE-IMS are investigated based on the generated product ions of the four exemplary model substances: acetonitrile, methanol, phosphine, and benzene. Table 2 presents the thermochemical properties of these substances relevant However, the reaction of O 2 + with methanol seems to be inefficient, proceeding significantly below the collisional rate. 26 In this work, acetonitrile, methanol, phosphine, and benzene serve as model substances enabling a first evaluation of the capabilities and limitations of HiKE-IMS regarding the ionization of substances with low PA in positive ion mode.
The recorded positive HiKE-IMS spectra of acetonitrile, methanol, phosphine, and benzene are shown in Figure 3A for six different reduced electric field strengths in the reaction region. Again, the reduced electric field strength in the drift region is kept constant at 110 Td, the sample gas humidity is 4000 ppm v , and the drift gas humidity is 50 ppm v . To identify the product ion species underlying the peaks in the ion mobility spectrum, the product ion peaks marked in Figure 3A are transferred to the coupled mass spectrometer. In Figure 3B, the recorded mass spectra corresponding to the marked product ion peaks are shown.
In the HiKE-IMS spectrum of benzene, the benzene cation C 6    the simulated results, this is mainly due to an increasing presence of NO + (H 2 O) n in the reaction region. Furthermore, at reduced electric F I G U R E 3 A, Positive high kinetic energy ion mobility spectrometry (HiKE-IMS) spectra of acetonitrile, methanol, phosphine, and benzene at different reduced electric field strengths in the reaction region and B, mass spectra corresponding to the marked product ion peaks in the HiKE-IMS spectra recorded by the HiKE-IMS-MS instrument operated in the selected-mobility mode. The sample gas humidity is 4000 ppm v . The other operating parameters are provided in Table 1 field strengths in the reaction region exceeding 70 Td, the measured In Figure 5B, the measured relative abundances of the protonated analyte molecules from acetonitrile, methanol, phosphine, Second, at these high reduced field strengths, the formation of large water clusters is inhibited due to collision-induced cluster dissociation, thus allowing for an efficient ionization of even nonpolar substances with low PA via PTRs with H 3 O + .

| Influence of humidity in the reaction region
When HiKE-IMS is used in field applications with a direct sample gas inlet, the sample gas humidity might range from 5% to 95% relative humidity (rH) at 298 K and 1013 mbar, leading to a relative humidity in the reaction region from 2.5% to 47%. Here, we state only the sample gas humidity, as this value is most relevant in the application.
Due to the mixing ratio of the sample gas and the drift gas of 1:1, the conversion factor from sample gas humidity to reaction region humidity is 0. ionization of these substances is hardly affected when the relative sample gas humidity is varied. However, the effect of the sample gas humidity is more pronounced when considering the HiKE-IMS spectra of benzene. As shown in Figure 6, when the relative sample gas humidity is varied, the measured and the simulated relative abundances of the singly charged benzene C 6

| Capabilities and limitations of HiKE-IMS
The results presented demonstrate three fundamental aspects regarding analyte ionization that should be considered when using HiKE-IMS.  Table 1 substances with low PA that are not detected or are difficult to detect at ambient pressure, HiKE-IMS would be beneficial. Furthermore, the much-shorter residence time of ions in the reaction region leads to a significant enhancement of the linear range and limited chemical cross-sensitivities in HiKE-IMS. Another important benefit of HiKE-IMS might be the possibility of changing the reduced electric field strength in the drift region independently of the reduced electric field in the reaction region to separate substances by their field-dependent ion mobility as known from field asymmetric ion mobility spectrometry and differential mobility spectrometry allowing for improved substance identification. Furthermore, high reduced electric field strengths can lead to fragmentation, also improving substance identification.