Protonation enhancement by dichloromethane doping in low-pressure photoionization

Doping has been used to enhance the ionization efficiency of analytes in atmospheric pressure photoionization, which is based on charge exchange. Compounds with excellent ionization efficiencies are usually chosen as dopants. In this paper, we report a new phenomenon observed in low-pressure photoionization: Protonation enhancement by dichloromethane (CH2Cl2) doping. CH2Cl2 is not a common dopant due to its high ionization energy (11.33 eV). The low-pressure photoionization source was built using a krypton VUV lamp that emits photons with energies of 10.0 and 10.6 eV and was operated at ~500–1000 Pa. Protonation of water, methanol, ethanol, and acetaldehyde was respectively enhanced by 481.7 ± 122.4, 197.8 ± 18.8, 87.3 ± 7.8, and 93.5 ± 35.5 times after doping 291 ppmv CH2Cl2, meanwhile CH2Cl2 almost does not generate noticeable ions itself. This phenomenon has not been documented in the literature. A new protonation process involving in ion-pair and H-bond formations was proposed to expound the phenomenon. The observed phenomenon opens a new prospect for the improvement of the detection efficiency of VUV photoionization.

where A represents the analyte molecules and R is the reagent which offers a proton or hydrogen atom. The reagent could be the analyte or solvent molecules. The use of dopants has been found to be very effective for enhancing the ionization efficiency of analytes 6,7,[9][10][11] in APPI and LPPI 12 via charge exchange: Scientific RepoRts | 6:36820 | DOI: 10.1038/srep36820 where D and A represent dopant and analyte molecules, respectively. Benzene (IE = 9.24 eV) 13,14 , acetone (IE = 9.70 eV) 10,[15][16][17] , toluene (IE = 8.83 eV) 6,7,[10][11][12][18][19][20] , and anisole (IE = 8.20 eV) 21 are often employed as dopants due to their excellent photoionization efficiencies under illumination of the krypton lamp. The resulting analyte ions may subsequently react with other molecules via proton transfer. The detection sensitivity could be enhanced by ~100 times via doping 22 . However, these dopants cannot be applied to the detection of methanol (CH 3 OH, IE = 10.84 eV), ethanol (C 2 H 5 OH, IE = 10.48 eV), and acetaldehyde (C 2 H 4 O, IE = 10.23 eV) due to their higher IEs. Dichloromethane has been chosen as a dopant for characterizing the molecular structures of analytes via secondary ion-molecule reactions, rather than for enhancing ionization efficiency 23 .
Our previous studies revealed that LPPI with a specially designed photoionizer was super sensitive (~1000 counts/ppbv) to many organic compounds [24][25][26] . However, the LPPI detection efficiency for CH 3 OH, C 2 H 5 OH, and C 2 H 4 O is very low due to their low ionization efficiencies. In this paper, we report a new phenomenon: The detection efficiencies of the three small volatile organic compounds (VOCs) can be remarkably enhanced via CH 2 Cl 2 doping. The results and experimental method are described in the following sections.

Results
Protonation enhancement of water and LPPI mass spectrum of CH 2 Cl 2 . Water (H 2 O) is an important protonation agent for PTR mass spectrometry. The IE of water is 12.62 eV, which indicates that it cannot be photoionized directly by the photons emitted from the krypton lamp. However, H 3 O + (m/z 19, 45 counts), (H 2 O) 2 H + (m/z 37, 214 counts), and (H 2 O) 3 H + (m/z 55, 24 counts) were observed in the LPPI mass spectrum of N 2 , as shown in Fig. 1(A). The concentration of water in the test chamber was < 5 ppmv, as a result of impurities in high-purity N 2 gas. Protonation of acetonitrile (IE = 12.20 eV) was observed in APPI with a krypton lamp as the VUV light source by Marotta et al. The authors speculate that photon irradiation leads first to the isomerization of acetonitrile molecules, affording species that exhibit IEs < 10 eV and that consequently are able to generate photoionization products 27 . The formation mechanism of protonated water and water clusters under illumination of 10.0 and 10.6 eV photons is not clear yet. In view of a tiny amount of N 2 + (m/z 28, 34 counts) observed in Fig. 1(A), the photoelectrons in the photoionization region might lead to the formation of protonated water and water clusters. Figure 1(B) shows the mass spectrum obtained after injecting 291 ppmv CH 2 Cl 2 into the chamber. Surprisingly, the signal intensities of H 3 O + , (H 2 O) 2 H + , and (H 2 O) 3 H + increased to 2.92 × 10 4 , 1.24 × 10 5 and 2.29 × 10 4 counts, respectively. The signal intensity of protonated water was averagely enhanced by 481.7 ± 122.4 times, measured from three independent measurements. This phenomenon has never been reported. CH 2 Cl 2 is a common solvent used in organic analysis. The IE of CH 2 Cl 2 is 11.33 eV. It cannot be directly ionized by the VUV photons emitted from the krypton lamp. As shown in Fig. 1(B), no noticeable ions were produced from direct photoionization of CH 2 Cl 2 . A small mass peaks at m/z 47 is assigned to ethanol residual in the test chamber or minor impurity in the CH 2 Cl 2 reagent.
Protonation enhancement of methanol, ethanol, and acetaldehyde. Methanol (CH 3 OH) is the simplest alcohol. Its IE is 10.84 eV, higher than the energy of the photons emitted from the krypton lamp. A weak signal of protonated methanol was observed when 4.6 ppmv methanol was sampled. Figure 2(A) shows the obtained LPPI mass spectrum of 4.6 ppmv methanol in nitrogen. The mass peaks at m/z 19, 37, and 55 correspond to H 3 O + , (H 2 O) 2 H + , and ( H 2 O) 3 H + , respectively. The mass peaks at m/z 33, 51 and 65 are assigned to (CH 3 OH) H + , (CH 3 OH·H 2 O)H + and (CH 3 OH) 2 H + , respectively. The moderate mass peak at m/z 47 is assigned to ethanol, the impurity in the methanol reagent. The peak intensities of (CH 3 OH)H + and (CH 3 OH) 2 H + are 559 and 171 counts, respectively. It is reported in the literature that dimers of methanol (CH 3 OH) 2 with IE equal to 9.72 eV coexist with methanol monomers under ambient conditions and that protonated methanol is generated from the dissociation of (CH 3 OH) 2 +28,29 . Figure 2(B) shows the LPPI mass spectrum of 4.6 ppmv methanol doped with 291 ppmv CH 2 Cl 2 . The signal intensities of the mass peaks at m/z 33 and 65 reach 1.48 × 10 5 and 6.06 × 10 4 counts, respectively. The signal intensity of protonated methanol was averagely enhanced by 197.8 ± 18.8 times, measured from three independent measurements.
The IE of ethanol (C 2 H 5 OH) is 10.48 eV, meaning it can be photoionized by the photons emitted from the krypton lamp (10.6 eV, 20%). Figure 3(A) shows the LPPI mass spectrum of 1.6 ppmv ethanol in nitrogen. As well as ions resulting from water and water clusters, mass peaks at m/z 45, 47, and 93 are assigned to ions produced from ethanol, i.e. C 2 H 5 O + (551 counts), (C 2 H 5 OH)H + (1923 counts), and (C 2 H 5 OH) 2 H + (222 counts). The mass peak of protonated ethanol was the strongest peak. After doping with 291 ppmv CH 2 Cl 2 , the intensities of the   Figure 4(B) shows the LPPI mass spectrum of 0.66 ppmv acetaldehyde in nitrogen doped with 291 ppmv CH 2 Cl 2 . The signal intensity of protonated acetaldehyde (m/z 45) increased to 7.04 × 10 4 counts, while the signal at m/z 61 slightly increased to 2107 counts. The signal intensity of protonated acetaldehyde was averagely enhanced by 93.5 ± 35.5 times, measured from three independent measurements. Additionally, a mass peak at m/z 63 assigned to (C 2 H 4 O·H 2 O)H + (1.71 × 10 4 counts) appeared.
Benzene (C 6 H 6 ) is an important chemical and atmospheric pollutant. Its IE is 9.24 eV, lower than the energy of VUV photons emitted from the krypton lamp. Benzene and its derivatives have excellent photoionization efficiencies under illumination of a krypton VUV lamp. Figure 5(A) shows the LPPI mass spectrum of 0.42 ppmv  Scientific RepoRts | 6:36820 | DOI: 10.1038/srep36820 benzene. The mass peak at m/z 78 is assigned to 12 C 6 H 6 + (6.42 × 10 4 ). Figure 5(B) shows the LPPI mass spectrum of 0.42 ppmv benzene in nitrogen doped with 291 ppmv CH 2 Cl 2 . The intensities of the mass peak at m/z 78 decreased by ~14% to 5.54 × 10 4 counts. The fluctuation of the signal intensities at m/z 78 was observed in separate experiments. No signal enhancement at m/z 79 (protonated benzene) was observed in all experiments.

Discussion
Pure CH 2 Cl 2 in LPPI almost does not generate noticeable ions as shown in Fig. 1, which implies that the observed protonation enhancement is not attributed to charge exchange. In order to reveal the mechanism of protonation enhancement, the doping effects of H 2 , CH 4 , CHCl 3 , and CCl 4 on the signals of methanol, ethanol, and acetaldehyde were also investigated. Among the four dopants, only CHCl 3 yielded a weaker enhancement on protonation of methanol, ethanol, and acetaldehyde compared with CH 2 Cl 2 . Under illumination of the krypton lamp, CH 4 , CHCl 3 , CH 2 Cl 2 , and CCl 4 have relatively strong absorption (~10 −17 cm 2 ) and are excited to Rydberg states 30,31 , while H 2 does not have absorption 32 . Shaw et al. reported that ion-pair states were observed in halogenated methanes excited by VUV light and ion pair states even existed below ionization potentials 33 . We speculate that CHCl 3 and CH 2 Table 1 lists IEs, PAs, molecular dipole moments, H-bond formation possibilities, and protonation enhancements of the compounds investigated. It is very interesting that protonation of benzene and self-protonation of dichloromethane were not observed in the experiment, while water and other three organics had significant protonation enhancements. The difference observed in protonation enhancements cannot be addressed simply by proton affinities or molecular dipole moments of the compounds. It is enlightening that the observed protonation enhancements of the compounds are coincident with their abilities to form a H bond as a H acceptor as shown in Table 1. The four compounds, water, methanol, ethanol, and acetaldehyde, are all capable of forming a H bond as a H acceptor, while benzene and dichloromethane are not. These phenomena may imply that the compounds are not protonated by free protons or protonated molecules. Based on experimental observations and the analyses above, we speculate that the following process might take place during CH 2 Cl 2 doping:  . This hypothesis rationalizes all the experimental observations. To the best of our knowledge, protonation via collision with excited-state molecules has not yet been documented. The heat of reaction (Δ r H°) of deprotonation of CH 2 Cl 2 (CH 2 Cl 2 = CHCl 2 − + H + ) is ~16.3 eV 34 . Considering the photon energy of VUV light (~10 eV) and PAs of analyte molecules (in the range of 7-9 eV) 35 , the total process of Reactions 5 to 7 is exothermic for most VOCs. Though the authenticity and intrinsic mechanism of the process still needs further elaborate investigation, the observed phenomenon opens a new prospect for the improvement of the detection efficiency of VUV photoionization.

Methods
The experimental setup has been described in detail elsewhere 25 . Briefly, it mainly consisted of a 120 L test chamber and a LPPI mass spectrometer.
The 120 L test chamber was mainly built with an open-head stainless steel drum and covered with a thin Tedlar bag to ensure one atmospheric pressure during sampling. A stainless steel fan driven by a magnetic field was  placed at the bottom of the test chamber to ensure quick mixing. Nitrogen was used as the buffer gas. An oil-free pump was used as the drain pump. Two mass flow controllers were used for gas samples. All experiments were performed under ambient atmospheric pressure and room temperature. The LPPI mass spectrometer was recently developed in our laboratory. It characterizes with a LPPI source with an optical baffle and a short reflectron time-of-flight mass spectrometer. The body of the LPPI source was a cylindrical stainless steel cavity 6 mm in diameter and 35 mm in length. A radio frequency-driven krypton VUV lamp was used as the VUV light source and coupled to the cylindrical stainless steel cavity with an MgF 2 window. The optical baffle was placed at the exit of the photoionization source to prevent the VUV light entering the mass spectrometer. The LPPI source was passivated with ~600 ppm CH 2 Br 2 under illumination of VUV light for ~8 hours to suppress photoelectron formation in the experiment. The krypton lamp was laboratory-assembled and emitted VUV photons with energies of 10.0 eV (~80%) and 10.6 eV (~20%). The sample gas was introduced into the photoionization source and controlled by a needle valve. The sample flow was maintained at ~1 cm 3 s −1 . The pressure in the photoionization source was 500-1000 Pa. The mass spectrometer was a simple V-shaped time-of-flight mass spectrometer with a free-field flight distance of 460 mm. The cycle time of detection was 10 s.
In the experiments, a small amount of bottle-contained chemical was first injected into the test chamber. Then, 100 μ L CH 2 Cl 2 was injected into the test chamber and the mass spectra were subsequently acquired after each injection. The amount of methanol, ethanol, acetaldehyde, and benzene injected into the nitrogen-filled test chamber was 1.0, 0.5, 0.5, and 0.2 μ L, respectively. The resulting mixing ratios for the prepared gases were 4.6, 1.6, 0.66, and 0.42 ppmv, respectively. The injection of 100 μ L CH 2 Cl 2 resulted in 291 ppmv in the mixing ratio.