Pulsed discharges in supersonic expansions containing the vapor of different precursors (formaldehyde, methanol) produce the m/z = 30 cations with formula [H2,C,O]+. The corresponding [H2,C,O]+ Ar complexes are produced under similar conditions with argon added to the expansion gas. These ions are mass selected in a time-of-flight spectrometer and studied with infrared laser photodissociation spectroscopy. Spectra in the 2300-3000 cm−1 region produce very different vibrational patterns for the ions made from different precursors. Computational studies with harmonic methods and various forms of anharmonic theory allow detailed assignment of these spectra to two isomeric species. Discharges containing formaldehyde produce primarily the corresponding formaldehyde radical cation, CH2O+, whereas those with methanol produce exclusively the cis- and trans-hydroxymethylene cations, HCOH+. The implications for the interstellar chemistry of these cations are discussed.

The structure of ions produced by molecular ionization and fragmentation processes has been a central focus of mass spectrometry for many years.1–3 Small organic cations are also of significant interest for interstellar chemistry.4–8 Radical cations are more reactive and are implicated in important reactions in interstellar gas clouds4–8 as well as in practical organic chemistry.9,10 To determine ion structures experimentally, and to provide optical signatures for their possible detection in space, high quality spectroscopy is needed. Unfortunately, ion spectroscopy is extremely difficult because of the low densities produced, the harsh conditions of most ionization experiments, and the complicated mixtures of neutral precursors and other ions present.11 Recent developments have made it possible to produce greater densities of cold ions with pulsed discharge sources including supersonic nozzle expansions, or with electron impact or electrospray sources followed by ion trapping and cryogenic cooling.12–21 Subsequent mass selection provides unambiguous ion identity, and tunable laser photodissociation measurements allow electronic or vibrational spectra to be obtained. In recent studies employing these methods, our group has reported infrared spectroscopy for several small organic cations.22–30 The present study provides the first infrared spectroscopy for the formaldehyde cation and its structural isomer, the hydroxymethylene cation.

Small organic cations have been the focus of infrared spectroscopy experiments for many years. The first measurements employed infrared laser absorption in various discharge configurations, often with electric field Doppler modulation to separate ion signals from those of neutrals in the same mixture.31–36 In subsequent work, Lee and co-workers demonstrated mass-selected photodissociation spectroscopy experiments using the method of rare gas atom or hydrogen molecule “tagging.”37–41 In this experiment, laser excitation eliminates weakly bound messenger atoms or molecules providing evidence of absorption. Although other methods have been developed for direct absorption experiments,42 tagging is now widely employed in both molecular beam and ion trap experiments.12–21 It has been applied by several groups other than our own for infrared43–54 or electronic55–57 spectroscopy of small organic cations. Our group has employed infrared photodissociation and tagging to study various carbocations,22–27 and more recently oxocations such as CH3OH+, CH2OH+, and C2H3O+.28–30 The present study on formaldehyde and its isomers extends this work.

[H2, C, O]+ radical cations have been studied extensively with mass spectrometry.58–61 Theory predicts three isomers, the expected formaldehyde cation CH2O+, the hydroxymethylene cation HCOH+ (cis and trans conformers), and the oxonium cation COH2+.62–65 CH2O+ and HCOH+ are the lowest energy structures, with CH2O+ lower in energy by about 6–10 kcal/mol. COH2+ is predicted to lie much higher in energy (∼60 kcal/mol) than CH2O+. The barrier for isomerization between CH2O+ and trans HCOH+ involving a 1,2-hydrogen shift has been calculated to lie 45–50 kcal/mol above CH2O+ minimum. Only CH2O+ and HCOH+ are thought to form experimentally.2,61 Spectroscopic information for these ions is very limited. Formaldehyde has been studied with photoelectron spectroscopy (PES), providing information about the cation’s electronic states and some limited vibrational information for its ground state.66,67 More recent high resolution PES measurements have determined the rotational constants.68 CH2O+ radical cations were also studied with electron spin resonance.69,70 The only information about hydroxymethylene cation comes from mass spectrometry, where the HCOH+ structure has been implicated as a stable isomer.59–61 The present work provides the first gas phase infrared spectroscopy for CH2O+ and HCOH+.

[H2, C, O]+ ions are produced in a pulsed discharge/ supersonic expansion of 5–14 atm of argon seeded with the ambient vapor of formaldehyde solution (30%, Baker) or methanol (99.9%, Fisher Scientific) at room temperature. The ions are mass selected in a reflectron time-of-flight spectrometer and observed with infrared photodissociation spectroscopy.22–30 Because single-photon infrared excitation cannot break the strong covalent bonds, we employ rare gas tagging.12–21,37–41 In this method, [H2, C, O]Ar+ ions are produced and mass selected, and IR absorption eliminates the argon. The spectrum is recorded as the [H2, C, O]+ yield versus the frequency of the infrared laser. The laser system is an infrared optical parametric oscillator/amplifier system (OPO/OPA) (LaserVision) pumped by a Nd:YAG laser (Spectra Physics Pro-230) equipped with an external AgGaSe2 crystal.

Theoretical computations were carried out at the CCSD(T) level as implemented in the CFOUR programming package.71 The atomic natural orbital (ANO) basis of Almlöf and Taylor is used for C, H, and O atoms. The Roos augmented double ζ ANO basis is used for Ar atoms. Smaller and larger contractions of the Taylor ANO basis are designated by ANO0 and ANO1, respectively. Harmonic vibrational frequencies resulting from computations were scaled for comparison to the experiment. The scaling factor for the symmetric and antisymmetric stretches of formaldehyde (0.947) was determined by comparing the computed frequencies for neutral formaldehyde vibrations to those known experimentally.72 A linewidth of 12 cm−1 was employed in the simulated spectra to more closely match the measured linewidths. Full computational details are available in the supplementary material file.

Anharmonic calculations used second-order vibrational perturbation theory (VPT2) as implemented in CFOUR.71 For this, isomer geometries were optimized at the CCSD(T)/ANO1 level and then a subsequent VPT2 analysis was performed at the MP2/ANO1 level and the anharmonic corrections obtained were applied to the frequencies obtained from the initial CCSD(T) optimization. VPT2 calculations were also performed at the B3LYP/6-311G(d,p) and MP2/6-311G(d,p) levels of theory/basis within the Gaussian 0973 program package. Two additional anharmonic treatments were employed that used the anharmonic frequencies and cubic force constants obtained as a part of the VPT2 calculations. The first anharmonic treatment focused on the effects of Fermi resonances involving the CH stretch fundamental, a bend overtone, and combination band that were found to be nearly degenerate. In the second anharmonic treatment, the CH stretch and a large amplitude Ar-molecule intermolecular vibration were treated within an adiabatic model that separates the higher frequency CH stretch of the formaldehyde cation from lower frequency intermolecular vibrations. The approach used is very similar to the one developed by Myshakin et al.,74 which was previously employed in a study of CH3O+.28 As quanta are added to the high frequency CH stretch, a shifted harmonic potential is produced that describes the low frequency intermolecular vibration. Additional details about these models can be found in the supplementary material.

Initial experiments designed to produce the formaldehyde cation from formaldehyde vapor employed solid paraformaldehyde heated to about 50 °C. This vapor produced the desired cation, but the signal was extremely unstable over time. Subsequent experiments used the aqueous solution of formaldehyde employed in biological labs, which has a small percentage of methanol added as a stabilizer. Although this resulted in extraneous mass signals from water- and methanol-containing ions, stable signal levels could be obtained for the desired [H2, C, O]+ cations in both pure and argon-tagged forms. A mass spectrum produced with this formaldehyde solution is presented in the supplementary material (Figure S1). Photodissociation experiments were conducted on these ions in the 1000–4000 cm−1 region. These experiments produced sharp resonances in the 2350–2850 cm−1 range, as shown in Figure 1. This spectrum contains six bands, with the most intense features at 2423 and 2653 cm−1. Weaker bands were also detected at 2377, 2460, 2523, and 2799 cm−1. For comparison to these resonances, the symmetric and antisymmetric C—H stretches of neutral formaldehyde occur at 2783 and 2843 cm−1.71 The photoelectron spectroscopy of formaldehyde reported a symmetric C—H stretch for the cation of 2580 cm−1, which is in the middle of the range of our most intense bands.67 The bands in the cation spectrum therefore occur at much lower frequencies than those in the neutral. We were not able to detect any signal in the region of the carbonyl stretching vibration or the HCH scissors vibrations (reported at 1675 and 1210 cm−1 in the photoelectron spectroscopy).67 This could be a result of weaker IR intensities in this region, the much lower laser power available in this region, or the fact that the binding energy of the argon (∼1000 cm−1) is close to the photon energy here.

FIG. 1.

The infrared photodissociation spectrum of [H2, C, O+]Ar produced from a formaldehyde precursor measured in the mass channel corresponding to argon elimination (top trace in black). Scaled harmonic infrared spectra resulting from CCSD(T)/ANO1 calculations are presented for two isomers corresponding to different argon binding positions on the formaldehyde cation structure. The isomer with argon on hydrogen provides a better match to the spacing and relative intensities of the two main bands in the experimental spectrum. The positions of the main bands are provided on the figure in cm−1.

FIG. 1.

The infrared photodissociation spectrum of [H2, C, O+]Ar produced from a formaldehyde precursor measured in the mass channel corresponding to argon elimination (top trace in black). Scaled harmonic infrared spectra resulting from CCSD(T)/ANO1 calculations are presented for two isomers corresponding to different argon binding positions on the formaldehyde cation structure. The isomer with argon on hydrogen provides a better match to the spacing and relative intensities of the two main bands in the experimental spectrum. The positions of the main bands are provided on the figure in cm−1.

Close modal

To investigate this spectrum further, computational studies were performed at the CCSD(T), B3LYP, and MP2 levels of theory, with basis sets as described in Sec. II B. Consistent across all levels of theory, the structure determined for the H2CO+ ground state is planar. Calculations performed at the CCSD(T) level of theory gave C—H and C = O bond distances of 1.11 and 1.20 Å, respectively and an H—C—H angle of 121.2°. Similar parameters were found with the B3LYP and MP2 levels of theory and are reported in the supplementary material. For comparison to this, the neutral formaldehyde molecule is computed with CCSD(T) level of theory to have C—H and C = O bond distances of 1.10 and 1.21 Å, and an H—C—H angle of 116.4°. These computed values agree reasonably well with the known experimental values for neutral formaldehyde of 1.11 and 1.21 Å, and 116.1°,75 confirming the validity of these levels of theory for the structures of these systems. Because the experiment is done on the argon-tagged ions, we also computed the structures of these ions. Two isomers were found with different binding sites for the argon, (1) above the plane of the molecular ion and (2) attached to a hydrogen atom. The latter structure is lower in energy than the former by 0.4 kcal/mol. As shown in the supplementary material, the binding of argon has little effect on the structure of the formaldehyde ion.

To investigate the spectroscopy, we compared the computed vibrational frequencies for the neutral formaldehyde molecule to the known experimental values72 to derive scaling factors for the symmetric and antisymmetric stretching vibrations appropriate for this level of theory. The factor obtained (0.947) was the same for both C—H stretching modes and this value is then used to produce a scaled vibrational spectrum for the tagged cations for comparison to the experiment. The resulting predicted scaled harmonic spectra for the two argon-binding isomers are also shown in Figure 1. These calculations on neutrals versus cations also help to explain why we did not detect the C = O stretch vibration. Whereas this vibration has reasonable IR intensities for the neutral (67 km/mol), in the cation its intensity is smaller (15.7 km/mol). For comparison, the C—H stretches detected here have computed intensities of 728.6 and 214.2 km/mol. The computed binding energy of argon is about 1000 cm−1, and so this may also be a factor in limiting our photodissociation sensitivity at lower energies.

As shown in Figure 1, the isomers of the formaldehyde cation with argon binding in different positions both have two main bands in the region of the experiment. The isomer with argon above the plane of the molecule has its two bands at higher frequency than those of the isomer with argon binding to hydrogen. These frequencies are quite close to those of neutral formaldehyde. The isomer with argon bound to one of the hydrogen atoms has bands that are more red shifted and the lower frequency feature has a greater intensity. The additional red shift that occurs for hydride stretches when argon binds here has been seen for several other organic ions studied previously.22,23,26,28,30 It is clear from the band positions and their relative intensities that the isomer with argon on hydrogen agrees better with the two main bands in the measured spectrum. This is the more stable isomer, although the computed energy difference between these argon-binding isomers is very small. The scaled harmonic spectrum agrees reasonably well with the experimental spectrum in terms of the spacings of the two most intense bands and their relative intensities. We therefore conclude that the 2423 and 2653 cm−1 bands are most likely the symmetric and asymmetric C—H stretches of the formaldehyde cation with argon attached on a CH. However, the frequencies are calculated to be 80–100 cm−1 higher than those observed for these features. Interestingly, our prediction for the frequency of the symmetric stretch of the formaldehyde cation without the argon tag is also higher than the experimental value from photoelectron spectroscopy by a similar amount (2665 cm−1 predicted vs 2580 cm−1 observed67), supporting our assignments. The scaling derived from the neutral spectroscopy is therefore not adequate for a quantitative description of the cation vibrations. Our measured position for the symmetric stretch of the argon complex (2423 cm−1) is red shifted from the frequency of this vibration in the photoelectron spectroscopy of the isolated ion without argon (2580 cm−1), providing an experimental determination of the shift induced by the argon tagging.

The scaled harmonic spectrum also does not explain the several weaker features in the experimental spectrum. These additional features could be coming from combination bands or overtones that pick up intensity through Fermi resonances. They could also arise from combination bands involving large amplitude vibrational motion of the Ar tag interacting with the CH stretch fundamentals. It is also conceivable that another isomer of [H2, C, O]+ is present. As noted above, other isomers of this ion have been suggested in previous work. We therefore investigated other structures of [H2, C, O]+ and the transition states connecting these species. Figure 2 and Table I present the computed energetics for this system. As indicated, the formaldehyde cation H2CO+ is the global minimum energy structure. The oxonium ion COH2+ lies much higher in energy (+57.8 kcal/mol) than this. The cis and trans conformers of hydroxymethylene HOCH+ lie only slightly higher in energy than the formaldehyde structure (+9.6 and +6.1 kcal/mol respectively), but are separated from it by a large activation barrier (+43.0 kcal/mol relative to H2CO+). Although these other structures are clearly less stable than the formaldehyde structure, it is still possible that they could be formed in the discharge source used here. If they are formed, the barriers between different minima could cause these other isomers to be cooled and stabilized. In a similar way, we have detected multiple isomers in the spectra of several other small organic ions in previous work.23,26,28–30 In the most extreme case, we found evidence for the methoxy cation co-existing with the protonated formaldehyde cation even though the former was over 90 kcal/mol higher in energy.28 

TABLE I.

The relative energies of stationary points on the potential energy surface of [C, H2, O]+ ⋅ with and without the argon tag. Calculations were performed at the CCSD(T) level with the ANO1 basis set (without tag) and with the ANO1 basis set for C, H, and O and the Roos augmented double ζ basis set for Ar (with tag). Relative energies (kcal/mol) are corrected for zero-point vibrational energy.

IonΔE [CCSD(T)]
CH2O+ 0.0 
trans HCOH+ +6.1 
cis HCOH+ +9.6 
COH2+ +57.8 
T.S. 1 +43.0 
T.S. 2 +88.2 
CH2O+Ar (a) (Ar on H) 0.0 
CH2O+Ar (b) (Ar above plane) +0.4 
trans HCOH+Ar (a) +3.3 
trans HCOH+Ar (b) +6.8 
cis HCOH+Ar +6.8 
COH2+Ar +54.5 
IonΔE [CCSD(T)]
CH2O+ 0.0 
trans HCOH+ +6.1 
cis HCOH+ +9.6 
COH2+ +57.8 
T.S. 1 +43.0 
T.S. 2 +88.2 
CH2O+Ar (a) (Ar on H) 0.0 
CH2O+Ar (b) (Ar above plane) +0.4 
trans HCOH+Ar (a) +3.3 
trans HCOH+Ar (b) +6.8 
cis HCOH+Ar +6.8 
COH2+Ar +54.5 
FIG. 2.

The schematic of the potential energy surface for formaldehyde cation, including the cis- and trans-hydroxymethylene and oxonium cations. Energies in kcal/mol calculated at the CCSD(T)/ANO1 level of theory/basis are given relative to the global minimum formaldehyde cation in black, and barriers relative to the trans-hydroxymethylene structure are given in red.

FIG. 2.

The schematic of the potential energy surface for formaldehyde cation, including the cis- and trans-hydroxymethylene and oxonium cations. Energies in kcal/mol calculated at the CCSD(T)/ANO1 level of theory/basis are given relative to the global minimum formaldehyde cation in black, and barriers relative to the trans-hydroxymethylene structure are given in red.

Close modal

To check for these other isomers as possible contributors to the features in Figure 1, we performed calculations on the argon complexes of the oxonium ion and the cis/trans hydroxymethylene species. The relative energies of these species are also presented in Table I. Scaled vibrational frequencies for each of these were used to generate predicted spectra, which are compared to the formaldehyde cation spectrum in Figure 3. As shown, the vibrations for these other isomers fall at frequencies higher than the bands measured, and therefore these other structures cannot account for the additional weak features. Figure S1 in the supplementary material shows that the mass peaks measured in this experiment are extremely weak. We therefore also considered the possible production of isobaric impurity ions (C2H6+ and CH2NH2+) and their argon complexes. Computations (see supplementary material) show that the spectra for these ions also do not agree with anything measured here. This spectrum must therefore be that of the formaldehyde cation, but with additional bands present other than its expected fundamentals.

FIG. 3.

The infrared photodissociation spectrum of [H2, C, O+]Ar produced from a formaldehyde precursor compared to spectra obtained at the CCSD(T)/ANO1 level of theory/basis for the oxonium and hydroxymethylene cations. The positions of the main bands are provided on the figure in cm−1.

FIG. 3.

The infrared photodissociation spectrum of [H2, C, O+]Ar produced from a formaldehyde precursor compared to spectra obtained at the CCSD(T)/ANO1 level of theory/basis for the oxonium and hydroxymethylene cations. The positions of the main bands are provided on the figure in cm−1.

Close modal

Having ruled out the presence of other isomers, we decided to investigate anharmonic effects and Fermi resonances as possible sources for the additional weaker vibrational bands in this spectrum. Figure 4 shows the experimental spectrum from the formaldehyde precursor compared to the spectra predicted using various anharmonic approaches. We first used vibrational perturbation theory (VPT2) calculations as implemented in CFOUR.71 In this approach, argon-binding isomer geometries are optimized at the CCSD(T)/ANO1 level and then a subsequent VPT2 analysis is performed at the MP2/ANO1 level, and the anharmonic corrections obtained are applied to the frequencies obtained from the initial CCSD(T) optimization. As shown in the second trace of Figure 4 (magenta), this approach produces the same two bands seen before for the isomer with argon attached to hydrogen, but with small shifts to lower frequencies compared to the scaled harmonic spectrum in Figure 1 (middle trace). The positions of these two bands are in better agreement with those in the spectrum, but they are still significantly higher in frequency. Additional VPT2 calculations were performed at the B3LYP (green trace in Figure 4) and MP2 levels of theory with the 6-311G(d,p) basis set as implemented in Gaussian 09.73 Scaled harmonic frequencies calculated with density functional theory (DFT) and the B3LYP XC functional were found to be only slightly blue shifted from the experimentally observed peaks. Anharmonic frequencies, computed with VPT2, for both B3LYP and MP2 were found to be slightly red shifted from the observed peaks. Regardless of the level of theory, the relative intensities of these two bands computed with the anharmonic VPT2 level of theory are in poorer agreement with the experimental spectrum and no additional bands are predicted that can account for the other weaker features. A similar treatment of the cation without argon produces a symmetric stretch frequency (2636 cm−1; see supplementary material) that is also higher than the frequency observed in the photoelectron spectrum (2580 cm−1).67 This level of anharmonic theory is therefore also inadequate to describe these cation spectra.

FIG. 4.

The infrared photodissociation spectrum of [H2, C, O+]Ar produced from a formaldehyde precursor compared to spectra predicted at the level of theory noted for the formaldehyde cation using the various forms of anharmonic theory described in the text. The positions of the main bands are provided on the figure in cm−1.

FIG. 4.

The infrared photodissociation spectrum of [H2, C, O+]Ar produced from a formaldehyde precursor compared to spectra predicted at the level of theory noted for the formaldehyde cation using the various forms of anharmonic theory described in the text. The positions of the main bands are provided on the figure in cm−1.

Close modal

To pursue this further, we carried out a reduced dimensional adiabatic coupling scheme like that described previously.28,30 There are two large amplitude intermolecular vibrations with frequencies close to the 100 cm−1 peak spacing between the dominant 2423 cm−1 and the 2523 cm−1 peaks in the top trace in Figure 4. This suggests the possibility that the 2523 cm−1 band reflects a combination involving CH stretching and one or both in-plane intermolecular vibrations of the Ar atom. This simple treatment, which is described more fully in the supplementary material, produces a partial spectrum of just this region (fourth trace of Figure 4 in blue). Such a combination band therefore seems to be a reasonable assignment for this band. It appears 100 cm−1 above the 2423 cm−1 band assigned to the symmetric C—H stretch, and the (unscaled) harmonic frequency computed for the argon stretch is 91 cm−1.

In another approach, we used a Fermi coupling scheme involving three vibrational levels: the symmetric C—H stretch (ν4), a combination band of the C—O stretch (ν5) with an in-plane hindered rotation of H2CO+7), and lastly, the overtone of the HCH bending (ν6). In this analysis, the zero-order frequencies were shifted to agree with the measured band origins, while the coupling constants were obtained within the VPT2 calculation, performed at the B3LYP level of theory. As shown in the lower trace of Figure 4 (red), this scheme produces (also a partial spectrum in just the C—H stretching region) a triplet of bands at 2375, 2413, and 2465 cm−1 that match nicely in position and relative intensities with the experimental bands at 2377, 2423, and 2460 cm−1. It therefore seems that a Fermi triad provides a reasonable explanation for these bands. The combined effects of combination bands and Fermi resonances therefore seem to be adequate to explain all the minor features in this spectrum of the formaldehyde cation.

To explore the possible formation of the other isomers implicated in previous experimental and computational work, we made the m/z = 30 cation from a methanol precursor. The fragmentation of methanol via electron impact ionization produces the mass 30 ion as a prominent fragment,76 and previous ion chemistry suggested that methanol ionization and fragmentation produce the hydroxymethylene cation.59–61 Discharges of methanol vapor in argon produce the protonated methanol cation, the methanol radical cation that we have studied previously,30 and the desired m/z = 30 cation, including their respective argon tagged species (see mass spectrum in Figure S2 of the supplementary material). Figure 5 shows the infrared spectrum of the m/z = 30 cations produced in this way compared to that produced with the formaldehyde precursor. The spectrum was measured between 1000 and 4000 cm−1, but signal was only detected between 2750 and 3050 cm−1. There are two prominent bands at 2757 and 2803 cm−1, as well as a series of weaker features extending to higher frequencies. Most striking, however, is that these bands do not match those measured from the formaldehyde precursor. Apparently, methanol fragmentation produces a different structure for the m/z = 30 cation. As shown in the figure, very weak features in the formaldehyde precursor spectrum line up with the strongest two bands in the methanol precursor spectrum. This could result if a small amount of the other isomer is produced from formaldehyde. However, the aqueous formaldehyde solution used has a small amount of methanol added to it as a stabilizer, and so these weak bands could also be produced from that small amount of methanol.

FIG. 5.

The infrared photodissociation spectrum of [H2, C, O+]Ar produced from a methanol precursor compared to that produced from a formaldehyde precursor. The two precursors produce different spectra corresponding to different ion structures. The positions of the main bands are provided on the figure in cm−1.

FIG. 5.

The infrared photodissociation spectrum of [H2, C, O+]Ar produced from a methanol precursor compared to that produced from a formaldehyde precursor. The two precursors produce different spectra corresponding to different ion structures. The positions of the main bands are provided on the figure in cm−1.

Close modal

Referring to the computational studies described above, and the predicted spectra shown in Figure 3, we can see that the most intense oxonium ion vibration and the bands predicted for the hydroxymethylene cations all fall in the wavelength region of the peaks measured here. However, the oxonium ion has a higher frequency band that is not detected, and it lies much higher in energy compared to the hydroxymethylene isomers. Hydroxymethylene was found in reaction studies of the ions produced from methanol.59–61 We therefore present the predicted spectra for the cis- and trans-hydroxymethylene cation complexes with argon again here below the experimental spectrum in Figure 6. Both conformers of this isomer are predicted to have two nearly overlapping bands in the 2800–3000 cm−1 region, with those belonging to the cis species lying lower in frequency. The more intense band for each is the O—H stretch where the argon attaches on the hydrogen, and the weaker band is the C—H stretch. Even though the O—H stretch is at higher frequency for the isolated molecule, the argon binds on this hydrogen, shifting this frequency lower than the C—H stretch in the argon complex. As shown in Figure 6, the band positions for the two hydroxymethylene species fall in the correct frequency region, but neither conformer alone can explain the measured spectrum. However, if we assume that both cis- and trans-hydroxymethylene species are present in roughly equal amounts, we can account for two strong bands in this region. This would not explain the weaker features at higher frequencies, but it would explain the two most prominent bands. Hydroxymethylene cation has been predicted in previous theoretical investigations,62–64 and suggested as a component in ion chemistry studies,59–61 but its spectrum has never been measured.

FIG. 6.

The infrared photodissociation spectrum of [H2, C, O+]Ar produced from a methanol precursor compared to that predicted by scaled harmonic theory at the CCSD(T)/ANO1 level for the cis- and trans-hydroxymethylene cations. The positions of the main bands are provided on the figure in cm−1.

FIG. 6.

The infrared photodissociation spectrum of [H2, C, O+]Ar produced from a methanol precursor compared to that predicted by scaled harmonic theory at the CCSD(T)/ANO1 level for the cis- and trans-hydroxymethylene cations. The positions of the main bands are provided on the figure in cm−1.

Close modal

To explore this spectrum further, we employed the anharmonic theory methods described above for the formaldehyde cation to the hydroxymethylene ions. Figure 7 shows the results of the CCSD(T)/VPT2 calculations for both cis- and trans-hydroxymethylene compared to the experimental spectrum generated with the methanol discharge. Both conformers have a predicted intense O—H stretch in the region of the two strong bands in the spectrum. The trans species has its predicted band at higher frequency within 12 cm−1 of the 2803 cm−1 experimental band, and the cis has its predicted band at 2706 cm−1, 51 cm−1 lower in frequency than the 2757 cm−1 band. Both conformers have weaker C—H stretches predicted at higher frequencies (2933 and 2909 cm−1 for trans and cis respectively; see Tables S9 and S10 in the supplementary material). The effect of anharmonicity is therefore mainly to separate the O—H and C—H stretches compared to their positions predicted in the harmonic spectra, mainly by shifting the O—H stretch much further to the red. Both calculated spectra show a combination band that is roughly 200 cm−1 to the blue of the OH fundamental. These combination bands are assigned to transitions to states with one quantum in both the O—H stretch and the Ar—H stretch. A third feature is evident in the calculated spectrum for the trans conformer and corresponds to the C—H stretch fundamental. Interestingly this feature is suppressed in the calculated spectrum for the cis-conformer. Given the similar frequencies of the C—H and O—H stretch fundamentals, it is expected that the calculated intensity will be very sensitive to the extent of mixing of these two vibrations at the harmonic level as well as possible contributions to the intensity due to quartic terms in the potential that couple the two fundamentals. These latter effects are not included in second order perturbation theory, while the amount of mixing of the normal modes is found to be sensitive to the level of electronic structure theory that is used.

FIG. 7.

The infrared photodissociation spectrum of [H2, C, O+]Ar produced from a methanol precursor compared to that predicted by anharmonic theory at the CCSD(T)/ANO1/VPT2 level for the cis- and trans-hydroxymethylene cations. The positions of the main bands are provided on the figure in cm−1.

FIG. 7.

The infrared photodissociation spectrum of [H2, C, O+]Ar produced from a methanol precursor compared to that predicted by anharmonic theory at the CCSD(T)/ANO1/VPT2 level for the cis- and trans-hydroxymethylene cations. The positions of the main bands are provided on the figure in cm−1.

Close modal

We also investigated the anharmonic spectrum (VPT2) obtained using B3LYP/6-311G(d,p), as presented in the supplementary material (Figure S16). This spectrum has similar combination bands, but with slightly different positions and relative intensities. Both anharmonic approaches therefore agree that the O—H and C—H stretches should be further separated than they are in the harmonic theory and that combination bands involving both the O—H stretch and the argon stretch should also be observed. The agreement with specific band positions is not perfect with either approach. However, if we take the overall result from theory that there should be two strong O—H stretches, two weaker C—H stretches, and one or two combination bands in the higher frequency region, overlapping spectra from cis and trans hydroxymethylene isomers in roughly equal abundances would seem to be adequate to explain the spectrum observed. Methanol ionization and fragmentation therefore produce hydroxymethylene exclusively, and the cold jet conditions do not allow any rearrangement to the more stable formaldehyde cation structure.

The results here show that formaldehyde ionization in the discharge source produces formaldehyde cation and that ionization followed by fragmentation of methanol produces a different ion, the hydroxymethylene cation, in roughly equal abundances for its cis and trans conformers. The final band assignments for both isomers are provided in Table II. The direct ionization of formaldehyde to produce its cation is not surprising, since this is the species favored by Franck-Condon overlap with the ground state and this is also the lowest energy structure for the cation. The hydroxymethylene cation has been suggested to form from methanol ion fragmentation on the basis of energy dependence of the ion yields and collisional dissociation of the ions formed.59–61 Its corresponding neutral is well known in formaldehyde photochemistry,77 and its infrared spectrum has recently been measured in matrix isolation78 and in helium droplets.79 Here we detect this cation directly with infrared spectroscopy. It is understandable that this ion should form from the fragmentation of methanol cation, because breaking its C—H bonds is a lower energy process than breaking the O—H bond.80 Likewise, there should be no particular preference for which C—H bonds break, and so the roughly equal amounts of cis and trans conformers also make sense. Interestingly, the neutral experiments on hydroxymethylene have only detected the trans conformer,78,79 whereas we detect both cis and trans species here for the cation.

TABLE II.

Observed positions and assignments for the bands observed in the spectra of ions formed from the formaldehyde solution or methanol precursors.

Band position (cm−1)Assignment
Formaldehyde solution precursor/formaldehyde cation 
2377 Member of Fermi triada 
2423 Member of Fermi triada 
2460 Member of Fermi triada 
2523 C—H sym. str. + Ar—H str. 
2653 C—H asym. Str. 
2799 trans HCOH+ O—H str. 
Methanol precursor/hydroxymethylene cation 
2757 cis HCOH+ O—H str. 
2803 trans HCOH+ O—H str. 
2833 cis HCOH+ C—H str. 
2892 cis HCOH+ O—H str. Ar—H str. 
2952 trans HCOH+ C—H str. 
3002 trans HCOH+ O—H str. + Ar—H str. 
Band position (cm−1)Assignment
Formaldehyde solution precursor/formaldehyde cation 
2377 Member of Fermi triada 
2423 Member of Fermi triada 
2460 Member of Fermi triada 
2523 C—H sym. str. + Ar—H str. 
2653 C—H asym. Str. 
2799 trans HCOH+ O—H str. 
Methanol precursor/hydroxymethylene cation 
2757 cis HCOH+ O—H str. 
2803 trans HCOH+ O—H str. 
2833 cis HCOH+ C—H str. 
2892 cis HCOH+ O—H str. Ar—H str. 
2952 trans HCOH+ C—H str. 
3002 trans HCOH+ O—H str. + Ar—H str. 
a

Fermi triad is comprised of the C = O stretch, combination band of C = O stretch and CH rock, and an overtone of the HCH bend.

As noted above, formaldehyde cation is of significant interest for interstellar chemistry. Both neutral formaldehyde and protonated formaldehyde have been detected in space, but not the formaldehyde cation, even though it is implicated to be there on the basis of reaction models of the interstellar chemistry.8 The present study confirms that this species can be produced, but also shows how its other isomer can also be formed, depending on the precursor employed. The spectra measured here are not of sufficiently high resolution to guide interstellar searches directly. Additionally, the O—H stretch fundamentals are severely shifted by the argon attachment. The C—H stretches are less perturbed, but are also shifted from those of the free molecule. On the other hand, as the Fermi interactions described above all involve intramolecular vibrations of H2CO+, they are expected to persist even in the absence of the argon atom. The limitations of studying such tagged ions are by now understood. However, our results also show that scaled harmonic theory, even when conducted at high levels on the argon-free cation, is likewise inadequate to predict the vibrations of this ion. It is our hope that the low resolution data provided here can guide higher resolution studies in other labs. Also relevant for interstellar chemistry is the likely abundance of these isomeric species in space. The neutral hydroxymethylene isomer has been produced in rare gas matrices and found to interconvert to the more stable formaldehyde isomer via tunneling on the time scale of several hours, even though the barrier to this interconversion was about 30 kcal/mol.78 Here, the barrier for hydroxymethylene-to-formaldehyde conversion is higher, but on the time scales of interstellar events its tunneling to form the more stable formaldehyde cation is also likely to occur. Unless there is an efficient path producing copious amounts of hydroxymethylene cation, interstellar studies would likely find the formaldehyde cation isomer.

Cations with m/z = 30 were produced with a discharge/ expansion source and were investigated using mass-selected infrared photodissociation spectroscopy and the method of argon tagging. The infrared spectra contain bands between 2350 and 2850 cm−1 when ions are produced from a formaldehyde solution precursor and different bands between 2700 and 3100 cm−1 are produced from a methanol precursor. Detailed analysis of these spectra indicates that the formaldehyde solution precursor produces the formaldehyde cation whereas the methanol precursor produces cis- and trans-hydroxymethylene cations. The theory indicates that the relative energy difference between these isomers is low, but large barriers prevent the rearrangement of the metastable hydroxymethylene cations to the more stable formaldehyde cation structure.

See supplementary material for calculated structures, energies, vibrational frequencies, and intensities of the described ions and additional information about the anharmonic calculations.

The authors gratefully acknowledge funding for this research by the National Science Foundation (M.A.D. Grant No. CHE-1464708 and A.B.M. Grant No. CHE-1619660). We appreciate helpful discussions with Professor John Stanton in the early stages of this work.

1.
Gas Phase Ion Chemistry
, edited by
M. T.
Bowers
(
Academic Press
,
New York
,
1984
).
2.
T.
Baer
,
C.
Ng
, and
I.
Powis
,
The Structure, Energetics, and Dynamics of Organic Ions
(
Wiley
,
New York
,
1996
).
3.
J. L.
Holmes
,
C.
Aubrey
, and
P. M.
Mayer
,
Assigning Structures to Ions in Mass Spectrometry
(
CRC Press
,
Boca Raton
,
2007
).
4.
The Molecular Astrophysics of Stars and Galaxies
, edited by
T. W.
Hartquist
and
D. A.
Williams
(
Clarendon Press
,
Oxford
,
1998
).
5.
A. G. G. M.
Tielens
,
The Physics and Chemistry of the Interstellar Medium
(
Cambridge University Press
,
Cambridge, UK
,
2005
).
6.
S.
Petrie
and
D. K.
Bohme
,
Mass Spectrom. Rev.
26
,
258
(
2007
).
7.
T. P.
Snow
and
V. M.
Bierbaum
,
Annu. Rev. Anal. Chem.
1
,
229
(
2008
).
9.
K. M.
Stirk
,
L. K. M.
Kiminkinen
, and
H. I.
Kenttämaa
,
Chem. Rev.
92
,
1649
(
1992
).
10.
11.
The Spectroscopy of Molecular Ions
, edited by
A.
Carrington
and
B. A.
Thrush
(
The Royal Society
,
London
,
1988
).
12.
T.
Ebata
,
A.
Fujii
, and
N.
Mikami
,
Int. Rev. Phys. Chem.
17
,
331
(
1998
).
13.
E. J.
Bieske
and
O.
Dopfer
,
Chem. Rev.
100
,
3963
(
2000
).
14.
M. A.
Duncan
,
Int. J. Mass Spectrom.
200
,
545
(
2000
).
15.
M. A.
Duncan
,
Int. Rev. Phys. Chem.
22
,
407
(
2003
).
16.
W. H.
Robertson
and
M. A.
Johnson
,
Annu. Rev. Phys. Chem.
54
,
173
(
2003
).
17.
T. R.
Rizzo
,
J. A.
Stearns
, and
O. V.
Boyarkin
,
Int. Rev. Phys. Chem.
28
,
481
(
2009
).
18.
T.
Baer
and
R. C.
Dunbar
,
J. Am. Soc. Mass Spectrom.
21
,
681
(
2010
).
19.
R.
Nagarajan
and
J. P.
Maier
,
Int. Rev. Phys. Chem.
29
,
521
(
2010
).
20.
A. B.
Wolk
,
C. M.
Leavitt
,
E.
Garand
, and
M. A.
Johnson
,
Acc. Chem. Res.
47
,
202
(
2014
).
21.
J.
Roithová
,
A.
Gray
,
E.
Andris
,
J.
Jašík
, and
D.
Gerlich
,
Acc. Chem. Res.
49
,
223
(
2016
).
22.
G. E.
Douberly
,
A. M.
Ricks
,
B. W.
Ticknor
,
W. C.
McKee
,
P. v. R.
Schleyer
, and
M. A.
Duncan
,
J. Phys. Chem. A
112
,
1897
(
2008
).
23.
G. E.
Douberly
,
A. M.
Ricks
,
P. v. R.
Schleyer
, and
M. A.
Duncan
,
J. Chem. Phys.
128
,
021102
(
2008
).
24.
G. E.
Douberly
,
A. M.
Ricks
,
P. v. R.
Schleyer
, and
M. A.
Duncan
,
J. Phys. Chem. A
112
,
4869
(
2008
).
25.
A. M.
Ricks
,
G. E.
Douberly
, and
M. A.
Duncan
,
Astrophys. J.
702
,
301
(
2009
).
26.
A. M.
Ricks
,
G. E.
Douberly
,
P. v. R.
Schleyer
, and
M. A.
Duncan
,
J. Chem. Phys.
132
,
051101
(
2010
).
27.
M. A.
Duncan
,
J. Phys. Chem. A
116
,
11477
(
2012
).
28.
J. D.
Mosley
,
T. C.
Cheng
,
A. B.
McCoy
, and
M. A.
Duncan
,
J. Phys. Chem. A
116
,
9287
(
2012
).
29.
J. D.
Mosley
,
J. W.
Young
, and
M. A.
Duncan
,
J. Chem. Phys.
141
,
024306
(
2014
).
30.
J. D.
Mosley
,
J. W.
Young
,
M.
Huang
,
A. B.
McCoy
, and
M. A.
Duncan
,
J. Chem. Phys.
142
,
114301
(
2015
).
31.
M. W.
Crofton
,
M. F.
Jagod
,
B. D.
Rehfuss
,
W. A.
Kreiner
, and
T.
Oka
,
J. Chem. Phys.
88
,
666
(
1988
).
32.
M.
Rosslein
,
C. M.
Gabrys
,
M. F.
Jagod
, and
T.
Oka
,
J. Mol. Spec.
153
,
738
(
1992
).
33.
M. W.
Crofton
,
M. F.
Jagod
,
B. D.
Rehfuss
, and
T.
Oka
,
J. Chem. Phys.
91
,
5139
(
1989
).
34.
M. F.
Jagod
,
M.
Rosslein
,
C. M.
Gabrys
,
B. D.
Rehfuss
,
F.
Scappini
,
M. W.
Crofton
, and
T.
Oka
,
J. Chem. Phys.
97
,
7111
(
1992
).
35.
C. M.
Gabrys
,
D.
Uy
,
M. F.
Jagod
,
T.
Oka
, and
T.
Amano
,
J. Phys. Chem.
99
,
15611
(
1995
).
36.
E. T.
White
,
J.
Tang
, and
T.
Oka
,
Science
284
,
135
(
1999
).
37.
M.
Okumura
,
L. I.
Yeh
,
J. D.
Myers
, and
Y. T.
Lee
,
J. Chem. Phys.
85
,
2328
(
1986
).
38.
M.
Okumura
,
L. I.
Yeh
,
J. D.
Myers
, and
Y. T.
Lee
,
J. Phys. Chem.
94
,
3416
(
1990
).
39.
M.
Okumura
,
L. I.
Yeh
, and
Y. T.
Lee
,
J. Chem. Phys.
83
,
3075
(
1985
).
40.
L. I.
Yeh
,
J. M.
Price
, and
Y. T.
Lee
,
J. Am. Chem. Soc.
111
,
5597
(
1989
).
41.
D. W.
Boo
and
Y. T.
Lee
,
Chem. Phys. Lett.
211
,
358
(
1993
).
42.
X.
Huang
,
A. B.
McCoy
,
J. M.
Bowman
,
L. M.
Johnson
,
C.
Savage
,
F.
Dong
, and
D. J.
Nesbitt
,
Science
311
,
60
(
2006
).
43.
S. A.
Nizkorodov
,
O.
Dopfer
,
T.
Ruchti
,
M.
Meuwly
,
J. P.
Maier
, and
E. J.
Bieske
,
J. Phys. Chem.
99
,
17118
(
1995
).
44.
S. A.
Nizkorodov
,
O.
Dopfer
,
M.
Meuwly
,
J. P.
Maier
, and
E. J.
Bieske
,
J. Chem. Phys.
105
,
1770
(
1996
).
45.
N.
Solcà
and
O.
Dopfer
,
J. Am. Chem. Soc.
126
,
1716
(
2004
).
46.
N.
Solcà
and
O.
Dopfer
,
J. Am. Chem. Soc.
126
,
9520
(
2004
).
47.
O.
Dopfer
,
R. V.
Olkhov
,
M.
Mladenović
, and
P.
Botschwina
,
J. Chem. Phys.
121
,
1744
(
2004
).
48.
S.
Chakraborty
,
A.
Patzer
, and
O.
Dopfer
,
J. Chem. Phys.
133
,
044307
(
2010
).
49.
B.
Chiavarino
,
M. E.
Crestoni
,
O.
Dopfer
,
P.
Maître
, and
S.
Fornarini
,
Angew. Chem.
124
,
5031
(
2012
).
50.
O.
Asvany
,
P.
Kumar
,
B.
Redlich
,
I.
Hegemann
,
S.
Schlemmer
, and
D.
Marx
,
Science
309
,
1219
(
2005
).
51.
O.
Asvany
,
K. M. T.
Yamada
,
S.
Brünken
,
A.
Potapov
, and
S.
Schlemmer
,
Science
347
,
1346
(
2015
).
52.
J.
Jašík
,
J.
Žabka
,
J.
Roithová
, and
D.
Gerlich
,
Int. J. Mass Spectrom.
354-355
,
204
(
2013
).
53.
J.
Jašík
,
D.
Gerlich
, and
J.
Roithová
,
J. Am. Chem. Soc.
136
,
2960
(
2014
).
54.
J.
Jašík
,
D.
Gerlich
, and
J.
Roithová
,
J. Phys. Chem. A
119
,
2532
(
2015
).
55.
V.
Dryza
,
N.
Chalyavi
,
J. A.
Sanelli
, and
E. J.
Bieske
,
J. Chem. Phys.
137
,
204304
(
2012
).
56.
N.
Chalyavi
,
J. A.
Sanelli
,
V.
Dryza
, and
E. J.
Bieske
,
J. Phys. Chem. A
117
,
11276
(
2013
).
57.
K. J.
Catani
,
J. A.
Sanelli
,
V.
Dryza
,
N.
Gilka
,
P. R.
Taylor
, and
E. J.
Bieske
,
J. Chem. Phys.
143
,
184306
(
2015
).
58.
J. H.
Beynon
,
A. E.
Fontaine
, and
G. R.
Lester
,
Int. J. Mass Spectrom.
1
,
1
(
1968
).
59.
J.
Berkowitz
,
J. Chem. Phys.
69
,
3044
(
1978
).
60.
C.
Wesdemiotis
and
F. W.
McLafferty
,
Tetrahedron Lett.
22
,
3479
(
1981
).
61.
P. C.
Burgers
,
A. A.
Mommers
, and
J. L.
Holmes
,
J. Am. Chem. Soc.
105
,
5976
(
1983
).
62.
W. J.
Bouma
,
J. K.
Macleod
, and
L.
Radom
,
Int. J. Mass Spectrom. Ion Proc.
33
,
87
(
1980
).
63.
N. L.
Ma
,
B. J.
Smith
, and
L.
Radom
,
Chem. Phys. Lett.
193
,
386
(
1992
).
64.
Y.
Osamura
,
J. D.
Goddard
,
H. F.
Schaefer
III
, and
K. S.
Kim
,
J. Chem. Phys.
74
,
617
(
1981
).
65.
O.
Wiest
,
J. Mol. Struct.: THEOCHEM
368
,
39
(
1996
).
66.
D. W.
Turner
,
C.
Baker
,
A. D.
Baker
, and
C. R.
Bundle
,
Molecular Photoelectron Spectroscopy
(
Wiley Interscience
,
London
,
1970
), p.
132
.
67.
B.
Niu
,
D. A.
Shirley
,
Y.
Bai
, and
E.
Daymo
,
Chem. Phys. Lett.
201
,
212
(
1993
).
68.
A. M.
Schulenburg
,
M.
Meisinger
,
P. P.
Radi
, and
F.
Merkt
,
J. Mol. Spectrosc.
250
,
44
(
2008
).
69.
S. P.
Mishra
and
M. C.
Symons
,
J. Chem. Soc. Chem. Comm.
1975
,
909
.
70.
L. B.
Knight
and
J.
Steadman
,
J. Chem. Phys.
80
,
1018
(
1984
).
71.
CFOUR, a quantum chemical program package written by J. F. Stanton, J. Gauss, M. E. Harding, and P. G. Szalay, with contributions from A. A. Auer, R. J. Bartlett, U. Benedikt, C. Berger, D. E. Bernholdt, Y. J. Bomble, L. Cheng, O. Christiansen, M. Heckert, O. Heun, C. Huber, T.-C. Jagau, D. Jonsson, J. Jusélius, K. Klein, W. J. Lauderdale, D. A. Matthews, T. Metzroth, L. A. Mück, D. P. O’Neill, D. R. Price, E. Prochnow, C. Puzzarini, K. Ruud, F. Schiffmann, W. Schwalbach, S. Stopkowicz, A. Tajti, J. Vázquez, F. Wang, and J. D. Watts; with the integral packages MOLECULE (by J. Almlöf and P. R. Taylor), PROPS (by P. R. Taylor), ABACUS (by T. Helgaker, H. J. Aa. Jensen, P. Jørgensen, and J. Olsen), and ECP routines (by A. V. Mitin and C. van Wüllen), 2010.
72.
T.
Shimanouchi
, in
NIST Chemistry WebBook
,
NIST Standard Reference Database Number 69
, edited by
P. J.
Linstrom
and
W. G.
Mallard
(
National Institute of Standards and Technology
,
Gaithersburg, MD
,
2016
), http://webbook.nist.gov (retrieved June 7, 2016).
73.
M. J.
Frisch
,
G. W.
Trucks
,
H. B.
Schlegel
,
G. E.
Scuseria
,
M. A.
Robb
,
J. R.
Cheeseman
,
G.
Scalmani
,
V.
Barone
,
B.
Mennucci
,
G. A.
Petersson
,
H.
Nakatsuji
,
M.
Caricato
,
X.
Li
,
H. P.
Hratchian
,
A. F.
Izmaylov
,
J.
Bloino
,
G.
Zheng
,
J. L.
Sonnenberg
,
M.
Hada
,
M.
Ehara
,
K.
Toyota
,
R.
Fukuda
,
J.
Hasegawa
,
M.
Ishida
,
T.
Nakajima
,
Y.
Honda
,
O.
Kitao
,
H.
Nakai
,
T.
Vreven
,
J. A.
Montgomery
, Jr.
,
J. E.
Peralta
,
F.
Ogliaro
,
M.
Bearpark
,
J. J.
Heyd
,
E.
Brothers
,
K. N.
Kudin
,
V. N.
Staroverov
,
R.
Kobayashi
,
J.
Normand
,
K.
Raghavachari
,
A.
Rendell
,
J. C.
Burant
,
S. S.
Iyengar
,
J.
Tomasi
,
M.
Cossi
,
N.
Rega
,
J. M.
Millam
,
M.
Klene
,
J. E.
Knox
,
J. B.
Cross
,
V.
Bakken
,
C.
Adamo
,
J.
Jaramillo
,
R.
Gomperts
,
R. E.
Stratmann
,
O.
Yazyev
,
A. J.
Austin
,
R.
Cammi
,
C.
Pomelli
,
J. W.
Ochterski
,
R. L.
Martin
,
K.
Morokuma
,
V. G.
Zakrzewski
,
G. A.
Voth
,
P.
Salvador
,
J. J.
Dannenberg
,
S.
Dapprich
,
A. D.
Daniels
,
Ö.
Farkas
,
J. B.
Foresman
,
J. V.
Ortiz
,
J.
Cioslowski
, and
D. J.
Fox
, gaussian 09, Revision E.01, Gaussian, Inc., Wallingford, CT, 2009.
74.
E. M.
Myshakin
,
K. D.
Jordan
,
E. L.
Sibert
, and
M. A.
Johnson
, “
Large anharmonic effects in the infrared spectra of the symmetrical CH3NO2 (H2 O) and CH3CO2 ⋅ (H2O) complexes
,”
J. Chem. Phys.
119
,
10138
(
2003
).
75.
NIST Computational Chemistry Comparison and Benchmark Database, NIST Standard Reference Database Number 101, Release 17b, edited by R. D. Johnson III, September 2015, http://cccbdb.nist.gov/.
76.
NIST Mass Spec Data Center,
S. E.
Stein, Director
, in
NIST Chemistry WebBook
,
NIST Standard Reference Database Number 69
, edited by
P. J.
Linstrom
and
W. G.
Mallard
(
National Institute of Standards and Technology
,
Gaithersburg, MD
,
2016
), http://webbook.nist.gov, (retrieved June 2016).
77.
C. B.
Moore
and
J. C.
Weisshaar
,
Annu. Rev. Phys. Chem.
34
,
525
(
1983
).
78.
P. R.
Schreiner
,
H. P.
Reisenauer
,
F. C.
Pickard
IV
,
A. C.
Simmonett
,
W. D.
Allen
,
E.
Mátyus
, and
A. G.
Császár
,
Nature
453
,
906
(
2008
).
79.
C. M.
Leavitt
,
C. P.
Moradi
,
J. F.
Stanton
, and
G. E.
Douberly
,
J. Chem. Phys.
140
,
171102
(
2014
).
80.
S. J.
Blanksby
and
G. B.
Ellison
,
Acc. Chem. Res.
36
,
255
(
2003
).

Supplementary Material