Singlet dihydroxycarbene ( HO C ̈ OH ) is produced via pyrolytic decomposition of oxalic acid, captured by helium nanodroplets, and probed with infrared laser Stark spectroscopy. Rovibrational bands in the OH stretch region are assigned to either trans,trans- or trans,cis-rotamers on the basis of symmetry type, nuclear spin statistical weights, and comparisons to electronic structure theory calculations. Stark spectroscopy provides the inertial components of the permanent electric dipole moments for these rotamers. The dipole components for trans, trans- and trans, cis-rotamers are (μa, μb) = (0.00, 0.68(6)) and (1.63(3), 1.50(5)), respectively. The infrared spectra lack evidence for the higher energy cis,cis-rotamer, which is consistent with a previously proposed pyrolytic decomposition mechanism of oxalic acid and computations of HO C ̈ OH torsional interconversion and tautomerization barriers.

Carbenes ( R 1 - C ̈ - R 2 ) are utilized as synthetic scaffolds in modern organometallic chemistry,1–6 and they have long been recognized as reaction intermediates in organic chemistry.7–9 Carbenes also play significant roles in many fundamental areas of chemical physics, including atmospheric,10,11 combustion,12,13 and interstellar chemistry.14–17 Singlet carbenes are composed of a divalent, sp2 hybridized C atom with an out-of-plane, vacant p-orbital, whereas triplet carbenes possess two half-filled p-orbitals.18 The relative stabilities of singlet and triplet electronic states depend partially on the propensity of substituent groups to back-donate electron density to the electron deficient C atom.10,18

There exists a long and rich history of the spectroscopy of carbene systems, including the seminal work on gas-phase methylene19–22 and other triatomic carbenes.10,23–25 Our understanding of the physical chemistry of these elusive species has profited tremendously from cryogenic matrix isolation spectroscopy,26 with the earliest reports on matrix isolated methylene appearing shortly after the introduction of the technique in the late 1950s.27 Indeed, the structure and reactivity of an impressive range of carbene systems have been probed with this method.26–33 

Relevant to the current report, the first direct observation of hydroxymethylene ( H C ̈ OH ) was reported by Schreiner and co-workers via the pyrolysis of glyoxylic acid followed by the isolation of the pyrolysate in an 11 K Argon matrix.34 It was discovered that H C ̈ OH rearranges to formaldehyde with a half-life of only 2 h by hydrogen tunneling, akin to Sheridan and co-workers’ earlier observation of carbon tunneling in matrix isolated 1-methylcyclobutylfluorocarbene.28 Despite this prompt disappearance, Fourier transform (FT) infrared (IR) spectra were obtained and assigned to the trans- H C ̈ OH conformer on the basis of comparisons to high-level anharmonic frequency computations. More recently, H C ̈ OH and its d1-isotopologue ( H C ̈ OD ) were isolated in He droplets at 0.4 K.35 Rotationally resolved spectra of these species isolated in the weakly perturbing He environment enabled an assignment of several bands in the X-H(D) stretching region to the trans- H C ̈ OH ( D ) isomers, in agreement with the predictions of anharmonic frequency computations and consistent with the assignments of the Ar matrix spectra.

Among the series of Ar matrix studies reported by Schreiner and co-workers, dihydroxycarbene, HO C ̈ OH , was shown to be similarly produced via the thermal decomposition of a readily available precursor, oxalic acid (1a and 1b in Fig. 1).36Ab initio computations show that dihydroxycarbene is stabilized by in-plane, π-electron donation from oxygen, implying production of HO C ̈ OH in its ground singlet state.36–38 This stabilizing transfer of electron density into the vacant, out-of-plane p-orbital results in a rather large singlet-triplet energy splitting (computed here to be 61.7 kcal/mol) and substantial torsional interconversion barriers. Computations of the singlet potential energy surface (PES)36,39 revealed three locally stable rotamers (2a, 2b, and 2c). The matrix FTIR spectra of the pyrolytic decomposition products of oxalic acid revealed the presence of trans,trans- and trans,cis- HO C ̈ OH (2a and 2b), but the higher energy cis,cis-rotamer (2c) was absent. Unlike H C ̈ OH , efficient tunneling of HO C ̈ OH to its tautomeric aldehyde, formic acid (3), was not observed.36 

FIG. 1.

Possible products of the pyrolytic decomposition of the cTc and cTt-oxalic acid conformers. The three lowest-lying isomers of dihydroxycarbene are trans,trans-, trans,cis-, and cis,cis-, although only the former two are observed, consistent with previous work.36 Barrier heights (numbers in parentheses) and relative ΔH0 values (bold numbers in parentheses) associated with each oxalic acid rearrangement and decomposition pathway are from Ref. 72. The barrier heights and relative ΔH0 values associated with dihydroxymethylene rotamerization/decomposition are from Ref. 36.

FIG. 1.

Possible products of the pyrolytic decomposition of the cTc and cTt-oxalic acid conformers. The three lowest-lying isomers of dihydroxycarbene are trans,trans-, trans,cis-, and cis,cis-, although only the former two are observed, consistent with previous work.36 Barrier heights (numbers in parentheses) and relative ΔH0 values (bold numbers in parentheses) associated with each oxalic acid rearrangement and decomposition pathway are from Ref. 72. The barrier heights and relative ΔH0 values associated with dihydroxymethylene rotamerization/decomposition are from Ref. 36.

Close modal

More recently, McCarthy and coworkers used a combination of microwave and millimeter wave spectroscopy for the structural determination of the gas-phase trans,cis-rotamer.40 Here, they confirmed a planar, Cs symmetry structure, consistent with previous theoretical computations. Their structural results further illustrate the critical role that the second electron-donating group in HO C ̈ OH plays in making this molecule resistant to isomerization. The apparent stability of this carbene to tunneling suggests a possible role for it in the atmospheric and astrochemistry already discussed for members of its isomeric family, formic acid, and the simplest Criegee intermediate, peroxymethylene.41–44 Extending upon these previous spectroscopic studies, the work described here probes the permanent electric dipole moments of He-solvated HO C ̈ OH using IR laser Stark spectroscopy.

The helium droplet methodology employed here has been described in detail previously.45–48 Nanodroplets of superfluid He are generated in a supercritical, subsonic expansion of He gas (99.9995% purity, 35 bars) through a cold (17 K), 5 μm diameter pinhole nozzle. Under these conditions, droplets consisting of 4500 He atoms on average are produced at a rate of 1012 s−1.45,49,50

The droplets pass into a differentially pumped “pickup” chamber, where gas-phase molecules are captured, solvated, and cooled to ≈0.4 K.51 A low-pressure, effusive pyrolysis source located in the pickup chamber produces gas-phase carbenes by thermal decomposition of organic precursor molecules.35 Gas-phase HO C ̈ OH was generated via pyrolysis of oxalic acid (C2H2O4), analogous to the procedure reported by Schreiner and co-workers.36 The pyrolysis region consisted of a 15 cm long quartz tube with an outer diameter of 0.6 cm, which is oriented perpendicular to the droplet beam. The sealed end of the tube, furthest from the beam, contained ≈1 g of oxalic acid. To achieve sufficient vapor pressure (≈10−5 Torr) in the pickup zone to optimize for the capture of single molecules, the sample was heated to 330 K by a Nichrome wire wrapped around the sealed end of the quartz tube. The temperature in this sample region was monitored by a K-type thermocouple. The pyrolysis region, adjacent to the droplet beam, was wrapped with a single coil of Ta wire, through which high current was passed. Pyrolysis of the precursor molecule and production of HO C ̈ OH were optimized at ≈1000 K (35 A).

The doped droplet beam passes into a Stark/multipass cell consisting of two 15 cm long, parallel, gold-coated mirrors situated on either side of the droplet beam and two parallel, stainless steel electrodes located above and below the beam axis. The electrode spacing is 3.10(2) mm and is calibrated via Stark measurements of He-solvated HCN.52 The tunable idler output from an IR continuous-wave, optical parametric oscillator (OPO) is used to vibrationally excite the molecules embedded in He droplets.53 The IR radiation directed into the multipass cell intersects the droplet beam 20-30 times in an approximately perpendicular arrangement. The polarization of the OPO idler output is aligned parallel to the applied Stark field.

When in resonance with a rovibrational transition of the dopant, the vibrational excitation energy is quenched by the evaporation of He atoms from the droplet surface. One He atom is lost for every ≈5 cm−1 of vibrational energy. This laser-induced reduction in the droplet geometric cross section leads to a reduction in the total ion signal produced upon electron-impact ionization in a downstream quadrupole mass spectrometer. The ion current from the mass spectrometer is processed with a lock-in amplifier as the OPO idler beam is amplitude modulated at 80 Hz and tuned continuously with ≈10 MHz resolution.53 The resulting IR spectrum, reflected in this vibrational excitation induced cross section modulation, is normalized to the power of the OPO idler wave, as measured directly next to the window into the vacuum system. Wedged optical components and dry-N2 purging are used to mitigate power modulations along the optical path leading from the OPO to the Stark/multipass cell. Stark spectra of molecules in He droplets reveal inertial components of the permanent electric dipole moment that have been shown, in most cases,52 to be equivalent to their gas-phase values, within the error of the measurement.

For comparison with experimental work, the singlet trans,trans-, trans,cis-, and cis,cis- HO C ̈ OH isomers were investigated with electronic structure theory, using the CFOUR quantum chemistry package.54 Equilibrium geometries were obtained at two separate levels of theory: frozen core, coupled cluster singles and doubles with perturbative triples corrections (CCSD(T))55/ANO1,56 and all-electron CCSD(T)/cc-pCVQZ.57 Ground state equilibrium rotational constants were determined at the CCSD(T)/cc-pCVQZ level of theory. Fundamental frequencies, corresponding infrared intensities as well as state-specific vibrationally averaged dipole moments and inertial components of the vibrational transition dipole moments were determined using second-order vibrational perturbation theory (VPT2)58 at the frozen-core CCSD(T)/ANO1 level using the GUINEA module of CFOUR.59 The optimized geometry of the lowest triplet trans,trans-isomer was also determined, revealing a zero-point corrected singlet-triplet energy gap of 61.7 kcal/mol at the frozen core CCSD(T)/ANO1 level of theory. Results from these calculations are presented in Table I.

TABLE I.

Experimental and computed molecular properties of dihydroxycarbene rotamers.a

trans,trans-(b2) trans,trans-(a1) trans,cis-(a′)
Helium Theoryb Helium Theoryb Helium Theoryb Gasc
| μ a | (D)  …  1.63(3)  1.674  … 
| μ b | (D)  …  0.661  0.68(6)  0.661  1.50(5)  1.705  … 
| μ a | (D)  …  1.66(3)  1.709  … 
| μ b | (D)  …  0.686  0.68(6)  0.687  1.50(5)  1.678  … 
A″  1.45  2.86d  1.45  2.86d  1.55  2.4945d  2.4853 
A′  1.6  …  1.6  …  1.55  …  … 
B ̄   0.276  0.40d  0.284  0.40d  0.306  0.3878d  0.3841 
B ̄   0.273  …  0.261  …  0.301  …  … 
(BC)″  0.12  0.05d  0.12  0.05d  0.15  0.0599d  0.0597 
(BC)′  0.12  …  0.12  …  0.15  …  … 
ν0e  3655.438  3660.3 (192.4)  3658.305  3664.1 (17.7)  3649.915  3656.1 (106.8)  … 
μ ̇ a / μ ̇ b f  a-type  a-type  b-type  b-type  2.63(6)  2.50  … 
trans,trans-(b2) trans,trans-(a1) trans,cis-(a′)
Helium Theoryb Helium Theoryb Helium Theoryb Gasc
| μ a | (D)  …  1.63(3)  1.674  … 
| μ b | (D)  …  0.661  0.68(6)  0.661  1.50(5)  1.705  … 
| μ a | (D)  …  1.66(3)  1.709  … 
| μ b | (D)  …  0.686  0.68(6)  0.687  1.50(5)  1.678  … 
A″  1.45  2.86d  1.45  2.86d  1.55  2.4945d  2.4853 
A′  1.6  …  1.6  …  1.55  …  … 
B ̄   0.276  0.40d  0.284  0.40d  0.306  0.3878d  0.3841 
B ̄   0.273  …  0.261  …  0.301  …  … 
(BC)″  0.12  0.05d  0.12  0.05d  0.15  0.0599d  0.0597 
(BC)′  0.12  …  0.12  …  0.15  …  … 
ν0e  3655.438  3660.3 (192.4)  3658.305  3664.1 (17.7)  3649.915  3656.1 (106.8)  … 
μ ̇ a / μ ̇ b f  a-type  a-type  b-type  b-type  2.63(6)  2.50  … 
a

Units in cm−1 unless otherwise specified.

b

Equilibrium ab initio rotational constants at the CCSD(T)/cc-pCVQZ level of theory. VPT2-corrected ab initio dipole moments at the CCSD(T)/ANO1 level of theory.

c

Reference 40.

d

The discrepancy between experimental He droplet and computed rotational constants is due to the well-known effect by which the He solvent contributes to the rotational moment of inertia of the molecule.46–48 

e

Uncertainties in He droplet band origins are ±0.001 cm−1. Theoretical band origins, intensities (in parentheses) and transition dipole moment components are obtained via VPT2 at the CCSD(T)/ANO1 level of theory. The computed band origin for the lower frequency, unobserved a′ band of the trans,cis-isomer is 3385.7 (6.2). Theoretical band origins and intensities for the cis,cis-isomer are (b2) 3258.7 (42.5) and (a1) 3308.6 (6.0). For comparison to the He droplet spectra, Argon matrix band origins of the trans,trans- (b2), trans,trans- (a1), and trans,cis- (a′) vibrations are 3625.1, 3633.2/3628.6, and 3618.3 cm−1, respectively.36 

f

Ratio of transition dipole moment components along the a and b inertial axes.

Electron impact ionization mass spectra (MS) of the neat and doped He droplet beam under various conditions are shown in Fig. 2. Figure 2(a) shows the MS of the neat droplet beam, and the peaks separated by 4 u are assigned to a distribution of He cluster cations produced via the following mechanism (Eq. (1)), where m = 2, 3, …, n:

( He ) n + e He + ( He ) n 1 + 2 e ( He ) 2 + ( He ) n 2 + 2 e ( He ) m + + ( n m ) ( He ) + 2 e .
(1)
FIG. 2.

The mass spectrum of the neat helium droplet beam is shown in frame (a). Frames (b) through (d) are mass spectra of the droplet beam after having passed through the heated oxalic acid source operated under various pyrolysis source conditions (filament current in A). Pyrolytic decomposition of oxalic acid is indicated by the rapid increase in m/z = 44 u (assigned to OCO+) with increasing filament current. Spectra of oxalic acid are obtained under condition (b), whereas spectra of dihydroxycarbene are obtained under condition (d).

FIG. 2.

The mass spectrum of the neat helium droplet beam is shown in frame (a). Frames (b) through (d) are mass spectra of the droplet beam after having passed through the heated oxalic acid source operated under various pyrolysis source conditions (filament current in A). Pyrolytic decomposition of oxalic acid is indicated by the rapid increase in m/z = 44 u (assigned to OCO+) with increasing filament current. Spectra of oxalic acid are obtained under condition (b), whereas spectra of dihydroxycarbene are obtained under condition (d).

Close modal

Upon heating the oxalic acid sample to 330 K, an intense peak at m/z = 45 u appears (Fig. 2(b)), which is consistent with the primary product (OCOH)+ observed upon electron-impact ionization of gas-phase oxalic acid.60 The ionization and fragmentation of He-solvated oxalic acid occur via the mechanism shown in Eq. (2). Because of the large mismatch between the ionization potential of He and oxalic acid (≈14 eV), the dopant is ionized via charge transfer and readily fragments, producing (OCOH)+,

( C 2 H 2 O 4 ) ( He ) n + e ( C 2 H 2 O 4 ) He + ( He ) n 1 + 2 e ( C 2 H 2 O 4 ) + ( He ) n + 2 e ( OCOH ) + + ( CHO 2 ) * + n ( He ) + 2 e .
(2)

The MS in Figs. 2(b)2(d) are obtained with increasing pyrolysis filament current, with the top MS recorded under conditions that lead to the near complete decomposition of the oxalic acid precursor. The signature of this decomposition is both the dramatic reduction of m/z = 45 u and the appearance of (OCO)+ and (HCO)+, the latter of which is the major ionization product of He-solvated HO C ̈ OH (vide infra).

The IR spectrum of oxalic acid in the OH stretch region was measured by monitoring the laser-induced depletion of ion signal in mass channel 45 u (OCOH)+, and the results are shown in the bottom frame of Fig. 3. Several bands are assigned to three separate conformers on the basis of comparisons to previous Ar matrix spectra of Fausto and co-workers.61 The most intense band at 3475.40 cm−1 corresponds to the lower-frequency OH-stretching mode of the doubly intramolecular hydrogen bonded cTc (cis-Trans-cis) conformer (C2h symmetry). The higher energy cTt and tTt conformers are also present, which give rise to three additional weak bands at 3520.66 cm−1, 3576.84 cm−1, and 3602.21 cm−1. The structures of the three lowest energy oxalic acid conformers, cTc, cTt, and tTt, are shown as insets in Fig. 3.

FIG. 3.

Survey spectra of the oxalic acid precursor (frame (a)) and the products of pyrolysis (frame (b)). Mass channels correspond to either 45 (OCOH+) or 29 (HCO+) for the precursor and pyrolysis spectra, respectively. Near complete decomposition of the oxalic acid precursor is achieved with a pyrolysis temperature near 1000 K, and a series of sharp bands are observed that can be assigned to either formic acid or dihydroxycarbene (top spectrum). The stick spectrum corresponds to computed band origins of trans,trans-(red), trans,cis-(blue), and cis,cis- HO C ̈ OH (green), and stick heights reflect the computed IR intensities. The arrows indicate the band origins reported for trans,trans-(red) and trans,cis- HO C ̈ OH (blue) isolated in an Argon matrix.

FIG. 3.

Survey spectra of the oxalic acid precursor (frame (a)) and the products of pyrolysis (frame (b)). Mass channels correspond to either 45 (OCOH+) or 29 (HCO+) for the precursor and pyrolysis spectra, respectively. Near complete decomposition of the oxalic acid precursor is achieved with a pyrolysis temperature near 1000 K, and a series of sharp bands are observed that can be assigned to either formic acid or dihydroxycarbene (top spectrum). The stick spectrum corresponds to computed band origins of trans,trans-(red), trans,cis-(blue), and cis,cis- HO C ̈ OH (green), and stick heights reflect the computed IR intensities. The arrows indicate the band origins reported for trans,trans-(red) and trans,cis- HO C ̈ OH (blue) isolated in an Argon matrix.

Close modal

In a previous report on the IR spectrum of hydroxymethylene, we found (HCO)+ to be a major ionization product of the H C ̈ OH doped He beam, and ion signal in this channel was substantially modulated by upstream laser-induced excitation of He-solvated H C ̈ OH . Our initial search for HO C ̈ OH under high-temperature pyrolysis conditions (≈1000 K) was therefore also carried out on mass channel 29 u. The top frame of Fig. 3 shows the result of this survey scan, which consists of three distinct regions that contain OH stretch bands. The broad feature near 3470 cm−1 is due to residual signal from the unpyrolyzed cTc oxalic acid conformer. A series of sharp peaks centered around 3570 cm−1 are assigned to a Fermi triad involving the OH stretch of formic acid, and these bands have been observed and analyzed previously.61 Another set of sharp peaks centered near 3655 cm−1 are due to neither the precursor nor formic acid, yet they are approximately 30 cm−1 to the blue of the OH stretch bands assigned to HO C ̈ OH trapped in an Ar matrix (arrows in Fig. 3). Computed anharmonic spectra are shown in the stick spectrum below the survey scan, where the red, blue, and green lines are OH stretch band origins for the trans,trans-, trans,cis-, and cis,cis- HO C ̈ OH rotamers, respectively. The computations for the trans,trans- (both OH stretch bands) and trans,cis-rotamers (highest frequency OH stretch band) are in good agreement with the set of sharp transitions observed near 3655 cm−1. However, there is no evidence for the cis,cis-rotamer in the survey spectrum, and there is no obvious spectral feature near the computed band origin of the much weaker, lower frequency OH stretch band of the trans,cis-rotamer.

A higher resolution scan of the region centered around 3655 cm−1 is shown in Fig. 4, revealing three rovibrational bands that can be assigned unambiguously to OH stretch bands of HO C ̈ OH rotamers. Line widths of individual features within each band are ≈0.07 cm−1, allowing for an analysis of rotational fine-structure. The red simulation below the experimental spectrum is based on a summation of three separate asymmetric top spectra, each with a 0.35 K rotational temperature and rotational constants summarized in Table I. The patterns associated with the bands centered near 3656 and 3660 cm−1 are consistent with a- and b-type selection rules, respectively. Moreover, to satisfactorily simulate the relative intensities within each band, a 1:3 nuclear spin weight ratio must be imposed for even:odd (Ka + Kc) rotational levels, consistent with a C2v symmetry species having equivalent hydrogen atoms. Given these spectral features, we assign these two bands to the a1 symmetric (b-type) and b2 antisymmetric (a-type) OH stretching bands of trans,trans- HO C ̈ OH . The band origins for these modes are 3658.305 and 3655.438 cm−1, respectively. The lowest frequency rovibrational band in Fig. 4 is an a,b-hybrid band, which has relative transition intensities consistent with a nuclear spin statistical weight ratio equal to 1:1. This hybrid band (a:b = 2.6), centered at 3649.915 cm−1, must therefore be associated with a Cs symmetry species having non-equivalent hydrogen atoms. Because of these characteristic signatures, we assign this band to the higher frequency OH stretch (a′) of the trans,cis- HO C ̈ OH rotamer. The structures and inertial axes of these two rotamers are shown as insets to Fig. 4. The assignments are further justified by the comparison of experimental band origins and relative transition moment projections to those computed with VPT2. Each of the three bands are 5–6 cm−1 to the red of the computed values, given the above assignments. In several previous studies of He-solvated molecules, it was found that the band origins of higher-frequency X-H stretching modes were shifted by less than 1 cm−1 from their gas-phase values.62–67 Therefore, the excellent agreement between theory and experiment observed here provides strong confidence in the above assignments. These assignments imply rather large Ar matrix shifts of the high-frequency OH stretch bands of dihydroxycarbene,36 which are ≈30 cm−1 to the red of the band origins of the He-solvated species.

FIG. 4.

Rovibrational spectrum of trans,trans- and trans,cis- HO C ̈ OH rotamers in the OH stretch region. A simulation (red) derived from an asymmetric top Hamiltonian is shown below the experimental (black) spectrum. Assignments are based on band-types and nuclear spin statistical weights. Pure b- and a-type bands are observed for the symmetric and antisymmetric OH stretching vibrations of the C2vtrans,trans rotamer, respectively. The a, b-hybrid band corresponds to the higher frequency OH stretch of the Cs symmetry trans,cis-rotamer. The spectrum was measured under relatively low laser power conditions, so as to minimize saturation effects.

FIG. 4.

Rovibrational spectrum of trans,trans- and trans,cis- HO C ̈ OH rotamers in the OH stretch region. A simulation (red) derived from an asymmetric top Hamiltonian is shown below the experimental (black) spectrum. Assignments are based on band-types and nuclear spin statistical weights. Pure b- and a-type bands are observed for the symmetric and antisymmetric OH stretching vibrations of the C2vtrans,trans rotamer, respectively. The a, b-hybrid band corresponds to the higher frequency OH stretch of the Cs symmetry trans,cis-rotamer. The spectrum was measured under relatively low laser power conditions, so as to minimize saturation effects.

Close modal

Stark spectra at several field strengths were recorded for the a, b-hybrid and pure b-type rovibrational bands. A selection of Stark spectra for the hybrid band is shown in Fig. 5, along with the zero-field spectrum for comparison. The electric field lifts the 2J + 1 rotational state M-degeneracy, which leads to shifts and splitting of individual rovibrational lines and the appearance of a peak near the band origin that gains intensity with increasing field strength. A satisfactory simulation of these effects requires ground state inertial dipole components μa = 1.63(3) and μb = 1.50(5) D, where the error bars are derived from both the uncertainty in the field strength and the sensitivity of spectral changes to small changes in dipole components. Moreover, to obtain the best agreement between experiment and simulation, a slight increase in μa is required upon vibrational excitation, the sign and magnitude of which is also found by the VPT2 calculations. Stark spectra of the pure b-type band are shown along with simulations in Fig. 6. Constraining the a component of the dipole moment to zero (C2v symmetry), the shifts and splitting of b-type lines and the rate at which the normally forbidden transition near the band origin grows are best simulated with μb = 0.68(6) D. Here, the uncertainty is somewhat higher because the spectral shifts are less sensitive to changes in electric field strength, in comparison to the trans,cis- Stark spectra. These experimental inertial dipole components are compared to the vibrationally averaged values from VPT2 in Table I. The rather good agreement between experiment and theory observed here provides additional support to the above assignments.

FIG. 5.

Zero-field (bottom) and Stark spectra of the highest frequency OH stretch a, b-hybrid band of trans,cis- HO C ̈ OH . The laser electric field is aligned parallel to the static, dc, Stark field, the strength of which is noted in each frame. Simulations are generated assuming a semi-rigid asymmetric top in a Stark field, and the constants that parameterize these simulations are given in Table I.

FIG. 5.

Zero-field (bottom) and Stark spectra of the highest frequency OH stretch a, b-hybrid band of trans,cis- HO C ̈ OH . The laser electric field is aligned parallel to the static, dc, Stark field, the strength of which is noted in each frame. Simulations are generated assuming a semi-rigid asymmetric top in a Stark field, and the constants that parameterize these simulations are given in Table I.

Close modal
FIG. 6.

Zero-field (bottom) and Stark spectra of the OH stretch b-type band of trans,trans- HO C ̈ OH . The laser electric field is aligned parallel to the static, dc, Stark field, the strength of which is noted in each frame. Simulations are generated assuming a semi-rigid asymmetric top in a Stark field, and the constants that parameterize these simulations are given in Table I.

FIG. 6.

Zero-field (bottom) and Stark spectra of the OH stretch b-type band of trans,trans- HO C ̈ OH . The laser electric field is aligned parallel to the static, dc, Stark field, the strength of which is noted in each frame. Simulations are generated assuming a semi-rigid asymmetric top in a Stark field, and the constants that parameterize these simulations are given in Table I.

Close modal

The abundance of individual carbene rotamers can be determined by normalizing the integrated band areas to computed intensities, giving a 2:1 trans,trans-to trans,cis-ratio. Again, there is no evidence for the cis,cis-rotamer in He droplets, and there was also no evidence for it in the Ar matrix.36 These observations are consistent with a unimolecular decomposition mechanism of oxalic acid,68–72 in which a concerted hydrogen migration and C—C bond cleavage produces dihydroxycarbene and CO2. Assuming this mechanism, it is expected that the thermal extrusion of CO2 from the cTc oxalic acid isomer produces trans,trans- HO C ̈ OH , whereas the cTt isomer decomposes to trans,cis- (see Fig. 1). Indeed, computations of Huang and co-workers predict barrier heights for these concerted processes to be 31.1 and 32.4 kcal/mol, respectively. The reverse barrier heights are both less than 10 kcal/mol, implying a rather low internal energy content for the carbene produced via this gas-phase unimolecular decomposition. Given the large torsional interconversion barriers (≈17 kcal/mol) and the rapid cooling provided by He atom evaporation, we expect the relative rotamer populations in He droplets to reflect the gas-phase populations, which are dictated by the abundance of gas-phase oxalic acid conformers at the pyrolysis source temperature. Using the energetics reported by Fausto and co-workers (B3LYP/6-31G**/ZPE corrected)61 and taking into account structural degeneracies, we find that a Boltzmann distribution over conformers at 1000 K leads to a cTc:cTt ratio of 1.7:1, which is certainly qualitatively consistent with the 2:1 trans,trans-to trans,cis- HO C ̈ OH ratio reported here. The absence of the cis,cis-rotamer is apparently due to the low internal energy content of gas-phase trans,cis- HO C ̈ OH , which cannot rotationally interconvert prior to being captured by a He droplet. These results strongly support the oxalic acid decomposition mechanism proposed by Lapidus et al.68–71 

It is more difficult to rationalize the extent to which formic acid is produced in these experiments. In the Ar matrix study, Schriener et al. reported a 1:5 ratio of dihydroxycarbene to formic acid upon pyrolysis of oxalic acid.36 Using the computed transition intensities to normalize the experimental spectra, we find an almost identical ratio here. Bands due to formic acid are absent with the pyrolysis source operated under Fig. 2(b) conditions, in which the pyrolysis zone and the sample region of the quartz tube are both at ≈330 K. We observe these bands to rise sharply as the pyrolysis zone is heated above ≈700 K, while at the same time, the oxalic acid sample region is kept close to 330 K. Therefore, we can rule out the possibility of formic acid being produced directly from the 330 K oxalic acid sample. In the hot pyrolysis zone, formic acid, and CO2 can, in principle, be produced via the gas-phase unimolecular decomposition of oxalic acid. However, computations by Huang and co-workers predict an approximately 70 kcal/mol barrier for the lowest energy unimolecular pathway,72 which makes gas-phase unimolecular dissociation of oxalic acid an unlikely source of formic acid at 1000 K and on the timescale of the experiment (≈15 μs transit time through pyrolysis zone). Indeed, the absence of the cis,cis- HO C ̈ OH rotamer implies, as discussed above, an oxalic acid average internal energy that is too low to overcome the barriers leading to formic acid. Huang and co-workers also discovered a bimolecular path, in which oxalic acid collides with trans,trans- HO C ̈ OH and catalyzes its interconversion to formic acid via a doubly intramolecular hydrogen bonded transition state.72 The barrier for this process is 13 kcal/mol above the carbene on the potential surface. Nevertheless, using rate constants from transition state theory,72 we estimate that the concentration of precursor molecules is simply too low in the pyrolysis zone, by several orders of magnitude, for this bimolecular path to account for the formic acid observed in the spectra. We note that the pyrolysis sources used in the He droplet and Ar matrix studies were rather similar implementations of a quartz tube furnace. Given the experimental evidence and available theory for the above mentioned gas-phase processes, it is quite possible that the formic acid derives instead from a surface catalyzed process, which facilitates the direct decomposition of oxalic acid to formic acid prior to its desorption from the quartz surface. Future work in this area may benefit from a systematic examination of alternative pyrolysis sources, such as SiC furnaces, to maximize the production of carbenes from the decomposition of α-keto carboxylic acid precursors.

Pyrolysis of gas-phase oxalic acid produces formic acid and two of three possible rotamers of singlet dihydroxycarbene. The products of this pyrolytic decomposition are captured by helium droplets and cooled to 0.35 K. In addition to bands previously observed for formic acid, IR spectra in the OH stretch region reveal three bands exhibiting rotational fine structure. These rotationally resolved bands are assigned to the C2vtrans,trans- and Cstrans,cis- HO C ̈ OH rotamers on the basis of relative transition intensities within each band (revealing the vibrational symmetry and nuclear spin statistical weight ratios) and the comparison of experimental band origins to VPT2 estimates. Stark spectroscopy reveals inertial components of the permanent electric dipole moment for each of the two rotamers. Experimental inertial components of the permanent electric dipole moment compare favorably to the vibrationally averaged predictions from VPT2 calculations. The observed 5:1 formic acid to dihydroxycarbene ratio cannot be rationalized on the basis of the energetics for gas-phase oxalic acid unimolecular decomposition or other proposed gas-phase bimolecular processes. These results seem instead to imply a surface catalyzed oxalic acid decomposition mechanism.

G.E.D. acknowledges support from the Office of Energy Research, Office of Basic Energy Sciences, Chemical Sciences, Geosciences and Biosciences Division of the U.S. Department of Energy (DOE) under Contract No. DE-FG02-12ER16298. J.F.S. and L.M. thank the Robert A. Welch Foundation of Houston, TX (Grant No. F-1283) and the National Science Foundation (Grant No. CHE-1361031) for support.

1.
H. G.
Raubenheimer
,
Dalton Trans.
43
,
16959
(
2014
).
2.
K. H.
Doetz
and
J.
Stendel
, Jr.
,
Chem. Rev.
109
,
3227
(
2009
).
3.
Y.
Wang
,
Y.
Xie
,
P.
Wei
,
R. B.
King
,
H. F.
Schaefer
,
P. von R.
Schleyer
, and
G. H.
Robinson
,
Science
321
,
1069
(
2008
).
4.
P.
de Frémont
,
N.
Marion
, and
S. P.
Nolan
,
Coord. Chem. Rev.
253
,
862
(
2009
).
5.
K.
Namitharan
,
T. S.
Zhu
,
J. J.
Cheng
,
P. C.
Zheng
,
X. Y.
Li
,
S.
Yang
,
B. A.
Song
, and
Y. R.
Chi
,
Nat. Commun.
5
,
7
(
2014
).
6.
C. T.
Check
,
K. P.
Jang
,
C. B.
Schwamb
,
A. S.
Wong
,
M. H. S.
Wang
, and
K. A.
Scheidt
,
Angew. Chem., Int. Ed.
54
,
4264
(
2015
).
7.
H.
Wanzlick
,
Angew. Chem., Int. Ed.
1
,
75
(
1962
).
8.
Kinetics and Spectroscopy of Carbenes and Biradicals
, edited by
M. S.
Platz
(
Plenum
,
New York
,
1990
).
9.
R. A.
Moss
,
M. S.
Platz
, and
M.
Jones
, Jr.
,
Reactive Intermediate Chemistry
(
Wiley
,
Hoboken, NJ
,
2004
).
10.
S. H.
Kable
,
S. A.
Reid
, and
T. J.
Sears
,
Int. Rev. Phys. Chem.
28
,
435
(
2009
).
11.
C.
Tao
,
C.
Ebben
,
H.-T.
Ko
, and
S. A.
Reid
,
Phys. Chem. Chem. Phys.
10
,
6090
(
2008
).
12.
C. A.
Taatjes
,
S. J.
Klippenstein
,
N.
Hansen
,
J. A.
Miller
,
T. A.
Cool
,
J.
Wang
,
M. E.
Law
, and
P. R.
Westmoreland
,
Phys. Chem. Chem. Phys.
7
,
806
(
2005
).
13.
P.
Maksyutenko
,
F. T.
Zhang
,
X. B.
Gu
, and
R. I.
Kaiser
,
Phys. Chem. Chem. Phys.
13
,
240
(
2011
).
14.
P.
Thaddeus
,
C. A.
Gottlieb
,
R.
Mollaaghababa
, and
J. M.
Vrtilek
,
J. Chem. Soc., Faraday Trans.
89
,
2125
(
1993
).
15.
M. C.
McCarthy
,
M. J.
Travers
,
A.
Kovacs
,
W.
Chen
,
S. E.
Novick
,
C. A.
Gottlieb
, and
P.
Thaddeus
,
Science
275
,
518
(
1997
).
16.
J.
Fulara
,
P.
Freivogel
,
D.
Forney
, and
J. P.
Maier
,
J. Chem. Phys.
103
,
8805
(
1995
).
17.
H.
Ding
,
T. W.
Schmidt
,
T.
Pino
,
A. E.
Boguslavskiy
,
F.
Guthe
, and
J. P.
Maier
,
J. Chem. Phys.
119
,
814
(
2003
).
18.
D.
Bourissou
,
O.
Guerret
,
F. P.
Gabbai
, and
G.
Bertrand
,
Chem. Rev.
100
,
39
(
2000
).
19.
G.
Herzberg
and
J. W. C.
Johns
,
Proc. R. Soc. London, Ser. A
295
,
107
(
1966
).
20.
G.
Herzberg
and
J. W. C.
Johns
,
J. Chem. Phys.
54
,
2276
(
1971
).
21.
P. F.
Zittel
,
G. B.
Ellison
,
S. V.
Oneil
,
E.
Herbst
,
W. C.
Lineberger
, and
W. P.
Reinhardt
,
J. Am. Chem. Soc.
98
,
3731
(
1976
).
22.
P. C.
Engelking
,
R. R.
Corderman
,
J. J.
Wendoloski
,
G. B.
Ellison
,
S. V.
Oneil
, and
W. C.
Lineberger
,
J. Chem. Phys.
74
,
5460
(
1981
).
23.
K. K.
Murray
,
D. G.
Leopold
,
T. M.
Miller
, and
W. C.
Lineberger
,
J. Chem. Phys.
89
,
5442
(
1988
).
24.
E. P.
Clifford
,
P. G.
Wenthold
,
W. C.
Lineberger
,
G. A.
Petersson
,
K. M.
Broadus
,
S. R.
Kass
,
S.
Kato
,
C. H.
DePuy
,
V. M.
Bierbaum
, and
G. B.
Ellison
,
J. Phys. Chem. A
102
,
7100
(
1998
).
25.
R. L.
Schwartz
,
G. E.
Davico
,
T. M.
Ramond
, and
W. C.
Lineberger
,
J. Phys. Chem. A
103
,
8213
(
1999
).
26.
W.
Sander
,
G.
Bucher
, and
S.
Wierlacher
,
Chem. Rev.
93
,
1583
(
1993
).
27.
D. E.
Milligan
and
G. C.
Pimentel
,
J. Chem. Phys.
29
,
1405
(
1958
).
28.
P. S.
Zuev
,
R. S.
Sheridan
,
T. V.
Albu
,
D. G.
Truhlar
,
D. A.
Hrovat
, and
W. T.
Borden
,
Science
299
,
867
(
2003
).
29.
P. S.
Zuev
and
R. S.
Sheridan
,
J. Am. Chem. Soc.
126
,
12220
(
2004
).
30.
R. A.
Moss
,
R. R.
Sauers
,
R. S.
Sheridan
,
J. Z.
Tian
, and
P. S.
Zuev
,
J. Am. Chem. Soc.
126
,
10196
(
2004
).
31.
A.
Nicolaides
,
T.
Matsushita
,
K.
Yonezawa
,
S.
Sawai
,
H.
Tomioka
,
L. L.
Stracener
,
J. A.
Hodges
, and
R. J.
McMahon
,
J. Am. Chem. Soc.
123
,
2870
(
2001
).
32.
N. P.
Bowling
,
R. J.
Halter
,
J. A.
Hodges
,
R. A.
Seburg
,
P. S.
Thomas
,
C. S.
Simmons
,
J. F.
Stanton
, and
R. J.
McMahon
,
J. Am. Chem. Soc.
128
,
3291
(
2006
).
33.
R. A.
Seburg
,
E. V.
Patterson
, and
R. J.
McMahon
,
J. Am. Chem. Soc.
131
,
9442
(
2009
).
34.
P. R.
Schreiner
,
H. P.
Reisenauer
,
F. C.
Pickard
,
A. C.
Simmonett
,
W. D.
Allen
,
E.
Mátyus
, and
A. G.
Császár
,
Nature
453
,
906
(
2008
).
35.
C. M.
Leavitt
,
C. P.
Moradi
,
J. F.
Stanton
, and
G. E.
Douberly
,
J. Chem. Phys.
140
,
171102
(
2014
).
36.
P. R.
Schreiner
and
H. P.
Reisenauer
,
Angew. Chem., Int. Ed.
47
,
7071
(
2008
).
37.
P. C.
Burgers
,
G. A.
McGibbon
, and
J. K.
Terlouw
,
Chem. Phys. Lett.
224
,
539
(
1994
).
38.
F. A.
Wiedmann
,
J. N.
Cai
, and
C.
Wesdemiotis
,
Rapid Commun. Mass Spectrom.
8
,
804
(
1994
).
39.
D.
Feller
,
W. T.
Borden
, and
E. R.
Davidson
,
J. Chem. Phys.
71
,
4987
(
1979
).
40.
C. C.
Womack
,
K. N.
Crabtree
,
L.
McCaslin
,
O.
Martinez
, Jr.
,
R. W.
Field
,
J. F.
Stanton
, and
M. C.
McCarthy
,
Angew. Chem., Int. Ed.
53
,
4089
(
2014
).
41.
O.
Welz
,
J. D.
Savee
,
D. L.
Osborn
,
S. S.
Vasu
,
C. J.
Percival
,
D. E.
Shallcross
, and
C. A.
Taatjes
,
Science
335
,
204
(
2012
).
42.
J. M.
Beames
,
F.
Liu
,
L.
Lu
, and
M. I.
Lester
,
J. Am. Chem. Soc.
134
,
20045
(
2012
).
43.
J. H.
Lehman
,
H. W.
Li
,
J. M.
Beames
, and
M. I.
Lester
,
J. Chem. Phys.
139
,
141103
(
2013
).
44.
K.
Samanta
,
J. M.
Beames
,
M. I.
Lester
, and
J. E.
Subotnik
,
J. Chem. Phys.
141
,
134303
(
2014
).
45.
C.
Callegari
,
K. K.
Lehmann
,
R.
Schmied
, and
G.
Scoles
,
J. Chem. Phys.
115
,
10090
(
2001
).
46.
J. P.
Toennies
and
A. F.
Vilesov
,
Angew. Chem., Int. Ed.
43
,
2622
(
2004
).
47.
M. Y.
Choi
,
G. E.
Douberly
,
T. M.
Falconer
,
W. K.
Lewis
,
C. M.
Lindsay
,
J. M.
Merritt
,
P. L.
Stiles
, and
R. E.
Miller
,
Int. Rev. Phys. Chem.
25
,
15
(
2006
).
48.
F.
Stienkemeier
and
K. K.
Lehmann
,
J. Phys. B: At., Mol. Opt. Phys.
39
,
R127
(
2006
).
49.
M.
Lewerenz
,
B.
Schilling
, and
J. P.
Toennies
,
Chem. Phys. Lett.
206
,
381
(
1993
).
50.
E.
Knuth
,
B.
Schilling
, and
J. P.
Toennies
, in
Proceedings of the 19th International Symposium on Rarefied Gas Dynamics
(
Oxford University Press
,
London
,
1995
).
51.
M.
Hartmann
,
R. E.
Miller
,
J. P.
Toennies
, and
A.
Vilesov
,
Phys. Rev. Lett.
75
,
1566
(
1995
).
52.
P. L.
Stiles
,
K.
Nauta
, and
R. E.
Miller
,
Phys. Rev. Lett.
90
,
135301
(
2003
).
53.
A. M.
Morrison
,
T.
Liang
, and
G. E.
Douberly
,
Rev. Sci. Instrum.
84
,
013102
(
2013
).
54.
CFOUR, Coupled-Cluster Techniques for Computational Chemistry, a quantum chemical program package by Stanton, J. F., Gauss, J., Harding, M. E., and Szalay, P. G., with contributions from Auer, A. A. et al. For detailed information, see www.cfour.de.
55.
K.
Raghavachari
,
G. W.
Trucks
,
J. A.
Pople
, and
M.
Head-Gordon
,
Chem. Phys. Lett.
157
,
479
(
1989
).
56.
J.
Almlöf
and
P. R.
Taylor
,
J. Chem. Phys.
86
,
4070
(
1987
).
57.
D. E.
Woon
and
T. H.
Dunning
,
J. Chem. Phys.
103
,
4572
(
1995
).
58.
I. M.
Mills
, in
Molecular Spectroscopy: Modern Research
, edited by
K. N.
Rao
and
C. W.
Mathews
(
Academic Press
,
New York
,
1972
), p.
115
.
59.
D. A.
Matthews
and
J. F.
Stanton
,
Mol. Phys.
107
,
213
(
2009
).
60.
“Mass Spectra” by NIST Mass Spec Data Center, S. E. Stein, director, in NIST ChemistryWebBook, NIST Standard Reference Database Number 69, edited by P. J. Linstrom and W. G. Mallard (National Institute of Standards and Technology, Gaithersburg MD, 20899), http://webbook.nist.gov (retrieved January 9, 2015).
61.
E. M. S.
Maçôas
,
R.
Fausto
,
M.
Pettersson
,
L.
Khriachtchev
, and
M.
Räsänen
,
J. Phys. Chem. A
104
,
6956
(
2000
).
62.
C.
Callegari
,
A.
Conjusteau
,
I.
Reinhard
,
K. K.
Lehmann
, and
G.
Scoles
,
J. Chem. Phys.
113
,
10535
(
2000
).
63.
P. L.
Raston
,
J.
Agarwal
,
J. M.
Turney
,
H. F.
Schaefer
, and
G. E.
Douberly
,
J. Chem. Phys.
138
,
194303
(
2013
).
64.
C. M.
Leavitt
,
C. P.
Moradi
,
B. W.
Acrey
, and
G. E.
Douberly
,
J. Chem. Phys.
139
,
234301
(
2013
).
65.
P. L.
Raston
and
G. E.
Douberly
,
J. Mol. Spectrosc.
292
,
15
(
2013
).
66.
L. F.
Gomez
,
R.
Sliter
,
D.
Skvortsov
,
H.
Hoshina
,
G. E.
Douberly
, and
A. F.
Vilesov
,
J. Phys. Chem. A
117
,
13648
(
2013
).
67.
A. M.
Morrison
,
P. L.
Raston
, and
G. E.
Douberly
,
J. Phys. Chem. A
117
,
11640
(
2012
).
68.
G.
Lapidus
,
D.
Barton
, and
P. E.
Yankwich
,
J. Phys. Chem.
70
,
407
(
1966
).
69.
G.
Lapidus
,
D.
Barton
, and
P. E.
Yankwich
,
J. Phys. Chem.
70
,
1575
(
1966
).
70.
G.
Lapidus
,
D.
Barton
, and
P. E.
Yankwich
,
J. Phys. Chem.
70
,
3135
(
1966
).
71.
G.
Lapidus
,
P. E.
Yankwich
, and
D.
Barton
,
J. Phys. Chem.
68
,
1863
(
1964
).
72.
J.
Higgins
,
X. F.
Zhou
,
R. F.
Liu
, and
T. T. S.
Huang
,
J. Phys. Chem. A
101
,
2702
(
1997
).