CD₃OD⁺ studied by zero-kinetic energy photoelectron spectroscopy and theoretical calculations.

Torsional energy levels of CH3OH+/CH3OD+/CD3OD+ studied by zero-kinetic energy photoelectron spectroscopy and theoretical calculations Zuyang Dai, Shuming Gao, Jia Wang, and Yuxiang Mo Citation: The Journal of Chemical Physics 141, 144306 (2014); doi: 10.1063/1.4896986 View online: http://dx.doi.org/10.1063/1.4896986 View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/141/14?ver=pdfcov Published by the AIP Publishing Articles you may be interested in The cyclopropene radical cation: Rovibrational level structure at low energies from high-resolution photoelectron spectra J. Chem. Phys. 141, 064317 (2014); 10.1063/1.4890744 Tunneling splittings in vibronic energy levels of CH 3 F + X ̃ 2 E studied by high resolution photoelectron spectroscopy and ab initio calculation J. Chem. Phys. 139, 064302 (2013); 10.1063/1.4817201 Spectroscopy and dynamics of methylamine. II. Rotational and vibrational structures of CH 3 NH 2 and CH 3 ND 2 in cationic D 0 states J. Chem. Phys. 118, 11040 (2003); 10.1063/1.1575735 Nonresonant two-photon mass analyzed threshold ionization and zero kinetic energy photoelectron investigation of the X̃ 2 B 1 ground state of CH 2 CO + and CD 2 CO + J. Chem. Phys. 117, 6546 (2002); 10.1063/1.1506157 Four-dimensional model calculation of torsional levels of cyclic water tetramer J. Chem. Phys. 109, 5404 (1998); 10.1063/1.477159

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 130.113.111.210 On: Fri, 19 Dec 2014 14:11:17

THE JOURNAL OF CHEMICAL PHYSICS 141, 144306 (2014)

Torsional energy levels of CH3 OH+ /CH3 OD+ /CD3 OD+ studied by zero-kinetic energy photoelectron spectroscopy and theoretical calculations Zuyang Dai, Shuming Gao, Jia Wang, and Yuxiang Moa) Department of Physics and State Key Laboratory of Low-Dimensional Quantum Physics, Tsinghua University, Beijing 100084, China

(Received 29 July 2014; accepted 22 September 2014; published online 9 October 2014) The torsional energy levels of CH3 OH+ , CH3 OD+ , and CD3 OD+ have been determined for the first time using one-photon zero kinetic energy photoelectron spectroscopy. The adiabatic ionization energies for CH3 OH, CH3 OD, and CD3 OD are determined as 10.8396, 10.8455, and 10.8732 eV with uncertainties of 0.0005 eV, respectively. Theoretical calculations have also been performed to obtain the torsional energy levels for the three isotopologues using a one-dimensional model with approximate zero-point energy corrections of the torsional potential energy curves. The calculated values are in good agreement with the experimental data. The barrier height of the torsional potential energy without zero-point energy correction was calculated as 157 cm−1 , which is about half of that of the neutral (340 cm−1 ). The calculations showed that the cation has eclipsed conformation at the energy minimum and staggered one at the saddle point, which is the opposite of what is observed in the neutral molecule. The fundamental C–O stretch vibrational energy level for CD3 OD+ has also been determined. The energy levels for the combinational excitation of the torsional vibration and the fundamental C–O stretch vibration indicate a strong torsion-vibration coupling. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4896986] I. INTRODUCTION

The internal rotation in methanol between CH3 and OH groups, which is also called torsional motion, provides a benchmark example of large amplitude vibrations and their coupling with other vibrational modes in molecules.1–15 Due to its fundamental importance and applications in astronomy, there are a great number of experimental and theoretical works on methanol. For example, microwave, infrared and laser transition lines in the range of 0–1258 cm−1 of CH3 OH have been compiled.1 The experimentally determined energy levels of CH3 OH have been assigned using sophisticated model Hamiltonians considering torsional, vibrational, and rotational couplings.4–7 Theoretical calculations of energy levels based on full dimensional ab initio potential energy surfaces of CH3 OH have also been reported.8, 9, 14 In comparison with the neutral molecules, only a few experimental works on the torsional motion of the cation CH3 OH+ have been reported. In 1977, Karlsson et al. recorded the photoelectron spectra of CH3 OH.16 Because of the low resolution (80 cm−1 ) and the broadening of the spectra by the torsional vibrations only the C–O stretch vibration was tentatively determined.16 Macneil and Dixon17 reported photoelectron spectra of CH3 OH, CH3 OD, and CD3 OD, which had similar resolution to that reported by Karlsson et al.16 However, they tried a deconvolution method to improve the resolution of the spectra and simulated the vibrational bands based on the torsional parameters of neutrals. From the simulation, they concluded that the barrier heights of the torsion are in the range 150–200 cm−1 , and the geometries of the a) Author to whom correspondence should be addressed. Electronic mail:

[email protected] 0021-9606/2014/141(14)/144306/8/$30.00

barriers have undergone a conformational change relative to those of the neutrals.17 However, the torsional energy levels have not been obtained due to the limited resolution of the spectra. Although there are several theoretical works on the interconversion of the methanol cation,18–21 it is surprising that only little work has been reported on the calculation of its torsional energy levels, which may be due to a lack of high resolution experimental data. To understand the properties of the cations CH3 OH+ /CH3 OD+ /CD3 OD+ , several important questions need to be answered: (a) the torsional energy levels and their splittings, (b) the barrier heights of the torsional potential, and (c) the minimum and saddle point geometric structures. In this work we try to answer the above questions by a combined experimental and theoretical study. The first several torsional energy levels for CH3 OH+ /CH3 OD+ /CD3 OD+ and the fundamental C–O stretch vibration for CD3 OD+ have been determined using VUV laser and zero-kinetic energy (ZEKE) photoelectron spectroscopy. In addition to these, we also report the adiabatic ionization energies (IEs) for CH3 OH/CH3 OD/CD3 OD with much improved accuracy compared to previous data. The torsional vibrational energy levels have also been calculated using a one-dimensional (1D) model. The potential energy curve was calculated at the RCCSD(T)-F12A/aug-cc-pVTZ level.22, 23 The calculated energy levels are in good agreement with experimental data.

II. EXPERIMENTAL METHOD

The experimental apparatus has been described previously.24–26 Therefore, only a brief description is

141, 144306-1

© 2014 AIP Publishing LLC


144306-2

Dai et al.

provided here. The coherent VUV radiation was generated by employing the resonance enhanced four-wave difference mixing in a pulsed Kr jet. Two dye lasers pumped by one Nd-YAG laser (20 Hz repetition rate) were used to prepare the fundamental frequency laser beams. The frequency (ω1 ) of one of the dye lasers with tripling unit was fixed so that the 2ω1 matches the transition 4p5 (2 P1/2 )5p[1/2]0 ← 4p6 (98 855.07 cm−1 ) of Kr. The wavelength of the other dye laser was tuned from 640 to 700 nm. A H2 Raman cell was used to red shift this dye laser wavelength to 870–990 nm. The two fundamental beams were merged by a dichroic mirror and focused by an achromatic lens (250 mm) into the pulsed Kr jet. The pulse voltages to ionize the molecules were varied from 1 to 6 V/cm with duration of 1 μs and a delay of 3 μs relative to the VUV laser pulses. It was found that there are no appreciable shifts of the vibrational bands in the ZEKE photoelectron spectra recorded using different electric field strengths. Therefore, the reported positions of the vibrational bands have not been corrected for the electric field induced shifts. The reported spectra were also not normalized to the VUV intensities. The wavelengths of the dye lasers were calibrated using He/Ne and He/Ar opto-galvanic lamps. The samples entered into the chamber by bubbling the pure liquid with high purity (99.995%) argon carrier gas (760 Torr). The isotopologues have deuterium purity better than 99%.

III. THEORETICAL METHOD A. Torsional potential energy curves

For CH3 OH+ , the torsional motion is much slower than other vibrational motions, which indicates that the torsional energy levels can be approximately calculated by keeping other vibrational coordinates fixed.15 We assumed that the torsional motion is along the steepest descent path (reaction path, RP) on the adiabatic potential energy surface (PES).27 This assumption has been used by many other authors in the discussion of the torsional energy levels of neutral methanol.8, 15, 28 There are other 1D treatments not using the RP path to determine the energy levels of internal rotation, see recent review.29 The RP was calculated from the torsional saddle point (torsional angle τ = 0) to the minimum energy point (torsional angle τ = π /3). The RP for the entire torsional motion was extended using symmetry requirements. In this paper, the calculation of the electronic energies was done using the MOLPRO software package at the RCCSD(T)-F12A/augcc-pVTZ level.22, 23 The previous reports showed that the zero-point energy (ZPE) corrections to the RP are important in torsional energy level calculations.8 In our work the zero-point energies (ZPEs) excluding the torsional motion were calculated at the energy minimum point and the saddle point with harmonic approximation. The barrier heights considering the ZPEs are thus calculated for the three isotopologues. To save the computational time, the corrections of the ZPEs at other points along the RPs are approximated by multiplying scaling fac-

J. Chem. Phys. 141, 144306 (2014)

FIG. 1. The Newman projection for the definition of angle τ describing the torsional motion between CH3 and OH in CH3 OH+ . The C–O bond is taken as the Z axis and the O atom is outside the plane. The dihedral angles φ 2 , φ 3 , and φ H are relative to the plane H1 -C-O. As defined in Eq. (1), τ = φH −

φ2 +φ3 3 .

The reader may also refer to Fig. 1 in Ref. 9 for the definition.

tors that equal to the ratios between the barrier heights with and without ZPE corrections. It is important to define a torsional angle that keeps the C3 symmetry of the torsional potential.9 The torsional motion can be described using the internal angle τ , φ2 + φ3 . (1) 3 Figure 1 shows the definition of the angle τ . This definition has been used by Bowman and his co-workers, and similar ones have also been used by other authors.8, 9, 14 To study the torsional motion in molecules, we have to separate it from the molecular rotation. Therefore, the origin of the coordinate is in coincidence with the center of mass of the molecule, and the axes are the principal axes of the rotation, as done in the usually treatment of molecular rotation. It is noted that there are overall rotations with respect to the C–O bond (Z axis, see Fig. 1) between neighboring geometries along the RP. The average angle of overall rotation between neighboring geometries can be calculated using the following equation: (n) (n) Iα α α α = O, C, H1 , H2 , H3 , H, (2) χ (n) = (n) Iα τ = φH −

α

IA(n)

where is the rotational inertia of atom A with respect to the Z axis at the nth geometry and (n) α is the rotational angle of atom A with respect to the Z axis between the nth and the (n − 1)th geometries on the RP. χ (n) is hence the average angle of overall rotation from the nth to the (n − 1)th geometries. For each geometric structure we rotate the nth geometry back ward with an angle ni=0 χ (i) or the sum of previous overall rotation angles. Therefore, the overall rotation with respect to the Z axis is approximately eliminated. For each isotopologue independent calculations are performed to remove the overall rotation.


Dai et al.

144306-3

J. Chem. Phys. 141, 144306 (2014) TABLE I. Geometric structures of CH3 OH+ at the saddle and minimum energy.a

B. Kinetic energy operators

The classical kinetic energy along the RP is expressed by 1 dxα 2 dyα 2 dzα 2 m + + T = , 2 α dt dt dt α α = H, O, C, H1 , H2 , H3 ,

(3)

where mα is the mass of atom α, and xα , yα , zα are the trajectories of atom α in the Cartesian coordinate. Because the trajectories of atoms are functions of the torsional angle τ , we have 2 1 dτ T = f (τ ) , (4a) 2 dt f (τ ) =

α

mα

dxα dτ

2

+

dyα dτ

2

+

dzα dτ

2 .

(4b)

In our numerical calculations the Cartesian trajectories of atoms along the RP were interpolated using cubic spline method and the derivatives were calculated with respect to the interpolated curves. Following Podolsky’s trick,30 the quantum kinetic energy operator is T =

1 1 1 1 p p , 1/4 1/2 2 f (τ ) f (τ ) f (τ )1/4

p = −i

∂ . ∂τ

(5)

To compare the torsional motion with the rotational motion we define the path dependent rotational constant as B (τ ) =

¯2 . 2f (τ )

(6)

IV. RESULTS AND DISCUSSION A. Potential energy curve and torsional energy levels

The ground electronic state of methanol cation CH3 OH+ is formed by removing an electron in the 2a orbital of the neutral CH3 OH, which is mainly located at the O atom. Therefore, the CH3 OH+ (X) has A symmetry at the energy minimum.14 The geometric parameters of CH3 OH+ at the energy minimum and at the saddle point for the torsional motion are listed in Table I. Panel (a) of Figure 2 shows the potential energy curve (PEC) of CH3 OH+ /CD3 OD+ /CH3 OD+ along the RP. For a comparison, the PEC along the RP for neutral CH3 OH/CD3 OD/CH3 OD is also shown in panel (b) of Fig. 2. The RP shown in Fig. 2 was calculated using the MOLPRO software package, which does not depend on the masses of the molecules.23 It is noted that MOLPRO uses a second order method to determine the steepest descent line.32 The RP is characterized by three equivalent wells and barriers and has the C3 symmetry. Three important parameters describing the torsional motion are the barrier heights, the locations of energy minima, and the saddle points in the RP. The torsional barrier heights of the cations shown in Fig. 2 are 157 cm−1 , while those of the neutrals are around 340 cm−1 .8–15

Geometryb

Saddle

Minimum

Ref. 31c

R(C–O) R(C–Hb ) R(C–Ha ) R(O–H) O-C-H b O-C-H a C-O-H H -C-O-H a b

1.377 1.116 1.082 0.986 108.1 109.6 112.5 125.9

1.367 1.119 1.083 0.985 105.3 115.5 113.7 128.8

1.370 1.123 1.090 0.983 105.6 115.3 113.1 128.3

a b c

The bond length is in Å and the bond angle is in degree. There are two equivalent Hb atoms outside the Cs symmetric plane. Minimum energy structure calculated at the CCSD(T)/6-311G(df,p) level.

The barrier heights taken account of the ZPE corrections for CH3 OH+ , CH3 OD+ , and CD3 OD+ are 226, 212, and 182 cm−1 , respectively. Therefore, scaling factors of 1.439, 1.350, and 1.159 are used to multiply the PEC shown in panel (a) of Fig. 2 to obtain the approximate PECs taken account of the ZPE corrections for the above three isotopologues, respectively. It is noted that the barrier height taken account of ZPE correction for the neutral CH3 OH is 364 cm−1 . The potential energy is at the maximum when the OH bond bisects two neighboring C–H bonds and is at the minimum when the OH bond is in the same plane with one of the C–H bonds. In other words, the minimum energy geometry of CH3 OH+ has eclipsed conformation, while the saddle point geometry of CH3 OH+ has staggered conformation, which is opposite to what is observed in neutral CH3 OH, as shown in Fig. 2. This result has been suggested by Macneil and Dixon based on the analysis of the intensity distribution of the photoelectron spectra of CH3 OH.17 We provided a concrete conclusion based on the high level ab initio calculations. Figure 3 shows the B(τ ) of CH3 OH+ , CH3 OD+ , and CD3 OD+ as a function of torsional angles. The B(τ ) is

FIG. 2. The potential energy curves (PECs) for the torsion along the reaction path: (a) CH3 OH+ /CD3 OD+ /CH3 OD+ and (b) CH3 OH/ CH3 OD/CD3 OD. The definition for the torsional angle τ is shown in Fig. 1. It is seen that the geometric structures of the neutral at the energy minimum and saddle point correspond to those of the cation at the saddle point and energy minimum, respectively.


144306-4

Dai et al.

J. Chem. Phys. 141, 144306 (2014)

FIG. 3. The reaction path dependent rotational constant B(τ ) defined in Eq. (6) as a function of torsional angles for CH3 OH+ , CH3 OD+ , and CD3 OD+ , respectively.

invariant with respect to 2π /3 rotations. The values of B(τ ) for CH3 OH+ , CH3 OD+ , and CD3 OD+ are in decreasing order due to the deuterium effect. Having determined the PECs along the RP with approximate ZPE corrections and the formal rotational constants B(τ ), the 1D Schrödinger equations were solved to determine the torsional energy levels. Table II shows the calculation results for CH3 OH+ /CH3 OD+ /CD3 OD+ and the experimentally determined levels that are introduced in Sec. IV B. If the torsional barriers are very low, the OH and CH3 groups can rotate freely. We have the so-called free internal rotation. If the barrier height is higher than the spacing between the torsional energy levels, we have a hindered-rotation and the torsional energy levels are split by the tunneling effect. As shown in Table II, tunneling energy splittings occur in the methanol cation.

tailed description of the simulation method can be found in our previous papers on the ZEKE spectra of BCl3 + and other molecules.24–26 Therefore, we provide only a short description here. The rotational states of CH3 OH (CH3 OH+ ) can be labeled by |τ J M (|τ + N+ M+ ), where τ (τ + ), J (N+ ) and M (M+ ) represent the numbering of the energy levels, the nuclear rotational quantum number, and the quantum number for the projection of nuclear rotational angular momentum on the space-fixed axis, respectively. For the cation we neglect the spin-orbit and spin-rotation interactions. The rotational state of the cation can be thus described using Hund’s case (b). The wave functions |τ J M and |τ + N+ M+ can be expanded using the basis functions of a symmetric top molecule, cτ K |J K M , (7) |τ J M = K

|τ + N + M + =

(8)

Here K and K+ are the quantum numbers related to the projection of J(N+ ) on the principal axis of the neutral and cation, respectively. Having known the rotational constants, the expanding coefficients Cτ K (Cτ + K + ) can be obtained by solving the Hamitonian equations. We assume that the rotational line strength I(N+ τ + , J τ ) can be expressed by the equation, I (N + τ + , J τ ) = r(N)(2J + 1)(2N + + 1)

2

E

cτ + K + cτ K μ (K) exp − J K , ×

+ kTrot

(9)

K K

N = N + − J , B. Experimental results

cτ + K + |N + K + M + .

K+

K = K + − K ,

(10)

1. Semi-empirical rotational simulation to determine the band origins

where r(N) and μ(K) are the scaling factors for different rotational branches and sub-bands, respectively. They can be adjusted to give the best fit of the experimental spectra.

To determine the vibrational energy levels accurately, we have performed semi-empirical rotational simulations. A de-

2. Torsional energy levels

TABLE II.

Torsional energy levels of CH3 OH+ , CH3 OD+ , and CD3 OD+ .a CH3 OH+

CH3 OD+

CD3 OD+

Symmetry

Calc.b

Expt.c

Calc.b

Expt.c

Calc.b

Expt.c

0(a1 ) 0˜ (e) 1(e) 1˜ a2 2(a1 ) 2˜ (e)

0 10.3 124.7 189.1 248.1 363.3

0 13(4) 114(4) 192(6) 236(6)

0 5.3 99.8 135.5 195.3 267.7

0 7(4) 92(4) 137(6) 190(6)

0 3.8 81.4 107.5 159.7 214.7

0 6(4) 88(4) 106(6) 160(6) 215(6)

Energy units are in cm−1 . Theoretical results using one-dimensional model with approximate zero-point energy corrections, see text for details. c Determined from the simulation of the experimental spectra. The estimated uncertainties are the numbers in the parentheses. a

b

The panels (a) and (b) of Fig. 4 show the ZEKE spectrum of CH3 OH and the simulated spectrum, respectively. The ZEKE spectrum shows clearly two resolved peaks. The first peak actually consists of the tunneling pairs of the ground ˜ states. The second state, which we named as 0(a1 ) and 0(e) ˜ 2 ). peak contains the first excited tunneling pair 1(e) and 1(a In the simulation the calculated torsional energy levels were first used as the initial input and then adjusted to give the best fit of the experimental spectra. The torsional energy levels determined from the semi-empirical simulations are listed in Table II. The parameters used in the empirical simulation are listed in Table III. A full list of such parameters for the simulation of the ZEKE spectra of CH3 OH, CH3 OD, and CD3 OD can be found in the supplementary material.33 It is seen in Table II that the torsional energy levels determined from the simulation of the experimental spectra are in good agreement


144306-5

Dai et al.

J. Chem. Phys. 141, 144306 (2014)

FIG. 4. (a) ZEKE spectrum of CH3 OH in the adiabatic ionization energy region and (b) semi-empirical simulation to determine the first adiabatic ionization energy of CH3 OH and the torsional energy levels of CH3 OH+ . The torsional progression, CH3 OH+ (v) ← CH3 OH (v = 0), is shown by the vertical bars. In panel (b) the simulated components of vibrational transitions, CH3 OH+ (v) ← CH3 OH (v = 0(a),1(e)), are shown. N + K a K c ← J K a K c represents the rotational excitation from the neutral to the cation.

with the calculation results. For example, the energy splitting determined from the experimental spectra of CH3 OH+ is 13 (±4) cm−1 , while that of the calculation is 10.3 cm−1 . The experimentally determined fundamental torsional energy levels are 114(±4) and 192(±6) cm−1 , the corresponding calculations are 124.7 (e) and 189.1(a2 ) cm−1 . The panel (b) of Fig. 5 shows the ZEKE spectra of CH3 OH/CH3 OD/CD3 OD in the energy region from the vibrational ground states of the cations to the first fundamental torsional energy levels. The assignments of the spectra are depicted in the figure. The VUV light intensities as a function of photon energies are shown in panel (a) of Fig. 5. The determined torsional energy levels are listed in Table II. It is seen that the tunneling splittings decrease with increasing degree of deuterium, as expected. Each level splits into a pair of levels with a and e symmetry in accordance with the C3 symmetric group. Because of the lower potential barrier in comparison with the neutrals, the energy splittings of the cations are larger. For example, the splitting of the ground state of CH3 OH+ is 13 (±4) cm−1 , while that of the neutral is 9 cm−1 .

3. Adiabatic ionization energies

The determined IEs for CH3 OH, CH3 OD, and CD3 OD are 87 427, 87 475, and 87 698 cm−1 , respectively, with un-

certainty of 4 cm−1 , as listed in Table IV along with previously reported results. The IEs reported in this work are about one order of magnitude more accurate than the previously reported ones. It is seen in Fig. 5 and Table IV that the IEs of CH3 OH/CH3 OD/CD3 OD increase with the more substitutions of D atoms for the H atoms in CH3 OH. The differences of the IEs mainly arise from the difference of ZPEs between the cation and neutrals. The result indicates that the differences of ZPEs between the cations and neutral molecules increase in the order of CH3 OH/CH3 OD/CD3 OD. Similar observations have been found in the case of CH3 Cl and CH3 F.25, 26, 34

4. C–O stretch vibration for CD3 OD+

The C–O stretch vibrational bands for CH3 OH+ , CH3 OD+ , and CD3 OD+ were observed by Macneil and Dixon at 926, 928, and 763 cm−1 with uncertainties of 30 cm−1 , respectively.17 Because the VUV light intensities are very weak in this energy region, we have only measured the ZEKE spectrum of CD3 OD, which is shown in panel (a) of Fig. 6. Panel (b) of Fig. 6 shows the simulation curve. The parameters used in the simulation can be found in the supplementary material.33


144306-6

Dai et al.

TABLE III. Parameters used in the semi-empirical simulation of the ZEKE spectrum of CH3 OH to determine the torsional energy levels of the cation.

J. Chem. Phys. 141, 144306 (2014) TABLE IV. First adiabatic ionization energies (IE) of CH3 OH, CH3 OD, and CD3 OD.a

Rotational constant (cm−1 )a A/A+ 4.254/4.165 B/B+ 0.824/0.883 C/C+ 0.793/0.832 Rotational temperature, 10 K Instrumental resolution (FWHM), 3 cm−1 Assignment Ionization energy (cm−1 )b 0(a1 ) 87 427(4) 0˜ (e) 87 440(4) 1(e) 87 541(4) 87 619(6) 1˜ a2 87 663(6) 2(a1 ) Line strength for transition ˜ 0.3 0 ← 0, 0˜ ← 0, ˜ ˜ 0 ← 0, 0 ← 0, 0.2 Scaling factor for 0 and 0˜ levels r (0) = 2.5, r (±1) = 0.5, r (±2) = 1, r (3) = 1.5, r (−3) = 0.5 μ (0) = 2, μ (±1) = 1, μ (±2) = 1.5, μ (−3) = 0.5, μ (3) = 0.5, μ (−4) = 0.5, μ (4) = 3 ˜ 2, and 2˜ levels Scaling factor for 1, 1, r (0) = 1, r (±1) = 0.5, r (2) = 0.5, r (−2) = 1, r (3) = 1.5, r (−3) = 0.5 μ (0) = 2, μ (±1) = 1, μ (−2) = 2, μ (2) = 1.5, μ (−3) = 0.5, μ (3) = 0.5 μ (−4) = 0.5, μ (4) = 2.5 a The rotational constants of the cation are from the ab initio theoretical calculations and the rotational constants of CH3 OH are from the reported experimental data. A full list of parameters for the semi-empirical simulation of the ZEKE spectra of CH3 OD and CD3 OD can be found in the supplementary material.33 b Estimated uncertainties are in the parentheses.

IE IE IE

CH3 OH

CH3 OD

CD3 OD

10.8396(5) 10.846(2) 10.853(5)

10.8455(5) 10.861(2)

10.8732(5) 10.885(2)

This work Ref. 17 Ref. 16

a

Energy units are in eV. The numbers in the parentheses are the uncertainties in the last digits.

The determined IE for the fundamental vibrational band of the CO stretch is 88 408 ± 4 cm−1 . Therefore, the fundamental vibrational energy level of the C–O stretch is 710 ± 4 cm−1 , which is much smaller than the previously reported data 763 ± 30 cm−1 . The combinational bands of the C–O stretch and the torsion are shown in Table V. The first torsional energy level of the combination excitation, 1(e), is located at 135 ± 4 cm−1 , while that of the pure torsion, 1(e), is only 88 ± 4 cm−1 . This indicates that the interaction between the C–O stretch and the torsional vibrations is very strong and the effective torsional barrier for the C–O stretch vibration should be higher than that of the pure torsion, which is different from those of neutral.8, 14 This fact also indicates that 1D model may not be simply applied to CD3 OD+ when the vibrational energy levels involve combinational excitation of C–O stretch vibration and torsional motion.

FIG. 5. (a) VUV light intensities as a function of photon energies and (b) the ZEKE photoelectron spectra of CH3 OH, CH3 OD, and CD3 OD in the adiabatic ionization energy region. The assignments of the torsional energy levels are shown in the figure. The ZEKE spectrum of CH3 OH is also shown in panel (a) of Fig. 4.


Dai et al.

144306-7

J. Chem. Phys. 141, 144306 (2014)

FIG. 6. (a) ZEKE spectrum of CD3 OD of the fundamental C–O stretch vibration and (b) semi-empirical simulation to determine the torsional energy levels of CD3 OD+ . The torsional progression, CD3 OD+ (1v) ← CD3 OD (v = 0), are shown by the vertical bars. In panel (b) the simulated components of vibrational transitions, CD3 OD+ (1v) ← CD3 OD (v = 0(a),1(e)), are shown. N + K K ← J K K represents the rotational excitation from the neutral to the cation. a

c

V. SUMMARY

We have measured the ZEKE spectra of CH3 OH, CH3 OD, and CD3 OD. The torsional vibrational energy levels of CH3 OH+ , CH3 OD+ , and CD3 OD+ have been determined for the first time. They have also been calculated using a one-dimensional model with approximate zero-point energy corrections and are in good agreement with the experimental data. The calculations showed that the cations at the energy minima and at the saddle points have eclipsed and staggered conformations, respectively, while the opposite is true for the neutral. The combinational excitation energy levels of the torsion and the C–O stretch in CD3 OD+ indicate a strong coupling between the torsion and the C–O stretch vibrations.

a

c

The adiabatic ionization energies for CH3 OH, CH3 OD, and CD3 OD are determined as 10.8396, 10.8455, and 10.8732 eV with uncertainties of 0.0005 eV, respectively. ACKNOWLEDGMENTS

The authors are grateful to Professor Toshinori Suzuki of Kyoto University, Japan, for providing us high resolution photoelectron spectra of CH3 OH and CH3 OD using Helium lamp. This work is funded by Projects 2010CB922900 and 2013CB834604 supported by the NKBRSF of China, and Projects 21327902 and 10734040 supported by the National Science Foundation of China. 1 G.

TABLE V. Torsional energy levels of CD3 OD+ : pure torsion vs. combinational excitation with the fundamental C–O stretch vibration.a Level

IEb

0(a1 )

0˜ (e)

1(e)

1˜ a2

2(a1 )

2˜ (e)

(1v)c (0v)d

88 408 87 698

0 0

6(4) 6(4)

135(4) 88(4)

174(6) 106(6)

212(6) 160(6)

272(6) 215(6)

Determined from the ZEKE spectra. Energy units are in cm−1 . The estimated uncertainties of the experimental data are in the parentheses. b Ionization energies with no torsional excitation. c Combinational energy levels of the fundamental C–O stretch and the torsional vibration. The energy levels are relative to the IE with no torsional excitation. d Energy levels of pure torsional vibration that are also listed in Table II. a

Moruzzi, B. P. Winnerwisser, M. Winnerwisser, I. Mukhopadhyay, and F. Strumia, Microwave, Infrared, and Laser Transitions of Methanol: Atlas of Assigned Lines from 0 to 1258 cm−1 (CRC Press, Boca Raton, 1995). 2 S. P. Belov, G. Winnesisser, and E. Herbst, J. Mol. Spectrosc. 174, 253 (1995). 3 R. M. Lees, L.-H. Xu, J. W. C. Johns, B. P. Winnewisser, and M. Lock, J. Mol. Spectrosc. 243, 168 (2007). 4 Y.-B. Duan, R. Wang, and I. Mukhopadhyay, Chem. Phys. 280, 119 (2002). 5 Y.-B. Duan, S. Yuan, R. Wang, and I. Mukhopadhyay, Chem. Phys. 330, 9 (2006). 6 L. Wang, Y.-B. Duan, R. Wang, G. Duan, I. Mukhopadhyay, D. S. Perry, and K. Takagi, Chem. Phys. 292, 23 (2003). 7 Y.-B. Duan, I. Ozier, S. Tsunekawa, and K. Takagi, J. Mol. Spectrosc. 218, 95 (2003). 8 B. Fehrensen, D. Luckhaus, M. Quack, M. Willeke, and T. R. Rizzo, J. Chem. Phys. 119, 5534 (2003).


144306-8 9 J.

Dai et al.

M. Bowman, X. Huang, N. C. Handy, and S. Carter, J. Phys. Chem. A 111, 7317 (2007). 10 L.-H. Xu and J. T. Hougen, J. Mol. Spectrosc. 173, 540 (1995). 11 L.-H. Xu and J. T. Hougen, J. Mol. Spectrosc. 169, 396 (1995). 12 L.-H. Xu, R. M. Lees, P. Wang, L. R. Brown, I. Kleiner, and J. W. C. Johns, J. Mol. Spectosc. 228, 453 (2004). 13 L.-H. Xu, J. Fisher, R. M. Lees, H. Y. Shi, J. T. Hougen, J. C. Pearson, B. J. Drouin, C. A. Blake, and R. Braakman, J. Mol. Spectosc. 251, 305 (2008). 14 E. L. Sibert III and J. Castillo-Chará, J. Chem. Phys. 122, 194306 (2005). 15 P. Jansen, L.-H. Xu, I. Kleiner, W. Ubachs, and H. L. Bethlem, Phys. Rev. Lett. 106, 100801 (2011). 16 L. Karlsson, R. Jadrny, L. Mattsson, F. T. Chau, and K. Siegbahn, Phys. Scr. 16, 225 (1977). 17 K. A. G. Macneil and R. N. Dixon, J. Electron. Spectrosc. Relat. Phenom. 11, 315 (1977). 18 W. J. Bouma, R. H. Nobes, and L. Radom, J. Am. Chem. Soc. 104, 2929 (1982). 19 W. J. Bouma, J. K. MacLeod, and L. Radom, J. Am. Chem. Soc. 104, 2930 (1982). 20 J. W. Gauld, H. Audier, J. Fossey, and L. Radom, J. Am. Chem. Soc. 118, 6299 (1996). 21 D. Fridgen and J. M. Parnis, Int. J. Mass Spectrom. 190, 181 (1999). 22 T. B. Adler, G. Knizia, and H.-J. Werner, J. Chem. Phys. 127, 221106 (2007).

J. Chem. Phys. 141, 144306 (2014) 23 H.-J.

Werner, P. J. Knowles, G. Knizia, F. R. Manby, M. Schütz et al., version 2010.1, a package of ab initio programs, 2010, see http://www.molpro.net. 24 J. Yang, Y. Mo, K. C. Lau, Y. Song, X. M. Qian, and C. Y. Ng, Phys. Chem. Chem. Phys. 7, 1518 (2005). 25 S. Gao, Z. Dai, W. Sun, H. Li, J. Wang, and Y. Mo, J. Chem. Phys. 139, 064302 (2013). 26 Z. Shao, H. Li, S. Zhang, J. Li, Z. Dai, Y. Mo, Y. J. Bae, and M. S. Kim, J. Chem. Phys. 136, 064308 (2012). 27 K. Fukui, Acc. Chem. Res. 14, 363 (1981). 28 A. M. Halpern and E. D. Glendening, J. Chem. Phys. 119, 11186 (2003). 29 A. Bauder, “Fundamentals of rotational spectroscopy,” in Handbook of High-Resolution Spectroscopy, edited by M. Quack and F. Merkt (Wiley, Chichester, 2011). 30 B. Podolsky, Phys. Rev. 32, 812 (1928). 31 J. W. Gauld, M. N. Glukhovtsev, and L. Radom, Chem. Phys. Lett. 262, 187 (1996). 32 J. Sun and K. Ruedenberg, J. Chem. Phys. 99, 5269 (1993). 33 See supplementary material at http://dx.doi.org/10.1063/1.4896986 for a complete list of parameters for the semi-empirical simulation of the ZEKE spectra of CH3 OH, CH3 OD and CD3 OD. 34 X. Xu, Z. Shao, W. Sun, Y. Zhou, Z. Dai, and Y. Mo, J. Mol. Spectrosc. 274, 35 (2012). MOLPRO ,


Superhalogen properties of BS2(-) and BSO(-): photoelectron spectroscopy and theoretical calculations.

Zero kinetic energy photoelectron spectroscopy of triphenylene.

Oxidation half-reaction of aqueous nucleosides and nucleotides via photoelectron spectroscopy augmented by ab initio calculations.

Charge Transfer Dynamics at Dye-Sensitized ZnO and TiO2 Interfaces Studied by Ultrafast XUV Photoelectron Spectroscopy.

Insights into the Photoelectron Spectroscopy of Chlorofluoroethenes Studied by Density-Functional and Coupled-Cluster Theories.

Zn0.8Al0.2O Heterojunction Studied by X-ray Photoelectron Spectroscopy.

Valence and Ionic Lowest-Lying Electronic States of Isobutyl Formate Studied by High-Resolution Vacuum Ultraviolet Photoabsorption, Photoelectron Spectroscopy, and Ab Initio Calculations.

Valence and ionic lowest-lying electronic states of ethyl formate as studied by high-resolution vacuum ultraviolet photoabsorption, He(I) photoelectron spectroscopy, and ab initio calculations.

The LVV Auger line shape of sulfur on copper studied by Auger photoelectron coincidence spectroscopy.

Conformational properties of cis- and trans-N-Cyclopropylformamide studied by microwave spectroscopy and quantum chemical calculations.

Variable photon energy photoelectron spectroscopy of tris-cyclopentadienyl lanthanides.

Zero Kinetic Energy Photoelectron Spectroscopy of Benzo[h]quinoline.

Electronic states of tetrahydrofurfuryl alcohol (THFA) as studied by VUV spectroscopy and ab initio calculations.

Probing the structural evolution of ruthenium doped germanium clusters: Photoelectron spectroscopy and density functional theory calculations.

Interaction of TiO2(-) with water: photoelectron spectroscopy and density functional calculations.

Revealing chemical ordering in Pt-Co nanoparticles using electronic structure calculations and X-ray photoelectron spectroscopy.

Microsolvation of LiBO2 in water: anion photoelectron spectroscopy and ab initio calculations.

Ge overlayer and surface alloy structures on Pt(100) studied using alkali ion scattering spectroscopy, x-ray photoelectron spectroscopy and x-ray photoelectron diffraction.

On the dissolution of lithium sulfate in water: anion photoelectron spectroscopy and density functional theory calculations.

Photoelectron spectroscopy and ab initio calculations of small SinSm⁻ (n = 1,2; m = 1-4) clusters.

Low-lying electronic structure of EuH, EuOH, and EuO neutrals and anions determined by anion photoelectron spectroscopy and DFT calculations.

Photoelectron spectroscopy and theoretical studies of anion-π interactions: binding strength and anion specificity.

Theoretical and Experimental Photoelectron Spectroscopy Characterization of the Ground State of Thymine Cation.

Covalently bound tetracoordinated organoborons as superhalogens: a combined negative ion photoelectron spectroscopy and theoretical study.