Synchrotron-Based Far-Infrared Spectroscopic Investigation and abInitio Calculations of 3‑Oxetanone: Observation and Analysis of theν7 Band and the Coriolis Coupled ν16 and ν20 BandsZiqiu Chen and Jennifer van Wijngaarden*

Department of Chemistry, University of Manitoba, Winnipeg Manitoba, R3T 2N2 Canada

*S Supporting Information

ABSTRACT: Rotationally resolved vibrational spectra of the four-membered heterocycle 3-oxetanone (c-C3H4O2) have been inves-tigated in the 360−720 cm−1 region with a resolution of 0.000 959cm−1 using synchrotron radiation from the Canadian Light Source.The observed bands correspond to motions best described as COdeformation out-of-plane (ν20) at 399.6 cm

−1, CO deformation in-plane (ν16) at 448.2 cm−1, and the ring deformation (ν7) at 685.0cm−1. Infrared ground state combination differences along withpreviously reported pure rotational transitions were used to obtainthe ground state spectroscopic parameters. Band centers, rotational and centrifugal distortion constants for the ν7, ν16, and ν20vibrational excited states were accurately determined by fitting a total of 10 319 assigned rovibrational transitions in a globalanalysis. The two adjacent carbonyl deformation bands, ν16 and ν20, were found to be mutually perturbed through a first-order a-type Coriolis interaction which was accounted for in the multiband analysis. The band centers agree within 3% of the ab initioestimates using DFT theory.

1. INTRODUCTION3-Oxetanone (c-C3H4O2), along with other four-memberedheterocycles, is a fundamentally important species in syntheticand pharmaceutical chemistry. It undergoes numerous ring-opening and ring expansion reactions and these, together withthe ability to transform its carbonyl group via reduction andGrignard reactions, make it a versatile starting material inmultistep organic synthesis.1 Such activity is inherently linkedto the instability of highly strained four-membered ringmoieties and the structural properties of the ring can clearlyinfluence the rate of nucleophilic substitutions at different sitesthat open the ring. In drug discovery, 3-oxetanone derivativesare preferred precursors for adding oxetane (c-C3H6O)subunits into pharmaceuticals that serve to block metabolicallyexposed sites in the bioactive molecule without increasing itslipophilicity.2,3 In this regard, the ring structure of the precursorobviously plays a role in the effectiveness of the blocking agent.To better understand and exploit the chemical and metabolicbehavior of 3-oxetanone and related heterocycles, a connectionof these activities to the fundamental properties of the ring isneeded.In this regard, modest effort has been put into the study of

the structure and dynamics of 3-oxetanone to date. Semi-empirical4 and ab initio5,6 calculations were used to investigatethe structure and ring puckering potential.7 Experimentally,from analysis of the microwave spectrum by Gibson andHarris,8 it has been shown that 3-oxetanone is a planarasymmetric top of C2v symmetry with a permanent dipolemoment of 0.887 D. The planar structure was rationalized byMeinzer and Pringle9 who investigated the balance between

torsional strain and ring strain and their connection to ringplanarity in 3-oxetanone and related compounds. Carreira andLord10 reported the first vibrational study of the ring puckeringmode of this molecule at 140 cm−1, and a more comprehensiveinfrared and Raman study in the gas and liquid phases wassubsequently presented by Durig et al.11 in which all 21fundamental bands were assigned and analyzed. To date, thevibrational spectrum of 3-oxetanone has not been studied byhigh resolution techniques.Our recent far-infrared spectra of the structural isomer 2-

oxetanone (β-propiolactone)12,13 demonstrate the value of highresolution investigations of small ring molecules. For example,our analysis of the ν13 and ν20 bands served to correct theassignment of the CO deformation in-plane and out-of-planemodes (separated by only ∼20 cm−1), which were misassignedin the original low resolution study of this molecule. Our highresolution spectrum also permitted the analysis of the Coriolisinteraction between these two modes leading to a morecomplete, accurate set of spectroscopic constants for thesestates.In the current work, we report the far-infrared rovibrational

assignment and analysis of the ν7, ν16, and ν20 bands of 3-oxetanone using data collected at Canada’s national synchro-tron facility. To the best of our knowledge, this is the first highresolution vibrational study of this molecule. On the basis of abinitio calculations of the harmonic and anharmonic vibrational

Received: August 2, 2012Revised: August 28, 2012Published: August 28, 2012

Article

pubs.acs.org/JPCA

© 2012 American Chemical Society 9490 dx.doi.org/10.1021/jp3076673 | J. Phys. Chem. A 2012, 116, 9490−9496

frequencies, these modes correspond to ring deformation andthe CO in-plane and out-of-plane bending modes,respectively. To properly account for the Coriolis interactionbetween the ν16 and ν20 bands, we first established accurateground state spectroscopic constants using ground statecombination differences from the assigned rovibrationaltransitions along with the previously reported microwavefrequencies.8 Once the ground state parameters were well-determined, a global analysis of 10 319 separate infraredtransitions corresponding to these three vibrations wasperformed to extract the key spectroscopic parameters thatcharacterize the vibrationally excited states. The experimentalresults are complemented by ab initio estimates of the bandcenters using DFT theory. These show close agreement for theν7, ν16, and ν20 bands reported here but also identify severalinconsistencies in the original assignment of the fundamentalbands of 3-oxetanone.11

2. EXPERIMENTAL DETAILS

The experiment was conducted at the Canadian Light Source(CLS), using the synchrotron as the infrared light source and aBruker IFS125 HR spectrometer capable of recording infraredspectra with an unapodized spectral resolution of 0.000 959cm−1. Synchrotron radiation (SR) offers a light source withhigher brightness than conventional thermal infrared sources.As SR is highly collimated, it passes through the small apertureneeded for high resolution experiments with minimal loss. Arecent infrared study of pyrrole compared the high resolutionvibrational spectrum recorded using SR from the CLS with thatrecorded using a conventional thermal (globar) source on thesame spectrometer.14 The signal-to-noise ratio of the SRspectrum showed a factor of 8 improvement compared to thatof the globar spectrum due to this “synchrotron advantage”.Rotationally resolved vibrational spectra of the ν7, ν16, and

ν20 bands of 3-oxetanone were recorded separately in twospectral regions at room temperature using vapor pressure from

a liquid sample (97% purity, Synthonix). For the weak ν7 band,708 mTorr of vapor was introduced into a 2 m multipass cell,which was set to achieve a total absorption path length of 72 m.Interferograms were recorded in the 550−1130 cm−1 regionutilizing the instrument’s entire 9.4 m optical path difference torealize the full spectral resolution of 0.000 959 cm−1. Thecalculated Doppler line width of 3-oxetanone in this region issuch that the observed line widths (fwhm) were instrumentlimited. For this particular spectral range, the spectrometer wasequipped with a KBr beamsplitter and a GeCu detecter. Thefinal spectrum of the ν7 band was obtained by averaging 758separate interferograms and applying a Fourier transform.Background interferograms were collected at lower resolution(0.015 36 cm−1), averaged and the result was transformed withappropriate zero filling before calculation of the absorptionspectrum. The data for the ν16 and ν20 bands were collected inthe range 360−550 cm−1 with the spectrometer outfitted with a6 μm Mylar beamsplitter and a GeCu detector at a reducedpressure of 188 mTorr to prevent saturation of the moreintense ν16 band. The spectrum was calculated from 454interferograms following the procedure described above.The ν7 spectrum was calibrated using a sample of N2O and

comparing the observed line positions in the 615−725 cm−1

range with those from the HITRAN database.15 The lowerfrequency spectrum was corrected using residual water lines inthe 360−490 cm−1 range. In the end, we found the measuredfrequencies in both spectral regions to be accurate to within0.0002 cm−1. A complete listing of the assigned transitions isprovided as Supporting Information.

3. FREQUENCY CALCULATIONS

The harmonic vibrational frequencies of 3-oxetanone werecalculated using Gaussian 03W16 at the B3LYP level with a 6-311++G(2d,3p) basis set following geometry optimization.Anharmonic corrections were subsequently determined viasecond-order perturbation analysis in G03W as our previous

Table 1. Comparison of the Experimentally Observed Vibrational Frequencies (cm−1)11 with Computational Predictions (ThisWork) (B3LYP/6-311++G(2d,3p))

fundamental ν experiment calculated harmonic calculated harmonic (scaled 0.9679)c calculated anharmonic normal mode symmetry

1 2931a 3037 2936 2889 CH2 sym str A1

2 1847 1888 1827 1851 CO str A1

3 1457 1488 1440 1472 CH2 scissoring A1

4 1134 1311 1269 1279 CH2 wag A1

5 979 991 959 971 ring breathing A1

6 832 840 813 822 ring def A1

7 683 688 666 680 ring def A1

8 2962b 3081 2982 2927 CH2 antisym str A2

9 1111b 1120 1084 1094 CH2 twist A2

10 1102b 770 745 755 CH2 rock A2

11 2924 3026 2929 2839 CH2 sym str B1

12 1430 1461 1414 1425 CH2 scissoring B1

13 1250a 1272 1231 1242 CH2 wag B1

14 1076 1087 1052 1051 ring def B1

15 955a 991 958 958 ring def B1

16 448 449 434 444 CO def in-plane B1

17 2956 3080 2981 2925 CH2 antisym str B2

18 1284a 1158 1121 1131 CH2 twist B2

19 1049 1070 1036 1048 CH2 rock B2

20 401 410 397 411 CO def out-of-plane B2

21 140 149 135 158 ring puckering B2

aFrom liquid IR spectrum. bFrom liquid Raman spectrum. cReference 21.

The Journal of Physical Chemistry A Article

dx.doi.org/10.1021/jp3076673 | J. Phys. Chem. A 2012, 116, 9490−94969491

work on the structural isomer 2-oxetanone12 showed that theanharmonic estimates were in better agreement with experi-ment in this region. This result was achieved at a steepcomputational price, however, as the anharmonic calculationstook 40 times longer to complete. The calculated frequenciesfor 3-oxetanone are listed in Table 1 along with descriptions ofthe corresponding motions. The band labels follow theconvention established by Durig et al.11

4. SPECTRAL ASSIGNMENT AND ANALYSIS

Earlier spectroscopic work has shown that 3-oxetanone is aplanar asymmetric rotor of C2v symmetry.8 The three modesstudied in this work were first identified by Durig et al.11 andcorrespond to motions involving CO deformation out-of-plane (ν20), CO deformation in-plane (ν16) and ringdeformation (ν7). The portion of the spectrum containingthe ν16 and ν20 bands is shown in Figure 1.The rovibrational assignment and analysis of the ν7, ν16, and

ν20 bands of 3-oxetanone is detailed below.4.1. Assignment of the a-Type ν7 Band. The ν7 ring

deformation band has A1 symmetry corresponding to a-typeselection rules eo↔ee and oe↔oo. Upon initial inspection,there were clear rotational patterns visible in the P and Rbranches. The estimated band origin from the early lowresolution IR study11 together with the ground state rotationalconstants previously determined from the microwave work8

were used to simulate the rovibrational spectrum of this band.Once the observed patterns were identified on the basis ofcomparison with those simulated, the assignment of thequantum numbers of the first few transitions was confirmedusing ground state combination differences. After severalprogressions were assigned and fit using Watson’s A-reducedHamiltonian, Ir-representation in Pickett’s SPFIT program,17

the simulated spectrum was refined and the process continuedin an iterative fashion until the strongest transitions wereassigned.Overall, 2461 transitions were assigned from the P and R

branches up to Ka =29 with J(max) = 58. Another 307 linesfrom the Q branch were assigned with Ka values between 2 and35 and J values up to 40. Although only nine excited state

parameters were varied (three rotational constants and fivecentrifugal distortion constants for the ν7 state plus the bandorigin), the rms error of the fit was only 0.000 114 cm−1 in thepreliminary band analysis.

4.2. Assignment of b-Type ν16 and c-Type ν20 Bands.The ν20 band at 399.6 cm−1 corresponding to CO out-of-plane deformation, has B2 symmetry which gives rise to c-typetransitions with selection rules oe↔ee and eo↔oo. The ν16CO in-plane deformation mode at 448.2 cm−1 has B1symmetry corresponding to the b-type selection rules eo↔oeand ee↔oo. The assignment of these two bands was difficultnot only because the ν20 band is inherently weak but alsobecause its R branch significantly overlaps the P branch of theν16 band resulting in a very congested spectrum in this region asseen in Figure 1. To aid in the assignment, Loomis-Wood plotswere constructed for the ν16 and ν20 bands to identify patternsof the strongest progressions of lines using the Loomis−Woodadd-in (LWA)18 program for IgorPro. During the preliminarystages, although there were many transitions that lined up wellwith the predicted frequencies, a number of transitions werenoticeably shifted as shown in Figure 2. The preliminary fits ofthe individual bands revealed that certain excited statesparameters (most notably A and ΔK) were inconsistent withthose of the ground state and, more importantly, that this set ofconstants could not account for the remaining unassignedtransitions. Such discrepancies are indicative of a perturbation.Also, as seen in the band profiles in Figure 1, the spectrum ofthe ν20 band exhibits a 1:3 intensity ratio between the P and Rbranches which suggests the presence of intensity borrowing asa result of interaction with nearby energy levels. To fullyaccount for these observations, we started by seeking a set of“perturbation free” ground state spectroscopic constants asoutlined below.

4.3. Determination of the Ground State SpectroscopicConstants. The ground state spectroscopic constants of 3-oxetanone were determined using both infrared and microwavetransitions from the current work and ref 8, respectively.Initially, we used rovibrational transitions from the P and Rbranches of the three bands under study (a data set whichincludes a-, b- and c-type transitions) to obtain infrared ground

Figure 1. Overview spectrum showing the overlapping c-type ν20 and the b-type ν16 bands of 3-oxetanone between 360 and 500 cm−1.

The Journal of Physical Chemistry A Article

dx.doi.org/10.1021/jp3076673 | J. Phys. Chem. A 2012, 116, 9490−94969492

state combination differences for direct fitting of the groundstate parameters. Ground state combination differencescalculated from a-/b-type lines provide information about thehigh J and Kc energy levels whereas those from a c-type

spectrum cover the high Ka levels. Normally, when transitionsof these types are combined, there is such broad coverage ofenergy levels that the ground state parameters are well-characterized. In this case, however, the centrifugal distortionconstants ΔK and δk remained poorly determined at this stagedue to incomplete assignment of the weak, perturbed c-type ν20band. To remedy this, we began assignment of the c-type ringpuckering band (ν21) of 3-oxetanone centered at ∼139.5 cm−1

and produced a final data set containing 2532 infrared groundstate combination differences spanning J quantum numbersfrom 0 to 61, covering Ka values from zero up to 34.19 This wassupplemented with the nine previously reported pure rotationaltransitions8 and used to determine a complete, accurate set ofground state spectroscopic constants using Pickett’s SPFITprogram (A-reduction, Ir representation).17 The uncertaintyassigned to the pure rotational transitions was 1/1500 that ofthe infrared transitions to reflect the greater precision of themicrowave data. The overall rms error of the ground state fitwas 0.000 159 cm−1. The ground state spectroscopic parame-ters were held fixed in the subsequent infrared analysis of theν7, ν16, and ν20 bands.

4.4. Perturbation Analysis of the ν16 and ν20 Bands.Under C2v symmetry, the ν16 and ν20 bands of 3-oxetanone haveB1 and B2 vibrational symmetries, respectively, which can giverise to a first-order a-type Coriolis interaction. Once the groundstate energy levels were characterized in isolation (to removeany effect of the untreated perturbation), we reanalyzed the ν16and ν20 bands simultaneously by incorporating a symmetry-allowed first-order a-type Coriolis interaction parameter Ga intothe Hamiltonian. This allowed the assignment of the shiftedinfrared transitions (as shown in Figure 2). In total, 4143transitions were assigned for the b-type ν16 band including 949transitions from the Q branch. The spanned quantum numbersof assigned transitions include those in the P and R branchesfrom J = 1 through 74 with a maximum Ka value of 37.Transitions in the Q branch were assigned with Ka betweenzero and 32 and J values up to 46. For the weaker c-type ν20band, 2814 P and R branch transitions between Ka = 4 and 45with J(max) = 45 were assigned. An additional 594 Q branchtransitions with Ka values up to 19 were included to a maximumJ of 52. The rovibrational analysis of the 7551 transitionsassigned for these two vibrational states yielded a rms error ofonly 0.000 108 cm−1.

Figure 2. Zoomed-in section of the P branch of the ν16 band showingthe effect of Coriolis interaction with the neighboring ν20 band. Thestick spectrum shows the predicted transition frequencies in theabsence of perturbation. The transition marked with a star isnoticeably blue-shifted.

Table 2. Spectroscopic Constants (cm−1) for the Ground State, ν20, ν16, and ν7 Bands of 3-Oxetanone

ground statea ν20 ν16 ν7

ν 0 399.588606(6) 448.194670(5) 684.994728(7)Rotational ConstantsA 0.4045571(3) 0.40533(2) 0.40484(2) 0.40406372(4)B 0.16532500(2) 0.16542474(3) 0.16536694(3) 0.16519845(3)C 0.12297756(2) 0.12309730(3) 0.122781360(17) 0.122850816(17)Centrifugal Distortion Constants (×109)ΔJ 21.73(6) 21.984(11) 21.882(4) 21.895(8)ΔJK 150.6(4) 140.66(16) 164.62(15) 152.03(5)ΔK 103.4(4) 102.56(14) 112.94(12) 104.10(5)δj 5.28(3) 5.213(6) 5.367(2) 5.299(5)δk 90.8(8) 101.5(3) 86.5(3) 91.90(8)Ga(16,20) 0.3402(17)rms error 0.000159 0.000110

aThe ground state parameters were derived from the MW spectrum (ref 8) and infrared ground state combination differences and subsequently heldfixed in the fit of the excited states.

The Journal of Physical Chemistry A Article

dx.doi.org/10.1021/jp3076673 | J. Phys. Chem. A 2012, 116, 9490−94969493

4.5. Global Analysis of the ν7, ν16, and ν20 Bands. Aglobal fit incorporating all 10 319 assigned rovibrationaltransitions from the ν7, ν16, and ν20 bands was performed inPickett’s SPFIT program using Watson’s A-reduced Hamil-tonian, Ir-representation17 with the ground state spectroscopicconstants held fixed. The rms error of the simultaneous threeband fit was 0.000 110 cm−1 and the resulting spectroscopicconstants are listed in Table 2.

5. DISCUSSIONThe low resolution survey of 3-oxetanone by Durig et al.11

provided a general overview of the approximate frequencies,selection rules and intensities of the fundamental vibrations ofthis molecule. This served as a preliminary guide for our highresolution work in the far-infrared region and allowed accuratecharacterization of the ground state as well as the ν7, ν16, andν20 excited states. The simultaneous fit of over ten thousandinfrared transitions with such a remarkably low rms error(0.000 110 cm−1) suggests that the model Hamiltonianemployed provides a good description of these three rovibra-tional bands. At a more fundamental level, the high resolutionspectra have also confirmed the nature of the vibrationalmotions that are involved in these transitions as the observedpatterns are consistent with selection rules related to primarilyin-plane (ν7, ν16) and out-of-plane (ν20) modes. Theunambiguous assignment of the rotational structure was criticalin this regard as the low frequency modes of the structuralisomer 2-oxetanone were originally misassigned when only theunresolved band profiles were observed.13

Although the microwave spectrum of 3-oxetanone offersgreater precision, the data set included only nine a-typetransitions that sampled energy levels to J = 3. In the presentwork, energy levels up to J = 61 and covering a broad range ofKa and Kc values were used to derive the ground statespectroscopic parameters via ground state combination differ-ences. This method removes the upper state dependence toisolate any perturbation and thus ensures that the resultingspectroscopic constants are physically meaningful rather thaneffective fit parameters. We believe that our treatment of theground state in this analysis has established a complete,accurate set of ground state spectroscopic parameters of 3-oxetanone as listed in Table 2 which are useful in this andfuture rotationally resolved studies of this heterocycle.During the initial assignment of the closely spaced ν16 and

ν20 bands, the rotational constant A and the centrifugaldistortion constant ΔK seemed to be effective constants andvaried more than expected from the analogous ground stateparameters as shown in Table 3. Furthermore, these effectiveparameters did not account for many of the weaker transitionsin this spectral region. The addition of the Coriolis termprovided better consistency among the spectroscopic constantsas seen in Table 3 although the excited state A rotationalconstants were less well-determined due to a high correlation

between Ga and A. Overall, the rms error of the fit was reducedby a factor of 3 when Ga was included, suggesting that this setof constants provides a better description of the 7551transitions assigned in the carbonyl deformation region of thespectrum.The Coriolis interaction term was found to have a value of

0.3402(17) cm−1, and the positive sign indicates that this is apositive perturbation in which the R branch of the lowerfrequency band (ν20) and the P branch of the higher frequencyband (ν16) gain intensity while the other branches are depleted.This leads to the asymmetry in the band profiles shown inFigure 1. Classically, this occurs when two perpendicularvibrational transitions (in this case the ν16 in-plane and ν20 out-of-plane CO deformations) are enhanced by the rotation ofthe molecule producing Coriolis forces in the same sense as thevibrational angular momentum of the adjacent mode.20

In this study, the band origins of the ν20 (399.588 606(6)cm−1), ν16 (448.194 670(5) cm−1), and ν7 (684.994 728(7)cm−1) modes were accurately determined for the first time andcompare well with the gas phase infrared spectrum reported byDurig et al. (401 cm−1, 448 and 683 cm−1).11 These values canbe compared with the ab initio predictions of the harmonic andanharmonic frequencies in Table 1. An additional columnlisting scaled harmonic frequencies is also included which weredetermined using an empirical scaling factor (0.9679) derivedfrom comparing experimental and ab initio frequencies for a setof 125 small molecules.21 When the band centers determined inthis work are compared with the harmonic frequencies derivedusing DFT theory (410, 449, and 688 cm−1), the calculatedvalues differ by less than 10 cm−1 from their experimentalcounterparts. The scaled harmonic (397, 434, and 666 cm−1)and anharmonic frequencies (411, 444, and 680 cm−1) provideno systematic improvement in this spectral region. This may beserendipitous or may suggest that these particular modes of 3-oxetanone are fairly harmonic. The latter assertion is consistentwith the ab initio derived potential for the ring puckering modeof 3-oxetanone, which has a small anharmonic correctioncompared with those of related heterocycles.7

Upon comparison of the band assignments of all 21vibrational fundamentals of 3-oxetanone provided by Durig etal.11 with the ab initio estimates in Table 1, several notablefeatures were identified. At this level of theory, the harmonicvalues, in general, match the experimental frequencies to within3% across the entire vibrational spectrum. The scaled harmonicvalues provide the best estimates (well within 1%) of thehighest energy vibrations in the CH stretching region above2900 cm−1 but provide no systematic benefit over the othermethods in predicting the frequencies of bending anddeformation motions. This suggests that empirical scalingfactors are of limited use for predicting far-infrared spectra ofsuch molecules. Our ab initio study also identified severalinconsistencies with the original assignments given in ref 11.The most obvious discrepancy is with the ν10 fundamental,

Table 3. Subset of Spectroscopic Constants (cm−1) That Illustrate the Effect of Including and Excluding the First-Order a-TypeCoriolis Interaction Term between the ν16 and ν20 Bands

without perturbation treatment perturbation treatment

ground state ν20 ν16 ν20 ν16

A 0.4045571(3) 0.40295009(7) 0.40721404(7) 0.40533(2) 0.40484(2)109ΔK 103.4(4) 11.52(13) 205.16(7) 102.56(14) 112.94(12)Ga 0.3402(17)rms error 0.000295 0.000108

The Journal of Physical Chemistry A Article

dx.doi.org/10.1021/jp3076673 | J. Phys. Chem. A 2012, 116, 9490−94969494

which was assigned on the basis of the Raman spectrum ofliquid 3-oxetanone. The experimental value of the band center(1102 cm−1) differs by more than 300 cm−1 from the calculatedvalue (770 cm−1). This misassginment is likely a result of thelow Raman intensity of the ν10 band, which is calculated to beless than 2% that of the nearest Raman active mode (ν7) lessthan 100 cm−1 away. Another discrepancy was discovered in theassignment of the ν4 (CH2 wag) and ν18 (CH2 twist) modes at1134 and 1284 cm−1, respectively. These are reversed accordingto the ab initio calculations. The former should be at higherfrequency (1311 cm−1 harmonic) than the latter (1158 cm−1

harmonic). This could easily be confirmed from a highresolution spectrum in this region in the future as the wagand twist modes obey a-type and c-type selection rules,respectively. As the band envelopes of these modes areapproximately 60 cm−1 wide, the observation of the ν4 andν18 bands in the original study was likely obstructed by theneighboring ν13 (1272 cm−1) and ν14 (1087 cm−1) vibrations,which are estimated to be approximately 1 and 2 orders ofmagnitude more intense, respectively. Finally, the band centersreported for the ν5 ring breathing and ν15 ring deformationbands differed by ∼15 cm−1 on the basis of the liquid IR andRaman spectra and the ν15 band was not assigned in the gasphase spectrum at all. On the basis of our calculations, the bandcenters for these two ring vibrations differ by only onewavenumber and the ν15 band is actually a factor of 3 moreintense! As we collected spectra up to 1130 cm−1 during ourstudy of the ν7 vibration, we serendipitously observed the ν5and ν15 bands which are centered near 982 and 983 cm−1,respectively. The former is weaker with a clear a-type Q branchand the latter is more intense and obeys b-type selection rulesas expected on the basis of the symmetry of the ring breathingand deformation motions. In fact, we have partially assigned theb-type band but there is evidence of a strong perturbationwhich will be difficult to unravel without assignment of theburied a-type spectrum.

6. CONCLUSIONS

In this work, we have assigned 10 319 rovibrational transitionsfrom the ν7, ν16, and ν20 modes of 3-oxetanone and derivedaccurate band centers and rotational and centrifugal distortionconstants for these three vibrational states. The perpendicularmodes corresponding to carbonyl deformation (ν16 and ν20)were mutually perturbed and a first-order a-type Coriolisparameter was employed to treat this interaction. The groundstate of 3-oxetanone has been extensively characterized throughfitting of infrared ground state combination differences thatsample quantum states up to J = 61 and Ka = 34 along with thepreviously reported microwave transitions.8 These parameterswill be useful for further rotationally resolved spectra of 3-oxetanone. On the basis of our ab initio calculations, we haveidentified several inconsistencies in the assignment offundamental bands in the original low resolution vibrationalstudy of 3-oxetanone. The current work has demonstrated theimportance of high resolution infrared spectroscopy incharacterizing low frequency vibrations of organic buildingblocks and has laid important groundwork for further studieson this and related molecules.

■ ASSOCIATED CONTENT*S Supporting InformationComplete listing of the assigned rovibrational transitions for theν7, ν16, and ν20 bands. This material is available free of chargevia the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: vanwijng@cc.umanitoba.ca. Phone: (204)474-8379.Fax: (204)474-7608.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis research was funded by the Natural Sciences andEngineering Research Council of Canada (NSERC) throughthe Discovery Grants program. The authors are grateful toBrant Billinghurst (Canadian Light Source) for technicalsupport for the experiments in Saskatoon.

■ REFERENCES(1) Dejaegher, Y.; Kuzmenok, N. M.; Zvonok, A. M.; De Kimpe, N.Chem. Rev. 2002, 102, 29−60.(2) Wuitschik, G.; Rogers-Evans, M.; Muller, K.; Fischer, H.; Wagner,B.; Schuler, F.; Polonchuk, L.; Carreira, E. M. Angew. Chem., Int. Ed.2006, 45, 7736−7739.(3) Wuitschik, G.; Rogers-Evans, M.; Buckl, A.; Bernasconi, M.;Marki, M.; Godel, T.; Fischer, H.; Wagner, B.; Parrilla, I.; Schuler, F.;Schneider, J.; Alker, A.; Schweizer, W. B.; Muller, K.; Carreira, E. M.Angew. Chem., Int. Ed. 2008, 47, 4512−4515.(4) Combs, L. L.; Rossie, M., Jr. J. Mol. Struct. 1976, 32, 1−7.(5) Pasto, D. J.; Chipman, D. M.; Worman, J. J. J. Phys. Chem. 1982,86, 3981−3989.(6) Thomson, C. J. Mol. Struct. THEOCHEM 1982, 88, 55−60.(7) Vansteenkiste, P.; Van Speybroeck, V.; Verniest, G.; De Kimpe,N. t.; Waroquier, M. J. Phys. Chem. A 2007, 111, 2797−2803.(8) Gibson, J. S; Harris, D. O J. Chem. Phys. 1972, 57, 2318−2328.(9) Meinzer, A. L; Pringle, W. C. J. Phys .Chem. 1976, 80, 1178−1180.(10) Carreira, L. A.; Lord, R. C. J. Chem. Phys. 1969, 51, 3225−3231.(11) Durig, J. R; Morrissey, A. C; Harris, W. C. J. Mol. Struct. 1970, 6,375−390.(12) Chen, Z.; van Wijngaarden, J. J. Mol. Spectrosc. 2009, 257, 164−169.(13) Chen, Z.; van Wijngaarden, J. J. Mol. Spectrosc. 2011, 268, 107−111.(14) Tokaryk, D. W.; van Wijngaarden, J. Can. J. Phys. 2009, 87,443−448.(15) Rothman, L. S.; Jacquemart, D.; Barbe, A.; Chris Benner, D.;Birk, M.; Brown, L. R.; Carleer, M. R.; Chackerian, C., Jr.; Chance, K.;Coudert, L. H.; Dana, V.; Devi, V. M.; Flaud, J.-M.; Gamache, R. R.;Goldman, A.; Hartmann, J.-M.; Jucks, K. W.; Maki, A. G.; Mandin, J.-Y.; Massie, S. T.; Orphal, J.; Perrin, A.; Rinsland, C. P.; Smith, M. A.H.; Tennyson, J.; Tolchenov, R. N.; Toth, R. A.; Vander Auwera, J.;Varanasi, P.; Wagner, G. J. Quantum Spectrosc. Radiat. Trans. 2005, 96,39−204.(16) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A.; , Jr., Vreven, T.;Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J.B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R.E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J.

The Journal of Physical Chemistry A Article

dx.doi.org/10.1021/jp3076673 | J. Phys. Chem. A 2012, 116, 9490−94969495

J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.;Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;Martin, R. L.; Fox, D. J.; Keith, T.; M. A. Al-Laham, Peng, C. Y.;Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,W.; Wong, M. W.; Gonzalez, C.; , Pople, J. A. Gaussian 03, RevisionC.02; Gaussian, Inc., Wallingford CT, 2004.(17) Pickett, H. M. J. Mol. Spectrosc. 1991, 148, 371−377.(18) Neese, C. F. Loomis-Wood Add-In (LWA) for Igor Pro, Version2.0; 56th Ohio State University International Symposium onMolecular Spectroscopy, Columbus, OH, June 20−24, 2005, http://fermi.uchicago.edu/freeware/LoomisWood.shtml.(19) Chen, Z.; van Wijngaarden, J. J. Mol. Spectrosc. 2012, JMS-12-148, DOI: 10.1016/j.jms.2012.08.006.(20) Mills, I. M. Pure Appl. Chem. 1965, 11, 325−344.(21) Andersson, M. P.; Uvdal, P. J. Phys. Chem. A 2005, 109, 2937−2941.

The Journal of Physical Chemistry A Article

dx.doi.org/10.1021/jp3076673 | J. Phys. Chem. A 2012, 116, 9490−94969496