This document is © Elsevier Science, 1994

Cross-conjugated compounds:
microwave spectrum of 4,4-dimethyl-2,5-cyclohexadien-1-one

Wolfgang Hutter and Hans-Karl Bodenseh

Abteilung Chemische Physik, University of Ulm, D-89069 Ulm (Germany)

Presented in part at the 12th Colloquium on High Resolution Molecular Spectroscopy, September 9th to 13th 1991 at Dijon.

Published in part in J. Mol. Struct., 291 (1993) 151-158

This file is formatted in HTML3.

Abstract

Microwave transitions of the cross-conjugated compound 4,4-dimethyl-2,5-cyclohexadien-1-one have been measured over extended regions from 11 to 40 GHz. The spectrum is a classical a-type spectrum of a near-prolate top with many interspersed lines presumably originating from excited vibrational states. No internal rotation splitting could be observed. 108 rotational transitions with J-quantum numbers up to 40 have been assigned for the vibrational ground state. The least squares fit yielded the three rotational constants A = 3332.158(13) MHz, B = 1193.07386(52) MHz, C = 1082.21280(45) MHz, as well as all five quartic centrifugal distortion constants from Watson's A-reduction: DJ=0.02149(73) kHz, DJK=1.1008(16) kHz, DK=-68.42(32) kHz, dJ = 0.00332(16) kHz, dK = -2.7513(99) kHz (representation IR used). The rotational constants of the n36=1 state are also given in this paper. The dipole moment was determined from the Stark shifts of the M-components of two transitions and was found to be 4.4522(83) D from 93 measurements.

Introduction

In the past the cyclohexadienone system has been the subject of numerous investigations. There are several reasons for this strong interest. The first one is the presence of this system in biologically important compounds [1]. Secondly, it represents an isomeric form of the phenol system. For this reason the dienone-phenol rearrangement has been investigated thoroughly using a variety of substituted cyclohexadienones [2,3]. For a review on this matter see [4]. A third area where intensive studies have been made is the photochemistry of the cyclohexadienone system [5] which has also been reviewed [6]. Last but not least these compounds belong to the class of so-called "cross-conjugated" compounds and are therefore of particular interest to organic theoretical chemists [1, 7].

Although the term "cross-conjugation" is found in many publications the concept itself seems to be rather unclear. This is evidenced by the fact that even a clear definition of cross-conjugation is lacking. One requirement that is basic to this concept and is agreed upon by most authors is that there have to be two double bonds that are in conjugation with a third one, but not in a linear arrangement. The p electronic system thus forms a bifurcation. So the simplest cross-conjugated hydrocarbon would be 3-methylene-1,4-pentadiene (I) that can be viewed as an ethene disubstituted geminally with two vinyl groups:

The structure implied by this illustration is not quite correct since the distance between the "inside" hydrogens on carbon atoms 1 and 5 would be smaller than the added van-der-Waals radii. The structure proposed by [8] is anti-skew with a dihedral angle of about 40 degrees between the planar butadiene fragment and the vinyl group.

Another requirement that some authors introduce is that the two "outer" double bonds are not in conjugation to each other. While this is the case for the aforementioned methylene-pentadiene, it is not true for its cyclic analogue fulvene (II). Although fulvene is considered to be cross-conjugated by many authors, it would not fit to this definition of cross-conjugation.

Some authors expand the definition of cross-conjugation even further: the "outer" double bonds can then be exchanged for "unsaturated centers". These include hetero-atoms with lone electron pairs as nitrogen or oxygen. By this definition even urea is said to be cross-conjugated [7].

In this paper we will use the first (and most common) definition of cross-conjugation, i.e. there have to be (at least) two double bonds that are conjugated to a "central" double bond in such a way that the p electronic system forms a bifurcation.

Cross-conjugated molecules according to this definition include quinones, fulvenes, organic dyes, fused aromatic hydrocarbons and as mentioned above many bioorganic compounds like steroids, terpenes and alkaloids.

One characteristic feature of many cross-conjugated compounds is an enhanced reactivity. Some compounds are even highly unstable. This is easily seen by looking at the most simple representatives fulvene, cyclopentadienone, cyclohexadienone or benzoquinone.

The first cross-conjugated compound examined by microwave spectroscopy was fulvene (II) [9,10]. Its oxygen analogue cyclopentadienone (III) has also been studied [11,12]. All of these works clearly showed that the five-ring system has a planar geometry. The question about the planarity of the six-membered rings is more difficult to answer and indeed the answer that have been given so far are contradictory. Six-membered heterocyclic systems like g-pyranone and its sulfur analogues were shown to have planar structures [13] by microwave spectroscopy whereas studies on 4,4-Dimethyl-1-methylene-2,5-cyclohexadiene (IV) by electron diffraction seem to indicate a non-planar structure [14]. The unmethylated parent compound has also been prepared [15]. But as it reacts rather quickly to form a mixture of toluene and dimeric products [16] only kinetic and thermodynamic studies have been carried out so far.

Note added in proof: In the meantime a microwave study on this compound has also been performed.

The situation is even more difficult for the corresponding dienone (V). It tautomerizes very rapidly to phenol. The detection of the molecule has been claimed only twice: with infrared spectroscopy at -196°C [17] and with the flowing afterglow technique [18], but no structural data are available. We tried to produce (V) by flash pyrolysis and to detect the unstable molecule by microwave spectroscopy, but were not able to find any lines originating from the desired product. Instead we measured only the spectrum of phenol [19] which indicates that even in the gas phase at very low pressures the molecule rearanges very rapidly.

Since the instability of the cyclohexadienone is caused by tautomerism, i.e. migration of a hydrogen atom from C-4 to the oxygen atom, it is evident that double methylation at the 4-position stabilizes this molecule, just as for the methylene compound (IV). So we decided to investigate the dimethyl-substituted compound (IX) that was characterized first by Yanagita, Tahara and Ohki [20].

Experimental

Due to plans for further studies and to get a purer product 4,4-Dimethyl-2,5-cyclohexadien-1-one (IX) was prepared from 4,4-Dimethyl-2-cyclohexen-1-one (VI) in a 3-step reaction rather than by the single-step methods described already [21]. In the first step the carbonyl group was converted to a trimethylsilylether (VII) using trimethylsilylchloride [22]. This was followed by bromination with cleavage of the Si-O-bond [23] to give the bromide (VIII). In the last step, hydrogen bromide was eliminated by quinoline [24] yielding the desired dienone:

The compound thus prepared is stable over a period of at least several weeks when stored under nitrogen or in an evacuated flask, but oxidizes slowly when exposed to air.

The spectra were recorded using a conventional Stark-spectrometer with a thermostatized 3m-cell that allows measurements at constant temperatures between +40°C and -70°C.

Microwave spectrum

As expected for a molecule with a high dipole moment and a near-prolate geometry most lines could be detected with extremely low Stark fields (about 10 V/cm), especially for high-KP-transitions.

The spectrum is typical of an a-type near-prolate rotor. There are large areas with no or few transitions and areas spaced by B+C where many lines lie very close together.

The assignment of some transitions proved to be rather difficult due to many lines originating from vibrationally excited states. We were not too interested in the vibrational behavior of this molecule and so only one state (n36=1) has been investigated in some detail.

The key to assigning the spectrum was to examine low-J-transitions with fewer K-components at low temperatures. The cooling however was limited by the vapor-pressure of about 0.008 mbar at -45°C. Most measurements were therefore carried out at about -30°C with a pressure of about 0.01 mbar. In the course of our investigation 108 lines of the vibrational ground state could be assigned (see Table 1) to give a least squares fit with a standard deviation of 30 kHz which corresponds well to the estimated accuracy of our measurements. The mesurements on the n36=1 state were made with low resolution. Therefore the fit for the 28 lines that have been assigned (see Table 2) shows a higher standard deviation of 534 kHz corresponding to the lower accuracy of these measurements. The spectroscopic constants are given in Table 3.

Internal rotation

The possible internal rotation splitting (two geminal methyl groups) could not be observed for the transitions assigned which is not surprising since a conformational analysis carried out by two different molecular modeling programs (SYBYL [25] and DISCOVER/INSIGHT [26]) predicted rotational barriers of about 17-20 kJ/mol.

Dipole moment

The dipole moment was determined by investigation of the Stark-effect splittings of the two transitions 606-505 and 707-606. From the 13 M-components of these two transitions a total of 93 measurements was made at field strengths from 350 to 1120 V/cm. The electric field in the Stark cell was calibrated with OCS using the known dipole moment of 0.71521 D [27]. The figure below shows a typical spectrum of the transition 707-606 along with a calculated spectrum. A least squares fit of the obtained frequencies yielded a dipole moment of 4.4522(83) D. This is in rough agreement with the values given by Bertelli and Andrews [28] who calculated the dipole moment with the CNDO/2 method and also determined it experimentally by dielectric constant measurements. The values obtained by them were 4.46 D (calc.) and 4.36 D (exp.) respectively.

Spectrum of the transition 707-606 registered at a static field of ca. 590 V/cm with a modulation field of 825 V/cm (above) and calculated using Lorentzian line-shapes with full half-widths of 500 kHz (below). The width of the recorded range represents a compromise between small ranges that would be desirable for high accuracy and wide ranges permitting to measure several lines at one time. The broken line shows the zero-field frequency of the transition. The component with M=6 lies outside the recorded range. The signals are labeled with their respective values of M. The signal labeled 3* represents the component M=3 at 590 + 825 = 1415 V/cm. All other M*-components are shifted outside the depicted range.

Ab initio and semi-empirical calculations

To get first indications to the rotational constants we had to expect, we carried out some theoretical investigations.

The molecule under study possesses CS symmetry if the ring is non-planar, whereas the symmetry is C2v if it is planar. So we started off with only CS symmetry imposed on the calculations.

We first investigated cyclohexadienone (V). For initial parameters we optimized its structure with the aid of a molecular modeling package called SYBYL [25]. The package used a TRIPOS-forcefield and the optimized structure was planar no matter if the starting-conformation was planar, boat, chair or half chair. So we twisted the molecule arbitrarily to a conformation where the C6-C1-C2-plane and the C3-C4-C5-plane were at plane angles of 2 and 10 degrees respectively to the C2-C3-C5-C6-plane. This conformation was then used as the initial conformation for the ab-initio optimizations with only CS symmetry required. These were carried out with the GAUSSIAN 86 [29] program suite on a DEC VAX-6440. We performed a full optimization at the HF-SCF level using the Berny algorithm and standard convergence criteria on forces and displacements.

The geometry we thus obtained with a RHF/6-31G* basis set had an essentially planar ring with dihedral angles of about 0.1 degrees. The resulting parameters were taken as starting values for the optimization of the 4,4-dimethylated-compound. The initial parameters for the methyl groups were: 1,10 Å for C-H distances, 109,5 degrees for H-C-H angles and 1,54 Å for C-C distances. As the previous calculation had indicated that the methyl torsion has a very high barrier, the methyl groups were "frozen" at their potential minima. Apart from this constraint all other parameters were allowed to relax simultaneously. As before, only CS symmetry was fixed. The resulting ring structure, again, was essentially planar with dihedral angles of about 0.004 degrees and so we restrained the molecule to the indicated C2v symmetry to obtain a conformation with a perfectly planar ring without having to increase computing time overly.

A planar ring conformation was also obtained using a STO-3G basis set and in semi-empirical calculations with MNDO, MINDO/3 and AM1 hamiltonians using the MOPAC [30] package. The resulting rotational constants and dipole moments of the RHF/6-31G* and AM1 calculations are compared to the experimental values in Table 4.

There is no possibility to decide whether the ring in dimethyl-cyclohexadienone is planar or not just from the rotational constants or the planar moments. This is due to the fact that the geometry of the methyl groups influences the planar moment Pcc more than the (non-)planarity of the ring. Whereas this planar moment varies by ca. 1.5% for different ring-conformations (dihedral angles below 10 degrees), the differences between two planar structures with different but reasonable geometries of the methyl-groups can be well over 3%. But since the ab-initio and semi-empirical calculations, which resulted in planar structures, predicted all three measured rotational constants with remarkable accuracy (the deviations with RHF/6-31G* are about 0.2% for A and 1% for B and C, with AM1 they are 1.1% for A and 0.3% for B and C) there are good reasons for assuming planarity for the ring.

Acknowledgments

Financial support by the Deutsche Forschungsgemeinschaft and the Fonds der Chemischen Industrie is gratefully acknowledged.

Note added in proof: Additional information on this work is contained in the dissertation thesis of W.H. (in German).


References

1W. Gramlich, Liebigs Ann. Chem., 1979, 121.
2K. L. Cook and A. J. Waring, J. Chem. Soc., Perkin Trans. 2, 1973, 88.
3V. P. Vitullo and N. Grossman, J. Am. Chem. Soc., 94 (1972) 3844.
4H. Krauch and W. Kunz, Reaktionen in der Organischen Chemie, 5th ed., p. 350, Hüthig, Heidelberg, 1976.
5H. E. Zimmerman, P. Hackett, D. F. Juers, J. M. McCall, and B. Schröder, J. Am. Chem. Soc., 93 (1971) 3653.
6P. J. Kropp in Organic Photochemistry (O. L. Chapman), 1st ed., Vol. I, pp. 1-90, Dekker, New York, 1967.
7N. F. Phelan and M. Orchin, J. Chem. Educ., 45 (1968) 633.
8A. Almenningen, A. Gatial, D. S. B. Grace, H. Hopf, P. Klaeboe, F. Lehrich, C. J. Nielsen, D. L. Powell, and M. Trætteberg, Acta Chem. Scand., A 42 (1988) 634.
9R. D. Brown, F. R. Burden, and J. E. Kent, J. Chem. Phys., 49 (1968) 5542.
10P. A. Baron, R. D. Brown, F. R. Burden, P. J. Domaille, and J. E. Kent, J. Mol. Spectrosc., 43 (1972) 401.
11P. A. Baron and R. D. Brown, Chem. Phys., 1 (1973) 444.
12K. Bestle and H.-K. Bodenseh, 12th Coll. High Res. Mol. Spec. Dijon (1991) F12.
13J. N. Macdonald, S. A. Mackay, J. K. Tyler, A. P. Cox, and I. C. Ewart, J. Chem. Soc., Faraday Trans., 2, 77 (1981) 79.
14M. Trætteberg, P. Bakken, A. Almenningen, W. Lüttke, and J. Janssen, J. Mol. Struct., 81 (1982) 87.
15H. Plieninger and W. Maier-Borst, Chem. Ber., 98 (1965) 2504.
16J. J. Gajewski and A. M. Gortva, J. Am. Chem. Soc., 104 (1982) 334.
17M.-C. Lasne, J.-L. Ripoll, and J.-M. Denis, Tetrahedron Lett., 21 (1980) 463.
18C. S. Shiner, P. E. Vorndam, and S. R. Kass, J.Am.Chem.Soc., 108 (1986) 5699.
19W. Hutter, Diploma Thesis, Universität Ulm (Germany), 1990.
20M. Yanagita, A. Tahara, and E. Ohki, J. Pharm. Soc. Japan, 71 (1951) 1060 (Chem. Abstr., 46 (1952) 5004i).
21K. L. Cook and A. J. Waring, J. Chem. Soc., Perkin Trans., 1, 1973, 529.
22H. O. House, L. J. Czuba, M. Gall, and H. D. Olmstead, J.Org.Chem., 34 (1969) 2324.
23L. Blanco, P. Amice, and J. M. Conia, Synthesis (1976) 194.
24F. G. Bordwell and K. M. Wellman, J. Org. Chem., 28 (1963) 2544.
25Tripos Inc., 1699 S. Hanley Road, Suite 303, St. Louis, MO, 63144, SYBYL 5.2. (1988).
26Biosym Technologies, 10065 Barnes Canyon Road, San Diego, CA, 92121,Insight II.
27J. S. Muenter, J. Chem. Phys., 48 (1968) 4544.
28D. J. Bertelli and T. G. Andrews, Jr., J. Am. Chem. Soc., 91 (1969) 5280.
29M. J. Frisch, J. S. Binkley, H. B. Schlegel, K. Raghavachari, C. F. Melius, R. L. Martin, J. J. P. Stewart, F. W. Bobrowicz, C. M. Rolfing, L. R. Kahn, D. J. Defrees, R. Seeger, R. A. Whiteside, D. J. Fox, E. M. Fleuder, and J. A. Pople, Gaussian 86, Carnegie-Mellon Quantum Chemistry Publishing Unit, Pittsburgh PA, 1984.
30aM. J. S. Dewar, J. Mol. Struct., 100 (1983) 41.
bM. J. S. Dewar, R. C. Bingham, and D. H. Lo, J. Am. Chem. Soc., 97 (1975) 1285.
cM. J. S. Dewar and W. Thiel, J. Am. Chem. Soc., 99 (1977) 4899.
dM. J. S. Dewar, E. G. Zoebisch, E. F. Healy, and J. J. P. Stewart, J. Am. Chem. Soc., 107 (1985) 3902.

TABLE 1

Rotational transitions of 4,4-dimethyl-2,5-cyclohexadienone (frequencies in GHz)
NJKPKO J'KP'KO' nexpncalcexp.-calc.
150540411.29421011.294197+0.000013
251441311.63410711.634143-0.000036
360650513.51098113.510941+0.000040
471761615.48979115.489756+0.000035
570760615.70925315.709291-0.000038
672662515.89420215.894298-0.000096
776166015.93824915.938227+0.000022
875265115.94156115.941523+0.000038
974464315.94759615.947543+0.000053
1074364215.94768315.947777-0.000094
1173563415.95570215.955731-0.000029
1273463315.96810615.968115-0.000009
1372562416.11172616.111751-0.000025
1471661516.25341616.253466-0.000050
1581871717.68653317.686501+0.000032
1680870717.89090417.890909-0.000005
1782772618.15103018.151039-0.000009
1887177018.21433718.214522-0.000185
1986276118.21715218.217497-0.000345
2085375218.22186418.222370-0.000506
2184574418.23102618.231183-0.000157
2291981819.87848619.878431+0.000055
2390980820.05924620.059254-0.000008
2492882720.40239220.402370+0.000022
2598188020.49072920.490751-0.000022
2697287120.49352720.493543-0.000016
2796386220.49771020.497720-0.000010
2895485320.50461220.504621-0.000009
2994684520.51686420.516850+0.000014
3094584420.51836520.518384-0.000019
3193783620.52619520.526165+0.000030
3292782620.81908920.819100-0.000011
331011091922.06579922.065767+0.000032
341001090922.21861222.218660-0.000048
35102992822.64776822.647753+0.000015
36109199022.76688822.766891-0.000003
37108298122.76959522.769579+0.000016
38106496322.77901722.779017+0.000000
39105695522.78841422.788396+0.000018
40105595422.78847322.788465+0.000008
41104794622.80473322.804658+0.000075
42104694522.80793322.807962-0.000029
43103893722.81153522.811528+0.000007
44103793622.88612322.886141-0.000018
45101991823.09969723.099779-0.000082
46102892723.17853623.178570-0.000034
47121111111027.31864627.318834-0.000188
48121021110127.32148427.321456+0.000028
491293119227.32481627.324797+0.000019
501284118327.32924127.329272-0.000031
511275117427.33556227.335628-0.000066
521266116527.34546027.345325+0.000135
531258115727.36153627.361355+0.000181
541257115627.36192127.361747+0.000174
5512310113927.37829127.378145+0.000146
561249114827.38676227.386705+0.000057
571248114727.39905427.398875+0.000179
58130131201228.67951428.679496+0.000018
59141141311330.77772930.777716+0.000013
60140141301330.83420230.834174+0.000028
61151151411432.94897332.948884+0.000089
62150151401432.99057032.990566+0.000004
63152131421234.85486634.854917-0.000051
64160161501535.14860835.148627-0.000019
65161151511436.31318536.313271-0.000086
66161511515036.42095736.420961-0.000004
67161421514136.42374336.423754-0.000011
68161331513236.42693736.426946-0.000009
69161241512336.43064036.430698-0.000058
70161151511436.43525636.435250+0.000006
71161061510536.44095436.440969-0.000015
721697159636.44853236.448445+0.000087
731688158736.45865036.458671-0.000021
74163141531336.46706536.467132-0.000067
751679157836.47346936.473446+0.000023
76166111561036.49623736.496212+0.000025
7716610156936.49637436.496412-0.000038
78165121551136.53257136.532543+0.000028
79165111551036.53811036.538122-0.000012
80164131541236.56938336.569426-0.000043
81164121541136.65748236.657455+0.000027
82163131531237.04037937.040405-0.000026
83171171611637.28628637.286306-0.000020
84170171601637.30813937.308140-0.000001
85171611616038.69604038.696046-0.000006
86171521615138.69893038.698923+0.000007
87171431614238.70217238.702152+0.000020
88171341613338.70587038.705868+0.000002
89171251612438.71027738.710264+0.000013
90171161611538.71563238.715629+0.000003
91171071610638.72238338.722403-0.000020
921798169738.73130838.731295+0.000013
931789168838.74352138.743502+0.000019
9417710167938.76116938.761196-0.000027
95176121661138.78853238.788470+0.000062
96176111661038.78885938.788867-0.000008
97175131651238.83103338.831027+0.000006
98175121651138.84065938.840667-0.000008
99174141641338.86578038.865780-0.000000
100174131641238.99651438.996629-0.000115
101181181711739.45328839.453274+0.000014
102193171911815.97491615.974913+0.000003
103334293343016.50616116.506161+0.000000
104223202212117.90736917.907364+0.000005
105313283132921.29089621.290890+0.000006
106263242612521.30168321.301202+0.000481
107403374033832.05442832.054436-0.000008
108256202542138.58196438.581968-0.000004
Standard deviation: 30 kHz

TABLE 2

Rotational transitions of 4,4-dimethyl-2,5-cyclohexadienone, n36=1 (frequencies in GHz)
NJKPKO J'KP'KO' nexpncalcexp.-calc.
11258115727.34977427.3490410.000733
21257115627.35035527.3494040.000951
312310113927.36693827.3664210.000517
4141141311330.78199330.7811730.000820
5140141301330.83903030.839298-0.000268
6161421514136.40895036.4087960.000154
7161331513236.41170936.411923-0.000214
8161241512336.41533436.415593-0.000259
9161151511436.42069436.4200380.000656
10161061510536.42553036.425615-0.000085
111697159636.43285136.432898-0.000047
121679157836.45699436.457224-0.000230
13165121551136.51472136.514766-0.000045
14165111551036.52052236.5199400.000582
15164131541236.55227336.5515310.000742
16162141521337.12735137.1272000.000151
17172161621538.17345738.174016-0.000559
18171611616038.68009438.680211-0.000117
19171521615138.68290538.683037-0.000132
20171431614238.68597738.686204-0.000227
21171341613338.68976838.689841-0.000073
22171251612438.69401638.694138-0.000122
23171161611538.69921238.699374-0.000162
2417810168938.72733138.7265180.000813
25176121661138.76943938.770247-0.000808
26175121651138.82097538.8207080.000267
27174141641338.84658838.846829-0.000241
28174131641238.96977638.970599-0.000823
Standard deviation: 534 kHz

TABLE 3

Spectroscopic constants and dipole moment of 4,4-dimethyl-2,5-cyclohexadienone (representation IR used).
(Numbers in parentheses are standard errors of the last digits given).
n36=0 n36=1
A3.332158(13) GHz3.332158(13) GHz
B1.19307386(52) GHz1.19307386(52) GHz
C1.08221280(45) GHz1.08221280(45) GHz
DJ0.02149(73) kHz0.02149 kHz*
DJK1.1008(16) kHz1.1008 kHz*
DK-68.42(32) kHz-68.42 kHz*
dJ0.00332(16) kHz>0.00332 kHz*
dK-2.7513(99) kHz-2.7513 kHz*
aA36 -8,2(22) MHz
aB36 -1,118(20) MHz
aC36 -0,229(20) MHz
k-0.901-0.903
m4.4522(83) D 
* Values restricted to those of vibrational ground state

TABLE 4

Comparison between measurements and theoretical calculations on 4,4-dimethyl-2,5-cyclohexadien-1-one.
Rotational constants are in GHz, dipole moments in Debye.
GAUSSIAN 86 (RHF/6-31G*)MOPAC (AM1)experimental
A3.338803.370323.33215*
B1.205231.197241.19307*
C1.092491.084411.08221*
m4.84.24.452(8)
* Error smaller than last digit