Gri © Ata Cenc Semi ACTA CCHEMICA SEANDINAVICA Microwave Spectrum, Conformation, Barrier to Internal Rotation, “N Quadrupole Coup g Constants, Dipole Moment and Quantum Chemical Calculations for Methyl Carbamate K.-M. Marstokk and Harald Mgllendal* Department of Chemistry, The University of Oslo, PO Box 1033 Blindern, N-0315 Oslo, Norway Marstokk, KM. and Mollendal, H., 1999. Microwave Spectrum, Conformation, Barrier 10 Internal. Rotation, N’ Quadrupole Coupling. Constants, Dipole Moment and Quantum Chemical Calculations for Methyl Carbamate. ~ Acta ‘Chem. Seand, 33° 79-84, © Acta Chemica Scandinavica 1999. ‘The microwave spectrum of methyl carbamate (H,NCO;CH) has been investigated in the 13.5-40.0 GHz spectral region at room temperature. One conformer was assigned, This rotamer has a symmetry plane (C, symmetty) and two outofplane hydrogen atoms. The methyl and the carbonyl groups are in the syn conformation. The batter to internal rotation of the methyl group is 4235.9(66)} mol". The dipole moment is (in units of 10" Cm) 1,=0.544(7), Wy=7.653(31), n= 0 (for symmetry reasons) and tyy=7.672(31). The “N Quadrupole coupling constants were Tound 10 be Yu=1-52(27) and yy S518) Miz "The microwave work has been assisted by ab initio computations at the HF) 63114 4G", MP2/6311++G** (Frozen core) levels of theory, as well as sensity theory cae ‘The carbamate group, -HNCO,-, appears in many biol- cogically active compounds'™ such as local anaesthetics, muscle relaxants, anticonvulsants, anti-cancer agents, antiepileptics, sedatives and hypnotics as well as industri- ally important polyurethane polymers.* Methyl carbamate is the prototype molecule of this class of compounds. Its structure and conformational properties have been studied by several physical and theoretical methods such as infrared spectroscopy in solution’ and in the gas phase,® X-ray erystallography,’ as well as semi-empirical and ab initio quantum chemical methods.” These studies all agree that the Syn con- formation (Fig. 1), which has the methyl and carbonyl ‘groups sym to one another, is the preferred form of the molecule in all phases. Quantum chemical calcula- tions**- predict that the same conformation of the free molecule is several kJ mol-! more stable than any other rotamer. ‘The non-hydrogen atoms of methyl carbamate are nearly planar in the crystal, where the molecules are hydrogen-bonded into planar layers.” The situation is of course different in the gas phase, where hydrogen bond- ing is absent. It has been claimed® that the nitrogen atom To whom correspondence should be addressed, Ecmal: haralé.mollendal@kjemi.uio.no (© Act Chemica Seandinavn 9 1899). 79-04 jons at the B3LYP/6-31G* level Ae rs sy Ant Fig. 1. The Syn and Anti conformations of methyl carbe- mate. Syn was assigned in this work. MP2/6-311+.1.G"* Caleviations predict Anti to be 35.5 kJ mol less stable than Syn. has a shallow pyramidal equilibrium conformation in the free state, The lowest energy level was found to lie above the barrier opposing planarity. No microwave (MW) studies have previously been reported for the ttle compound. Since MW spectroscopy can provide accurate information about geometrical structures, barriers to rotation of methyl groups, dipole ‘moments and nuclear quadrupole coupling constants, it ‘was decided to carry out an investigation for this import= ant prototype compound. The experimental work has been assisted by quantum chemical calculations made ”MARSTOKK AND MOLLENDAL at different andjor higher levels than previously reported. Experimental The sample (m.p. 56-58 °C specified to be at least 98% pure) utilised in this work was purchased from Aldrich and used as received. No impurities were seen in the MW spectrum. The MW spectrum was studied using the Oslo spectrometer which is described in Ref. 11. The 13.5-40 GHz spectral region was investigated with the microwave absorption X-band brass cell held at room ‘temperature. The pressure was about $Pa when the spectra were recorded and stored electronically using the computer programs written by Waal.!? The accuracy of the frequency measurements is presumed to be better than 0.10 MHz, Results and discussion Quantum chemical calculations. The Gaussian 94 program package"? running on the IBM RS6000 cluster in Oslo ‘was employed in all the quantum chemical calculations, Computations were made for the Syn conformer (Fig. 1) at three different levels of theory. Hartree-Fock (HF) and Moller-Plesset (MP2) perturbation calculations with frozen-core electrons! were made using the 6- 3114+G** basis set, In addition, density functional theory computations employing the B3LYP method'® ‘were carried out using the 6-31G* basis set. Calculations for Anti were made only at the highest level (MP2/ 6311+ +G**). Full geometry optimisation was made in each case with the results shown in Table | (atom num- bering in Fig. 1), which also includes selected bond lengths and bond angles taken from the X-ray structure” of the crystalline phase for comparison. Vibrational frequencies (not included in Table 1) were computed in each case. No imaginary frequencies were found for any of the two conformers, which are thus assumed to be true minima on the potential energy hypersurface."® Comments are warranted for some of the results in this table. The MP2 and the B3LYP structures of Syn are remarkably similar. The bond lengths are shorter in the case of the HF computations than in the other two ‘cases, presumably because these calculations do not take lectron correlation into consideration as the MP2 and BILYP methods do. The bond lengths of the crystalline phase are similar to the theoretical predictions with the exception of the CINI bond length, which is shorter in the crystal. The difference is presumably caused by hhydrogen bonding that is present here. ‘The non-hydrogen atoms are computed to be nearly co-planar in Sym, as can be seen from the O1CLO2C2 and C202CINI dihedral angles. The amide group is calculated to be non-planar at all three levels of theory, as seen from the OICINIH4 and OICINIHS dihedral angles, which deviate 10-20" from the values they would have had in the case of a completely planar amide group. 80 ‘The pyramid around the nitrogen atom is calculated to be more shallow in the HF than in the MP2 or B3LYP calculations. The non-planarity computed for the amide group manifests itself in the value of [,+/)~f,. This value is typically 3.20% 10°? um? for a molecule possessing a symmetry plane and two out-of-plane hydrogen atoms attached to a sp? hybridised carbon atom. An increase in this value is characteristic for a compound that does not possess such a plane. The MP2 structure of Sym has the most non-planar amide group and the largest value of et Iy~ I, (3.467 x 10-* um; Table 1). The non-hydrogen atoms of Anti are predicted (Table 1) to be slightly non-planar, since the O1C102C2 and C202CINI dihedral angles are calculated to be 171.9 and 168.2, respectively. A non-planar amide group is computed for this conformation as well. The Anti conformer is calculated (see entries in Table 1) to be as much as 35.5 k mol” less stable than Sym at the MP2/6- 311++G** level MV spectrum and assignment of Syn. The MW spectrum of methyl carbamate is rather dense and comparatively strong. Theory predicts that all transitions should be split into two components of equal intensity owing to tunnelling of the methyl group. One of these components. hhas the non-degenerate 4 symmetry whereas the other hhas the degenerate £ symmetry. The quadrupole inter- action of the "*N nucleus with the molecular rotation should lead to additional splittings of both the A and the E species lines. Survey spectra revealed many transitions with charac- teristic splitting patterns. In addition, several strong lines with resolved Stark effect were noticed. ‘The quantum chemical computations above and previ- ‘ous work®""” indicate that Syn is the preferred form of the molecule, The largest dipole moment component of this rotamer is 4, (Table 1), Searches were first made for the strong *Q-branch transitions using the rotational constants obtained in the MP2 computations as the starting point. These transitions were soon identified close to their predicted frequencies. The b-type R-branch transitions were identified next. Spectroscopic constants and barrier to internal rotation of the methyl group. The assigned lines with J<7 were selected for fitting to Herschbach’s ‘principal axes" Hamiltonian'” using the computer program described in detail in Ref. 18. The rotational constants, the barrier to internal rotation of the methyl group, Vs, the direction cosines of the symmetry axis of the methyl group with respect to the principal inertial axes, 2, by, , and the ‘moment of inertia of the methyl group around its sym- metry axis, J,, are the parameters of this Hamiltonian. Centrifugal distortion and nuclear quadrupole inter action is not taken into account in this Hamiltonian, ‘The $4 lines with J+<7 used for fitting are shown in Table 2. Centrifugal distortion is assumed to be of litleTable 1. Structure® of methyl carbamate at various levels of theory and X-ray structure.” [MW SPECTRUM OF METHYL CARBAMATE Conformer: Syn Method: MP2/6-3114+G** HF/6311++G"* BSLYP/GS1G* Xray MP2/63114.46"" Bond lengthypm c1o1 1213 89 15 12272) 1207 croz 1356 132.4 1360 133.82) 1359 o2c2 143.4 1416 1433 144.03) 127 cout 109.1 108.1 1093 1093 cane 109.1 108.1 1093 1095 C2H3 1088 1079 109.1 1089 cm 1374 1352 1370 133.412) 1388 Nia 1008 99.2 1009 1010 IHS 1008 99.1 1003 1008 Bond angles,” orci02 1286 1283 1247 122812) 1203 c102c2 1137 179 145115211) 1195 o2caHt 108 nos 08, m3 02c2H2 1105 nos nos. m5 0262H3 1053, 1079 1055 1055 o1cini 1256 1248 1254 124.2(2) 1238 CINTHS 143, nT 1159 ma CINTHS 1790 1197 1187 1187 Dihedral angles)” orcr02ce Ww os 08 m9 c102c2H1 608 608 60.8 535, Cro2c2H2 605, 605, 608 700 c102c2H3 1798 1798 -1798 1723 c2o2cint 1763 1784 15 168.2 O1cINIHa 196 107 160 148 O1CINTHS 1621 169.4 1685 1509 Total energy/kJ mol"? —745096.4 7428836 1467903 =7450609 Rotational constants/MH2 A 10625 11129 10608 10136 8 4407 4438 4372 4456 c 3183 3238 3161 3171 Tyla 1gN0°% um? 3.487 3.208 3.353 3.95 Dipole moment4/10°% Cm Me 0.16 oz 1.03 12.16 Me 7.88 42 Tae 1401 Me 281 162 243 273 ot 837 858 789 1875 *See Fig. 1 for atom numbering. "Taken from Ref. 7. “Tyr ly and Ie are the principal moments of inertia, Conversion factor: 1505379.05 um? MHz. 1 debye= 3.33564 x 10-*° Cm, importance for these transitions. It was also assumed that methyl carbamate has a symmetry plane (see below). 2, is then the only independent variable direction cosine. The moment of inertia of the methyl group around its symmetry axis, J,, was kept constant at 3.20 x 10°? um? in the fit while the other parameters of this Hamiltonian were allowed to vary. The result of the least-squares fit is shown in Table 3. The A-species transitions depend only on the even- order terms of the angular momentum operator,"” and ray thus be fitted to Watson’s Hamiltonian."° The result of such a fit of 92 selected transitions with comparatively small “'N quadrupole splittings is seen in Table 4." In this fit the maximum value of the J quantum number was 33. It was necessary to vary one of the sextic centrifugal distortion constants (@,) in order to get a fit with a root-mean-square deviation comparable to the experimental uncertainty of $0.10 MHz. ‘The A-species spectroscopic constants in Table 4 "The spectrum of the 4 species is available from the authors ‘upon request, or from the Molecular Spectra Data Center, National Institute for Standards and Technology, Molecular Physics Division, Bldg. 211, Room B265, Gaithersburg, MD 20899, USA, where it has boon deposited 81MARSTOKK AND MOLLENDAL Table 2. MW spectrum of the ground vibrational stato of methy! carbamate. Table 4. Spectroscopic constants of the ground vibrational state (A species) of meth! carbamate, Observed" MHz Transition a alent tien ie eiy Ya ware YAM 2 thy (3871656 3502 3662 eo iy 191198.71 3399 34.32 Bs Ap 168631355683 33 2 26080816252 ai 5317715558567 By Ba 1802180343 3.35 ai By 24gt222 401-388 2 317275831105 11.08 aS a 27118402277 22.98 fos ke 5261849270 2.48 Bra dey 2506340 19.12 1908 5c See 1956736 6236.34 bs + Bue 1730955 «1184 1202 Ba Bre -3037889 2096 20.89 fe See = «3980983 131-127 Sis Brg «3145881 2540 28.77 Be Bea 35010 1530 14.71 Bis Bra «35817828399 83.75 Gs Oe 54741 7607.48 Gs + Ge 18458301086 1097, Ss Ge | 427769 2088 ©2058 G+ Be 2953340865897 Gu Ss 37182704868 48.95, Tes = Bre 2850705 51.18 51.13 Te = Tey 3214847933 9.48 eS tay 2765868 19291952 Tes + Tha 20962-73910 9.23 +£0.10 Miz, Table 3. Molecular constants* ofthe ground vibrational state (of methiy! carbamate Rotational constants? MHz Ag=10715.151(25) 8)=4398,9521183) Cy =3182.851(10) Principal-axes moments of inertia")10-#? um? 1g=47.16890(11) [y= 114.88628124) 1 Tar lyle=3.269 28153) Moment of intetia of the methyl group around its symmetry axis/10- um? 1,=3.20¢ Direction cosine of the methyl group symmetry 2,=0.9110135) Barrier to internal rotation of the methyl group/J mol”* Vs=4236.9166) 158.781 86146) “Uncertainties represent one standard deviation. °As defined bby Herschbach."” “Conversion factor 50537805 um* MHz “Assumed; see text should predict the frequencies of a- and c-type A-species transitions accurately, Such transitions were searched for ‘but not found, presumably because they are very weak. This is evidence that the components of the dipole moment along the a- and c-inertial axes are small, or perhaps zero in the case of 4, (see below), because the 2 No. of transitions: 92 Fums. dev./MMz 0.102 AyiNHe 10719.383 6192), By/MH2 4399.149 4/32), cayMHe 3182.8696(35), Age? 0.723113) Aigikie 43011) aokHe 8.806462) ayikie 0.221 9196), 287118) 0.0845115) 3.246 42112) *Areduction, I“representation.” Nuclear quadrupole inter- action has been neglected. "Uncertainties represent one Standard deviation. “Root mean-square deviation. *Further sextic. centrifugal distortion constants pre-set at 22r0; see text. *Same conversion factor as in Table 7 intensities are proportional to the square of the dipole ‘moment component. Interestingly. the rotational constants (Table 3) agree to within better than 1% with those calculated in the MP2 procedure (Table 1). Moreover, 2, calculated from the MP2 structure in Table ! is 0.909, close to the experimental value 0.9110(35) given in’ Table 3. It is assumed that the good agreement seen for the rotational constants as well as for Ay is not fortuitous, but reflects the fact that MP2/6-311++G** structures in general are quite accurate, as has already been noted.*° ‘The barrier of 4235,9(66) J mol ' is compared to the barriers of other substituted methyl formates, eg, 4769(12) in methyl formate,?* $081(371) in methyl acet- ate," 4447(24) in fluoroformate,” 4866(12) in methyl ‘eyanoformate,** $297(126) in methyl propiolate** and 5105(126)3 mol" in methyl acrylate.* The somewhat smaller barrier seen for the ttle compound is perhaps a result of x clectron delocalisation in the amide moiety Dipole moment and planarity of the amide group. A non- planar conformation of the amide group would produce 4 component of the dipole moment along the principal inertial canis (y.) that would be non-zero, The theoretical structures in Table | indicate that even a small deviation from planarity produces a significant y, for Syn. The dipole moment is in fact a rather sensitive test of planarity The 15 Stark coefficients shown in Table 5 of -species transitions with comparatively small quadrupole split {ings were used to determine the dipole moment following the standard procedure. Field strengths were typically in the 500-1000 V.em~ range. Quadrupole interaction is small and was not taken into account in the least- squares procedure. Initially, all three principal axes dipole moment components were fitted. However, a small and ‘imaginary value was found for tH, (H.)?= 0.012]. This shows that this dipole moment component must be zeroTable 6. Stark coefficients and dipole moment of methyl carbamate.” ‘av E-#/10-® M2 V Transition? (M| Obs. Cale. hee has #18202) 148 3 0.4886) ote Bes Ba 5 2.8913) 273) 4 1842) 178 3 1081) 1.00 Ss Be 8 0.990110) 1.03, 7 073718) 0772 6 052715) 0543 5 034014) 0.355 Se ey 91.8841) 168 8 1301) 130 7 0:369(70) 0.387 6 07218) one foe = ha 1084818) ose af Ga #1212) 121 Dipote moment*/10- Cm Hy=0.54417) —y=7.653(31) p,=0.0% pe =7.672(81) “Uncertainties represent ono standard deviation. °A species ©T debye~ 3.335 64 x 10" Cm, “Pre-set at 2er0; S00 text. or very close to zero. This is a strong indication that the amide group is indeed planar, or very nearly so, and that the molecule has a true symmetry plane (C, symmetry). In the final fit yw, was pre-set at zero yielding the results shown in Table 5. The total dipole moments obtained in the three theoretical computations (Table 1) are all close to the experimental result Another evidence that methyl carbamate has a sym- metry plane comes from the value of I+ /y—/e= 3.269 28(53) x 10-*° um? (Table3). If the contribution from the out-of-plane hydrogen atoms of the methyl group is arbitrarily assumed to be (same units) 3.20, a pseudo-inertial defect of 0.069 x 102° um? is obtained. For a completely rigid molecule this value would be exactly zero. The corresponding values are: ~0,006510(2) for formamide,”” —0.122 for acet- amide, 0.078 for FCH,CONH3.” 0.184 for CH,OCH,CONH,,” 0.314 for propionamide” and 0.156 for acetidinone.”* all of which are assumed to have planar or very nearly planar, amide groups. It can be concluded from the preceding that the MP2 structure of Table | is likely to be very accurate with the exception of the geometry around the nitrogen atom, which is more shallow than the MP2 computations indicate. This shortcoming of the MP2 procedure in dealing with amide group nitrogen atoms has been discussed before. "4N quadrupole coupling constants. All transitions were split by @ small amout by quadrupole interaction, Well resolved quadrupole components were only observed in a few cases. The well-esolved A-species components appearing in Table 6 were used to determine the quadru- [MW SPECTRUM OF METHYL CARBAMATE Table 6. “N nuclear quadrupole splitings (Eq) of methyl carbamate and diagonal elements of the “*N’ quadrupole coupling tens EMH Teansition® " e F Obs Cale Bae 5 oe 8-167 176 ae 6 5 6 17 175 tes 75 7 21738 165 Bye 305 3 047 OMT Ba 35 3 142-140 Be + 4 5 4 050 04s ne 45 4-130 -1.32 aS 5 oF 8 128-427 eo 6 © 6 129 128 % 8 5 8 1200-41 te 30 5 2 029 036 te 2 5 4 137-146 ae 3 5 2 074 ope NV nuclear quadrupole coupling constants®/MHz 5227) __xy=3.51120) “A species. Nuclear quadrupole interaction has been neg: lected. "Uncertainties represent one standard deviation, pole coupling constants of the "*N nucleus following the procedure of Ref. 26. The spectroscopic constants in Table 4 were used to predict the “unperturbed” centre frequencies. The “N nuclear quadrupole splittings (E, in ‘Table 6) were found by subtracting the calculated centre frequencies from the frequencies of the resolved quadru- pole components. A rigid-rotor Hamiltonian was used in the least squares fitting procedure in order to determine the 'N nuclear quadrupole coupling constants 7, and You- The results are shown in Table 6. Conclusions ‘The Syn conformer of methyl carbamate is stabilised by several kJ mol”! relative to any other rotamer. The amide moiety is planar (or very nearly so) and the molecule has C, symmetry. Ab initio calculations at the HF and MP2 levels as well as density theory computa- tions at the B3LYP level fail to predict a planar structure for the amide group. The MP2 and B3LYP methods predict structures that are believed to be accurate with the exception of the hydrogen atoms of the amide group. The calculated total dipole moments at all three levels of theory are close to the experimental one. The barrier to internal rotation methyl group is about 10-20% less than found in other substituted methyl formates. Acknowledgement. Anne Horn is thanked for the art work and for assistance. This work has received support from The Research Council of Norway (Programme for Supercomputing) through a grant of computer time. The authors are grateful to one of the referees for a thorough reading of the manuscript, which helped us to remove several obscurities and misprints. 83MARSTOKK AND MOLLENDAL References 1 2 3 10. 2, B, 14 84 Wolll, M.E., Fd, Burger's Medicinal Chemistry, 4th Edn, Parts Hand IIL, Wiley, New York 1979, 1981 O'Brien, In: Ariens, E. J. Drug Design, Vol It, Academic Press, London 1971, pp. 162-212. Hartley, G. S. In: Drug Design, Vol. VI, Academic Press, London 1975, pp. 298-381 David, D. J. and Staley, H. B. In: Mark, H., Flory, C. S., Marl, C.S., Melville, H.W. (Eds) High Polymers, Vol XVI, Part 3, Wiley-Interscience, New York 1969, ‘Oki, M. and Nakanishi, H. Bull Chom Soc: Jpn 44 agin) 3148, Kydd, R. A. and Rauk, A. J. Mol Struct. 77 (1981) 227. Sepehrning, B., Ruble, J. R, and Jelfey, G. A. deta Crystallogr’ C43 (1987) 249. Manning, J., Klimkowski, V. J., Siam, K., Ewbank, J. D. land Schafer, LJ. Mol Struct. (Theochem) 139 (1986) 305, Remko, M. and Scheiner, 8. J. Mol. Struct. (Theochem) 204 (1990) 331 Bandekar, J, and Okurumi, Y. J. Mol. Struct. (Theochem) 281 (1993) 113, Guirgis, G. A. Marstokk, KM. and Mellendal, H. Acta Chem. Scand. 45 (1991) 482, Waal. O., Personal communication 199. Frisch, M. J. Trucks, G. W., Schlegel, H. B.,Gill,P. MW. Johnson, B. G., Robb, M. A., Cheeseman, J. R., Keith, T Petersson. G. A.. Montgomery. J. A.. Raphavachari. K. AbLakam, M.A. Zakrzewski, VG. Ortiz, J. V, Foresman, J. B., Peng, C. Y., Ayala, B. Y., Chen, W., Wong, M.W., Andres, . L, Replogle, E.S., Gomperts, R Martin, RL, Fox, D. J.” Binkley, J. 8, Defrees, D. J, Baker, J., Stewart, J. P,, Head-Gordon, M., Gonzalez, C. and Pople, J. A. Gaussian 94, Revision B.3, Gaussian Inc, Pitisburgh, PA 1995, Moller, C. and Plesset, M. S. Phys. Rev. 46 (1934) 618, Becke,’A.D. J. Chem: Phys. 98 (1993) S648. 16, Hehre, W, J., Radom, L., Scbleyer, P. v. R. and Pople, J.A, Ab Initio Molecuiar Orbital Theory, Wiley, New York 1985, p. 227 1, Herschbach, D. R. J. Chem. Phys. 31 (1959) 91 18. Marstokk, K-M. and Mollendal, H. J. Mol Suet. 20 (3974) 257, 19, Watson, 1. K. G. In; Durig, J. R., Ed, Vibrational Spectra and Structure, Elsevier, Amsterdam 1977, Vol. 6, p. 20, Matstokk, KM, and Mollendal, H. Aeta Chem. Scand. $2 (1998) 296, 21. Demaison, J., Boucher, D., Dubrulle, A. and van Eijck, BP. J. Mol. Spectrase. 102 (1983) 260, 22. Sheridan, J, Bossert, W. and Bauder, A. J. Mol Spectrosc. 80 (1980) 1 23. Groner, P., Tolley, C. L, and Durig, J. R. J. Mol Struct. 223 (1990) 471, 24. Durig, JR, Groner, P. Lin, J. and van der Veken, B. J J. Chem. Phys. 96 (1992) 8062 25. Williams, G., Owen, N. L-and Sheridan, J. Tans. Faraday Soc. 67 (1971) 922. 26. Marstokk, KM. and Mollendal, H. Acta Chem. Scand, Ser. A 36 (1982) 517, 27. Hirota, E., Sugisaki, R., Nielsen, C. J. and Sorensen, G. O. J. Mol. Spectrosc, 49 (1974) 231 28. Koijima, T., Yano, F., Nakagawa, K. and Tsunekawa, 8. J. Mol. Spectrosc, 122 (1987) 408, 29. Marstokk, K-M. and Mollendal (1974) 283, 30, Marstokk, KM. Mollendal, H. and Samdal, S. Acta Chem. Scand. 50 (1996) 845, 31. Marstokk, K-M., Mallendal, H. and Samdal, $. J. Mol Struct. 376 (1996) 11 32, Marstokk, KM. “Mollendal, H., Samdal, S. Ugecrud, E. Acta Chem. Scand. 43 (1989) 351 33, Samdal, 8.1 Mol. Soruct 440 (1998) 165, HJ. Mol. Struc 22 and Received July 3, 1998