Effects of Electronic Resonance Interaction on the Methoxy Group NMR Parameters. Theoretical and Experimental Study of Substituted 2-Methoxypyridines

In order to increase the understanding of the interactions which define the planar conformation of the methoxy group with respect to the aryl ring in methyl aryl ethers and the effect these interactions have on the methoxy NMR parameters, 170 and I3C spectra were measured and a b initio optimized geometries were calculated for three different conformations of the methoxy group in 2-methoxy-5-X-pyridines (X = H, N02, and NH2). ''0 and 13C chemical shifts were also calculated using the LORG approach. It was found that, contrary to what has been commonly assumed, the oxygen and the carbon of the methoxy group aredeshielded when the electronic resonance interaction is increased. Therefore, the large 13C deshielding effect observed for a conformation with an out-of-plane methoxy group in aryl methyl ethers and related compounds has to be attributed to the inhibition of the attractive van der Waals forces between the methyl moiety and the aromatic o r t h w i s carbon atom.


Introduction
In a recent paper Bond and Schleyerl reported the results of an ab initio study indicating that the preferential s-cis conformation in methyl vinyl ether and related compounds originates mainly due to the electrostatic interaction between the methyl and vinyl moieties. The methoxy group in unhindered methyl aryl ethers also adopts a planar conformation2 as in these methyl vinyl ethers, suggesting that electrostatic interactions between the methyl group and the proximate ring atoms may also play an important role in defining the preferential conformation in the methyl aryl ethers. Therefore, it is expected that this electrostatic interaction would complement the resonance interaction between the a-type oxygen lone pair and the aromatic a-electron system in defining the planar conformation.
In this paper, the I7O and I3C N M R shieldings of 2-methoxypyridine (I), 2-methoxy-5-nitropyridine (11), and 2-methoxy-Saminopyridine (111) are used to estimate the importance of the resonance interaction and the electrostatic interactions in determining the N M R parameters. Compounds I1 and I11 were selected because the nitro and amino groups, placed para to the methoxy group, can interact with the methoxy group only by electronic resonance through the a-system. As the nitro and amino groups are of opposite electronic nature (i.e., electron donating versus electron withdrawing), these two compounds are adequate to assess the effect of the resonance interaction on the methoxy I7O and I3C chemical shifts and the methoxy carbon's IJ(CH) coupling constant. In previous studies I7O chemical shieldings have been found to be very sensitive to the a-resonance interaction. 3 Ab initio optimized geometries were calculated for three different conformations of the methoxy group in each compound, namely, planar cis to the N atom, A, planar trans to the N atom, B, and perpendicular to the pyridine ring, C (see Scheme I conformation. In all three cases A is expected to be the preferred conformation.5 Experimental Section NMR Measurements. Natural abundance 13C NMR spectra were recorded on a Varian FT-80A spectrometer using DzO as an external lock. The spectra were recorded at room temperature in tetrahydrofuran. 13C chemical shifts were referenced internally to TMS. Natural abundance I7O N M R spectra were recorded on a Varian XL-400 spectrometer using 5-mm sample tubes. Spectra were taken at room temperature in pyridine-d5. 170 chemical shifts were externally referenced to water at 28 OC. All substances were of commercial origin. Calculations. Geometry optimizations were carried out for compounds 1-111, for the three different conformations A-C, using the GAUSSIAN-86 molecular package. 6 In the optimization of the A and B structures the C, symmetry was enforced.
The D95 basis set was e m p l~y e d .~ Methoxy I7O and I3C magnetic shielding constants were calculated for all optimized geometries using the LORG program4 with the same basis set.

Results and Discussion
NMRMeasurements. Themethoxy I7Oand chemicalshifts and 1J(CH) coupling in 1-111 are shown in Table I. A large 170 deshielding is observed in I relative to the shielding measured in anisole (45 ppm).8 This deshielding effect can be attributed to the resonance interaction between two substituents placed ortho to one another, assuming that electronically the ring nitrogen atom of pyridine acts like a nitro group attached to a benzene ring9  Compound I1 contains an electron acceptor, the nitro group, placed para to the methoxy group, which is an electron donor, and it can be assumed that the main interaction between the two groups is a resonance interaction. The increase of the resonance effect yields a deshielding of the methoxy I7O chemical shift of 16 ppm along with a slight deshielding (1.3 ppm in tetrahydrofuran and 1.89 ppm in dimethyl-&,sulfone) of themethoxy I3Cchemical shift. In compound 111 an electron donor substituent, the amino group, is placed para to the methoxy group producing an inhibition of the electronic resonance. A shielding effect of 13 ppm is observed for the I7O chemical shift, and a slight shielding effect, 0.11 ppm, is observed for the methoxy 13C chemical shift. Therefore, it is observed that although the methoxy I7O chemical shift is notably more sensitive to the resonance interaction than the I3C chemical shift, both of them are affected in the same sense; i.e., when increasing the resonance interaction, both nuclei are shielded.
It is important to note that this trend does not support the common assumption that the deshielding effect of about 5 ppm observed for an out-of-plane conformation of a methoxy group, with respect to the planar conformation, originates in an inhibition of resonance.1w14 According to values reported in Table I, an inhibition of resonance would yield a shielding chemical shift and not a deshielding value.
According to values reported in Table I, the OMe lJ(CH) coupling also seems to be a fairly sensitive probe to this resonance interaction, with the coupling increasing as the resonance interaction increasing.
Calculations. The calculated total energies for the three conformations depicted in Scheme I are shown in Table 11. According to those values, in all the compounds studied here the preferred conformation is planar-cis, as is expected from the experimental data.s The trans-cis difference increases slightly when increasing the resonance interaction, going from 30.46 kJ/ mol in 111 to 32.17 kJ/mol in 11, indicating that the resonance interaction favors the planar-cis over the planar-trans conformation. While for compound I1 the planar-trans conformation is preferred over the perpendicular orientation, for I and 111 the opposite holds.
Other calculated molecular parameters are also affected by the resonance interaction. The bond orders and bond lengths of  effects with the former being lengthened and the latter being shortened when the resonance interaction is increased.
In Table IV the LORG calculated chemical shieldings are compared with the experimental values. It is observed that the LORG method with D95 basis set systematically underestimates the chemical shieldings by about 40 ppm but successfully reproduces the substituent effects when the experimental values are compared with those calculated for the A conformations. The I3C chemical shieldings are overestimated by about 10 ppm, but again the qualitative trends are well reproduced.
The theoretical calculations confirm the experimental trend that resonance interactions affect both nuclei in the same sense; Le., when the resonance interaction is increased both nuclei are deshielded, and when there is an inhibition of the resonance interaction both nuclei are shielded.
For the B conformations the I7O SCS are essentially the same as those obtained for the A conformations. As 170 chemical shifts are quite sensitive to resonance interaction^,^ this indicates that the resonance interaction between the methoxy and nitro or amino groups is essentially the same for both planar conformations.
It is important to observe that for the methoxy out-of-plane conformation, C, the SCS effects are smaller than those observed for the planar forms, but they are still significant. This is a clear indication that even for an out-of-plane conformation the methoxy group is undergoing a rather strong resonance interaction with the ?r electronic system. This result is in agreement with recent experimental results on sterically crowded methoxy groups in substituted anis01es.l~ The calculated I3C shieldings for the C conformations show the rather strong I3C deshielding effect of about 6 ppmpreviously observedin other aromaticcompounds.11J6

Conclusions
Both experimental and theoretically calculated I7O and 13C chemical shifts indicate that in compounds 1-111 the increase in the resonance interaction produces a deshielding effect on both the carbon and oxygen of nuclei of the methoxy group. Moreover, according to the calculations this holds for both an in-plane and an out-of-plane methoxy group conformation. As a similar behavior can be expected in other methyl aryl ethers, the wellknown deshielding effect of the methoxy I3C chemical shift for an out-of-plane conformation cannot originate in an inhibition of resonance as it has been commonly assumed.I*l3 In substituted anisoles the conformational effect of the methoxy group on the chemical shift of the aromatic ortho carbon atom placed cis to the methyl moiety is a shielding effect of approximately 8 ppm;i6J7 therefore, it can beconcluded that the proximity between the methyl moiety and the aromatic ring yields a shielding effect on both 13C nuclei. According to Li and Chestnut,18J9 when two atoms of the same molecule are in close proximity to one another, a shielding effect is observed on both if the van der Waals forces yield an attractive interaction, while a deshielding effect is observed if there is a repulsive interaction. These considerations suggest that the observed shielding effect is due to an attractive interaction.
These attractive interactions are similar in nature to those described as electrostatic interactions by Bond and Schleyerl when they analyzed the interactions that define the preferential s-cis conformation in vinyl methyl ether and related compounds. Therefore, it can be concluded that in methyl aryl ethers two different types of interactions cooperate to definea planar methoxy conformation, namely, the conjugative interaction between the oxygen r-type lone pair and the aromatic r electronic system and the electrostatic attraction between the methyl moiety and the aromatic ortho-cis carbon atom.