Abstract
Substituents on the pyridinium ring of N-methylpyridinium derivatives, especially those on the 2- or 4-positions, have a large effect on the 1H and 13C NMR chemical shifts of the N-methyl group. Reasonable correlations between the chemical shift changes and the resonance substituent constants are observed. The dual substituent parameter approach provides an excellent correlation when a combination of polar and resonance substituent constants is employed.
The determination of the electronic effect that substituents have on reactions is very important in the elucidation of reaction mechanisms, especially the nature of the transition state.1 The Hammett constants, σmeta and σpara, were first calculated from the comparison of ionization constants of substituted benzoic acids.2 Two new types of σ constants, σ+ and σ−, were introduced for structures in which the substituent is able to come to direct resonance interaction with electron-deficient or electron-rich reaction sites, respectively.3,4 Later, the separation of polar and resonance parts of the substituent effect has led to the establishment of the polar substituent constant, σI, and four sets of resonance substituent constants, σ°R, σR(BA),σ− R and σ+ R.1,5 The four sets of resonance parameters are used for relatively unperturbed systems, substituted benzoic acids, electron-deficient and electron-rich benzene rings, respectively.1,5 In this report, we will describe the excellent correlation between the 1H and 13C NMR chemical shifts of substituted N-methylpyridinium and the substituent constants σI and σ+ R using the dual substitutent parameter approach (eq. 1).
(1) |
We have recently examined the acidity of the α–CH group of 2-methoxy and 4-methoxypyridinium compounds through kinetic measurement of hydrogen-deuterium exchange.6 The pKa values of these two compounds in aqueous solution were found to be surprisingly high and a large solvent effect was observed for the hydrogen-deuterium exchange reactions. In our effort to investigate the effect of various substitutents on the rate of hydrogen-deuterium exchange, we have found that these groups, and other π-electron donating groups, on the 2- or 4- positions have a large effect on the chemical shifts of the N-methyl group in pyridinium compounds. The upfield chemical shifts are a result of the lower charge density on the nitrogen atom due to the resonance structures as shown in If a good correlation were to be found between the chemical shift of the N-methyl group and the ability of substituents to donate electrons, it would be a valuable tool in the measurement of the resonance effects of substituents due to the ready availability of the N-methylpyridinium compounds. Therefore, we have prepared a series of substituted N-methylpyridinium molecules to investigate the relationship between the chemical shifts of the N-methyl group and the nature of the substituents.7 The 1H and 13C NMR chemical shifts and the relevant substituent constants are listed in Table 1.
Table 1.
Polar and Resonance Substituent Constantsa and Corresponding 1H and 13C NMR chemical shifts of the N-methyl group in substituted pyridinium compounds
Substituent | σm | σI | σ°R | σR(BA) | σ+R | δ (1H, ppm) | δ (13C, ppm) |
---|---|---|---|---|---|---|---|
−H | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 4.426 | 51.043 |
−NH2 | −.16 | .12 | −.48 | −.82 | −1.61 | 3.834 (2-) | 44.055 (2-) |
4.243 (3-) | 50.813 (3-) | ||||||
3.936 (4-) | 47.552 (4-) | ||||||
−NHAc | — | .26 | −.25 | −.36 | −.86 | 4.243 (2-) | 47.079 (2-) |
4.235 (4-) | 49.445 (4-) | ||||||
−OCH3 | .12 | .27 | −.34 | −.61 | −1.02 | 4.276 (2-) | 44.153 (2-) |
4.390 (3-) | 51.230 (3-) | ||||||
4.182 (4-) | 49.071 (4-) | ||||||
−CH3 | −.07 | −.04 | −.11 | −.11 | −25 | 4.258 (2-) | 48.419 (2-) |
4.366 (3-) | 50.747 (3-) | ||||||
4.330 (4-) | 50.153 (4-) | ||||||
−Cl | .37 | .46 | −.23 | −.23 | −.36 | 4.415 (2-) | 50.390 (2-) |
4.431 (3-) | 51.357 (3-) | ||||||
−Br | .39 | .44 | −.19 | −.19 | −.30 | 4.415 (3-) | 51.204 (3-) |
−CN | .56 | .56 | .13 | .13 | .13 | 4.636 (2-) | 51.457 (2-) |
4.511 (3-) | 51.982 (3-) | ||||||
4.521 (4-) | 52.206 (4-) | ||||||
−CO2CH3 | .37 | .30b | .14 | .14 | .14 | 4.486 (4-) | 51.608 (4-) |
−COCH3 | .38 | .28 | .16 | .16 | .16 | 4.517 (3-) | 51.478 (3-) |
4.501 (4-) | 51.416 (4-) |
N-Methylpyridinium compounds substituted at the 2- or 4- positions were treated separately because the chemical shifts were quite different for the two sets of molecules. A better fit was found for 4-substituted pyridinium compounds. This result was reasonable considering the fact that the substituent constants were derived with para-substituted benzene derivatives. When Δδ (the differences between the substituted and parent pyridinium molecules) were plotted against resonance substituent constants alone, σ° R, σR(BA) and σ+ R gave reasonable fits. The poor correlation with σ− R was fully expected because of the electron-deficient nature of the pyridinium ring.
One interesting observation is the relative electron-withdrawing ability of the cyano (−CN) and the ester group (−CO2CH3). The cyano group appears to be less electron-withdrawing than the ester group, based on the comparison of their resonance substituent constants. However, it is apparent from the reactivity of benzyl bromide derivatives in their oxidation to benzaldehyde derivatives that a p-CN substituent is more electron-withdrawing than a p-CO2CH3 group.8
One explanation is the strong polar effect associated with the cyano group, as evident from the larger σI value. When the dual substituent parameter approach was applied to the data, an excellent correlation (R2 = 0.99) was observed between the chemical shift changes and the sum of polar and resonance effects (ρIσI+ρRσR) when σ+ R was used as shown in Figure 2. The ratio of the coefficients ρI and σR (as indicated by λ = ρR/ρI) signified the larger contribution from the resonance effect. A good correlation (R2 = 0.96) was also observed when σR(BA) was employed.9 The better fit with σ+ R is expected because of the electron deficient nature of the system studied here.
Figure 2.
Correlation of 13C (A) and 1H (B) NMR chemical shifts of the N-methyl group in substituted N-methylpyridiniums with the polar and resonance substituent constants (σI and σ+ R)
For the 2-substituted pyridinium compounds, the correlation was not as good when the chemical shift changes were plotted against the resonance substituent constants alone or a combination of polar and resonance substituent constants,ρIσI+ρRσR. This kind of poor correlation has been regularly observed with ortho-substituted benzene systems and was attributed to the proximity of the substituents to the reaction site and thus contributions from the proximity effect.5
For the 3-substituted pyridinium compounds, no good correlation was found between the Δδ and σmeta or σI, although a reasonably good correlation has been reported between the 19F NMR chemical shifts of fluorobenzene derivatives and σI.10 When the dual substitutent approach was employed using σI and σ° R, a good correlation (R2 = 0.94) was found for 13C but not 1H NMR chemical shifts.
The coefficients ρI, ρR and λ are 1.37, 0.74, 0.54 for 13C NMR, respectively.11 These numbers indicate the greater contribution from the polar effect, as expected. However, the contribution from resonance effect does play a role, presumably by reducing the overall charge of the pyridinium ring.
In summary, reasonable correlations between the 1H and 13C NMR chemical shifts of the N-methyl group on pyridinium compounds and the resonance substituent constants has been observed. The best fit was found with 4-substituted pyridinium molecules. This could be a useful way to assess the resonance electronic effects of various substituents. When both the polar substituent constant (σI) and the resonance substituent constants (σ+ R) were considered, excellent correlations were observed. The results have further demonstrated the advantage of the dual substituent parameter approach.1,5
Figure 1.
N-Methylpyridinium Derivatives and Representative Resonance Structures with Electron-Donating Groups
Acknowledgment
This investigation was supported by the National Institutes of Health, MBRS SCORE Program – Grant #5 S06 GM52588. We thank Wee Tam for obtaining the NMR spectra. The NMR facility was funded by the National Science Foundation (DUE-9451624 and DBI 0521342). We also thank Professors James Keeffe and Ihsan Erden at SFSU for helpful discussions. In addition, we are indebted to Joan Wu for her assistance in the statistical analysis of the data.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References and Footnotes
- 1.Lowry TH, Richardson KS. Mechanism and Theory in Organic Chemistry. 3rd ed. New York: Harper & Row; 1987. Chapter 2, Section 2.2. [Google Scholar]
- 2.Hammett LP. J. Am. Chem. Soc. 1937;59:96–103. [Google Scholar]
- 3.Ritchie CD, Sager WF. Prog. Phys. Org. Chem. 1964;2:323–400. [Google Scholar]
- 4.Stock LM, Brown HC. Adv. Phys. Org. Chem. 1963;1:35–154. [Google Scholar]
- 5.Ehrensen S, Brownlee RTC, Taft RW. Prog. Phys. Org. Chem. 1973;10:1–80. [Google Scholar]
- 6.Wong FM, Capule CC, Chen DX, Gronert S, Wu W. Org. Lett. 2008;10:2757–2760. doi: 10.1021/ol800892d. [DOI] [PubMed] [Google Scholar]
- 7.Typical experimental procedures: Pyridine (1 g) was dissolved in 5 mL anhydrous ethyl ether in a pressure tube. Methyl iodide (3 g) was added to the solution and the mixture was stirred overnight at room temperature. The resulting solid was filtered, washed with anhydrous ethyl ether, and dried to give N-methylpyridinium iodide as a white solid. In cases when the pyridine derivatives are not soluable enough in ether, ethyl acetate was used as the reaction solvent. All chemicals are commercially available and are used without further purification. NMR spectra were obtained with either QE 300 or Bruker 500 instruments. All chemical sifts are referenced to sodium 3-trimethylsilylpropionate at 0.00 ppm.
- 8.Chen DX, Ho CM, Wu QYR, Wu PR, Wong FM, Wu W. Tetrahedron Lett. 2008;49:4147–4148. doi: 10.1016/j.tetlet.2008.04.124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.The dual substituent parameter approach using σR may underestimated the contribution from the polar effect in this case, as evident from the coefficients. The coefficients ρI, ρR and λ are 0.61, 4.12, 6.78 for 13C NMR and 0.04, 0.54, 13.62 for 1H NMR, respectively.
- 10.Taft RW, Price E, Fox IR, Lewis IC, Anderson KK, Davis GT. J. Am. Chem. Soc. 1963;85:709–724. [Google Scholar]
- 11.When σR was used, a reasonable correlation was also found for 13C NMR chemical shifts only (R2 = 0.92). The coefficients ρI, ρR and λ are 1.37, 0.49, 0.36 for 13C NMR, respectively.