Resonance substituent parameters

The absolute rate constants for reaction of the carbon-substituted silenes 2b, 6, 7, 23 and 85a-c are plotted against the resonance substituent parameter o 86 in Figure 10. The data appear to correlate reasonably well, if one assumes that the curvature in the plot results from the fact that MeOH for the more reactive derivatives in the series approaches... 

A plot against Hammett s cr-constants of the logarithms of the rate constants for the solvolysis of a series of Mz-substituted dimethylphenylcarbinyl chlorides, in which compounds direct resonance interaction with the substituent is not possible, yielded a reasonably straight line and gave a value for the reaction constant (p) of — 4 54. Using this value of the reaction constant, and with the data for the rates of solvolysis, a new set of substituent parameters (cr+) was defined. The procedure described above for the definition of cr+, was adopted for... 

In this equation, the substituent parameters and reflect the incremental resonance interaction with electron-demanding and electron-releasing reaction centers, respectively. The variables and r are established for a reaction series by regression analysis and are measures of the extent of the extra resonance contribution. The larger the value of r, the greater is the extra resonance contribution. Because both donor and acceptor capacity will not contribute in a single reaction process, either or r would be expected to be zero. 

A more ambitious goal is to separate completely resonance effects from polar effects. This involves using separate substituent constants to account for resonance and polar effects. The modified equation, called a dual-substituent-parameter equation, takes... 

Examples of the successful application of such dual substituent parameter (DSP) treatments with field and resonance parameters will be discussed in Sections 8.3 (hydroxy-de-diazoniation) and 10.5 (Sandmeyer reaction). 

A table of correlations between seven physicochemical substituent parameters for 90 chemical substituent groups has been reported by Hansch et al. [39]. The parameters include lipophilicity (log P), molar refractivity MR), molecular weight MW), Hammett s electronic parameters (a and o ), and the field and resonance parameters of Swain and Lupton F and R). 

The reactivities of carbenes toward alkenes have been correlated with the inductive and resonance effects of the carbene substituents, log k — a Eat + fcEaR+ + c.m Analogous correlations cannot be obtained for the reaction rates of carbenes with alcohols, neither with the substituent parameters used by Moss,109 nor with related sets.110 In particular, the substituent parameters do not describe the strong, rate-enhancing effect of aryl groups. For a detailed analysis, see the discussion of proton affinities (Section V.A). 

The most fruitful treatment of the electronic effects of ozt/zo-substituents involves the use of the same cr/ and correlation analysis for meta- and para-substituents by means of the dual substituent-parameter equation 91 or the extended Hammett equation 95 (Section II.B). Obviously it is a considerable assumption that these are valid for ort/zo-substituents and the implication is that in the correlation analysis any peculiarities may be adequately expressed through the coefficients of the inductive and resonance terms. Really satisfactory correlation analysis for any given reaction system requires a large amount of data and can only rarely be accomplished. 

Taft and Topsom s article151 and also Topsom s171 should be consulted for details of the setting up of the scales of substituent parameters. The equation has been applied to a wide range of gas-phase reactivities. (In the multiple regressions an intercept term is often permitted, but usually this turns out to be indistinguishable from zero, as it should be if equation 20 is valid.) For aliphatic and alicyclic saturated systems the resonance term is duly negligible. The roles of field, resonance and polarizability effects are discussed and the interpretat of the various p values is attempted. 

Topsom, 1976) and to treat them separately. In this review we will be concerned solely with polar or electronic substituent effects. Although it is possible to define a number of different electronic effects (field effects, CT-inductive effects, jt-inductive effects, Jt-field effects, resonance effects), it is customary to use a dual substituent parameter scale, in which one parameter describes the polarity of a substituent and the other the charge transfer (resonance) (Topsom, 1976). In terms of molecular orbital theory, particularly in the form of perturbation theory, this corresponds to a separate evaluation of charge (inductive) and overlap (resonance) effects. This is reflected in the Klopman-Salem theory (Devaquet and Salem, 1969 Klop-man, 1968 Salem, 1968) and in our theory (Sustmann and Binsch, 1971, 1972 Sustmann and Vahrenholt, 1973). A related treatment of substituent effects has been proposed by Godfrey (Duerden and Godfrey, 1980). 

Despite these difficulties, it appears that the most potent substituent scales are those where a DSP set is used (Topsom, 1976) instead of a single a-value for the electronic effect. While one cr-inductive (cr,) parameter is used in all molecular situations, it seems preferable to apply several o-resonance (cir) parameter sets. Here, the system to which the substituent is attached is taken into account. This corresponds to the fine tuning above mentioned. Some values for common substituents are given in Table 1 (Topsom 1976). 

The DSP approach nicely answers the controversial question about which substituent parameters should be employed to correlate pKa data for 4-substituted pyridinium ions. Statistically, the best correlation is given by Eq. (9), which has values to measure the resonance contribution of a substituent, a result in keeping with chemical intuition. This correlation is statistically superior to a Hammett treatment, where both resonance and inductive effects of a group are combined into a single parameter, p or ap.53,54 Moreover, now it is possible to rationalize why a simple Hammett treatment using ap works so well. Equation (9) reveals that the protonation equilibrium is much more sensitive to an inductive effect (p, — 5.15) than to a resonance effect (p = 2.69). Hence, substituent parameters, such as erp, which are derived from a consideration of the dissociation constants for benzoic acids where resonance contributions are small serve as a useful approximation. The inductive effect is said to have a larger influence on pKa values for pyridinium ions than for benzoic acids because the distance between the substituent and the reactive site is shorter in the pyridine series.53... 

Alkenes and aromatics. The resonances for these classes of compounds appear in the same region (80-140 p.p.m. downfield from TMS) since in both cases the carbon atoms are sp2-hybridised. Empirical rules for calculating the position of absorption in acyclic alkenes have been developed the appropriate substituent parameter is added to the value for carbon in ethylene (123.3 p.p.m.). 

C chemical shifts in aromatic compounds are dependent on the polarity of the substituent. Appendix 3, Table A3.14 shows the substituent effects for a range of substituted benzenes. The 13C spectra of substituted benzenes can often be interpreted on the basis of these substituent parameters in association with data from off-resonance decoupled spectra. 

Conjugation of the nitrogen lone pair with the adjacent phosphoryl or carbonyl function was tested by the 13C NMR parameters of the N-phosphorylated and N-acetylated aniline, as well as of their complexes with Lewis acids. (h ). The inductive and resonance constants for the neutral and charged amide groups were determined using the dual substituent parameter (dsp) approach. ( 5). Results are given in the Table. 

The 3,4-dihydroisoquinoline system is also encountered in this family of alkaloids. The assignment of chemical shifts to the aromatic carbon atoms of the substituted 3,4-dihydroisoquinolines (21-25 in Fig. 3 and Table III) followed directly from the application of the appropriate substituent parameters to the shifts reported for 20 (22) and from a consideration of the resonance effect of the carbon-nitrogen double bond. This latter point is especially evident in the methiodide salts, 24 and 25, where charge delocalization causes C-4a, C-6, and C-8 to appear at lower field than their counterparts C-8a, C-l, and C-5, respectively. Carbon-1 was readily recognized as the lowest field resonance because of its imine character. 

---