Solvent acidity the SA scale

As stated above, a probe for solvent acidity should possess a number of features to be usable in UV-Vis spectroscopy. One is that it should be basic enough in its electronic ground state 

Our group has shown that the extremely strong negative solvatochromism of the chromophore stilbazolium betaine dye (11), about 6500 cm, is not a result of a change in the non-specific effect of the solvent but rather of change in acidity. This was confirmed by a study of the solvatochromic effect of a derivative of (11), o,o -di-tert-butylstilbazolium betaine dye (DTBSB) (12) in the same series of solvents. In this compound, the basic site in the betaine dye (its oxygen atom) is protected on both sides by bulky tert-butyl groups the specific effect of the solvent is hindered and the compound exhibits only a small solvatochromic effect that can be ascribed to non-specific solvent effects.  

Since the carbonyl group of stilbazolium betaine possesses two lone pairs in the plane of the quinoid ring, one can have two (11), one (TBSB, 13) or no channels (DTBSB) to approach hydrogen bond-donor solvents, depending on the number of o-tert-butyl groups present in the molecular structure of the (11) derivative concerned. Our group has also shown that, for at least 20 alkanols, the wavenumber difference between the maximum of the first absorption bands for (11) and TBSB, and for TBSB and DTBSB, is virtually identical. 

These results show that an oxygen lone pair in (11) is basic enough, as a result of hydrogen bonding, for (11) to be used as an acidity probe in UV-Vis spectroscopy thus, the absorption maximum for TBSB shifts by about 1000 cm from 1-decanol to ethanol, whereas that for DTBSB shifts by only 300 cm. This sensitivity to acidity, and the structural likeness of the probe and its homomorph -which must endow them with a similar sensitivity to solvent basicity and dipolarity/polarizability-, suggest that the two compounds make an appropriate probe-homomorph couple for developing a pure solvent acidity scale. For this purpose, the first visible absorption band for these compounds exhibits quite an in- 

The SA acidity scale was estabhshed by comparing the solvatochromism of the probes TBSB and DTBSB using flic method of Kamlet and Taft. In this method, the solvatochromism of DTBSB in a solvent is used as die reference for zero acidity. Consequently, non-acidic solvents obey flic equation 

Based on eq. [ 10.3.11 ] for non-acidic solvents, the SA value for a given solvent can be obtained from the following expression, where a value of 0.4 is assigned to ethanol, the acid solvatochromic behavior of which is exhibited at 1299.8 cm  

---

The SA scale was established using the solvatochromic comparison method of Kamlet and Taft [224, 226]. A good linear correlation (r = 0.961) has been found between the wavenumbers of the absorption maxima of dyes (52) and (53), measured in fifty non-acidic or non-HBD solvents, according to Eq. (7-40c) ... 

In going from tetrahydrofuran to ethanol, these Av values increase from —39 to 1300 cm In order to obtain a normalized scale of near-zero HBD acidity for non-HBD solvents and a value of roughly unity for HBD solvents, the SA scale was defined as follows [296, 337a] ... 

Evaluating the acidity of weakly acidic solvents entails using a probe basic enough to afford measurement of such an acidity as a result, the probe is usually protonated by strongly acidic solvents and useless for the intended purpose. Stilbazolium betaines are subject to this constraint as they have pK values of about 10 -(11) and DTBSB have a pKa of 8.57 and ca. 10, respectively, which makes them unsuitable for the evaluation of solvents more acidic than methanol. This forced us to find a suitable probe with a view to expanding the SA scale to more acidic solvents. This problem is not exclusive to our scale in fact, it affects all acidity scales, which usually provide little information about strongly acidic solvents. 

Finally, the ambitious approach of Catalan et al. to introduce complete new comprehensive scales of solvent dipolarity/polarizabihty [SPP scale), solvent basicity SB scale), and solvent acidity SA scale) must be mentioned [296, 335-337]. These three UV/Vis spectroscopic scales are based on carefully selected positively solvatochromic and homomorphic pairs of probe dyes and include values for about 200 organic solvents for a recent review, see reference [296]. The molecular structures of the three pairs of homomorphic indicator dyes proposed are as follows ... 

The general SPP scale of solvent dipolarity/polarizability and the specific SB and SA scales of solvent HBA basicity and HBD acidity, respectively, are orthogonal to one another and they can be used in the correlation analysis of solvent effects in single- or, in combination with the others, in two- or three-parameter correlation equations, depending on the solvent-influenced process under consideration see also Section 7.7. Examples of the correlation analysis of a variety of other solvent-dependent processes by means of SPP, SB, and SA values, including those used for the introduction of other solvent polarity parameters, can be found in references [335-337, 340-342]. In particular, comparisons with Kamlet and Taft s n scale [340] and Winstein and Grunwald s Y scale [341] have been made. 

The use of 3,6-diethyl-l,2,4>S-tetrazine 401 as a new solvatochromic probe enabled the extension of the solvent acidity (SA) scale to carboxylic acids, phenols, haloalcohols, and halocarboxylic acids (Figure 6) <1999EJ0885>. 

Substitution into alcoholic compounds allows the entire range of the solvent acidity scale to be spanned. Thus, 2-octanol can be considered to be seareely HBD (SA=0.088) and ethylene glycol to be highly HBD (SA=0.717). Perfluoroalkanols are even more highly HBD. 

Reported scales for describing solvent polarity [f(8,n), Jt " and S ], basicity (DN and P ) and acidity (AN and a ) were previously analyzed against our SPP, SA and SB scales in the originating references, so no fiirther comment is made here. 

Recently, Sapre et al. showed that a plot of the maximum fluorescence of Neutral Red (NR) against the solvent polarity function Af is clearly bilinear (see Figure 10.3.8). Accordingly, they concluded that, in solvents with Af > 0.37, the emitting state changes to a much more polar, ICT state. The analysis of this spectroscopic data in the light of our scales reveals that, in fact, the solvatoehromism of NR is normal, albeit dependent not only on the polarity of the solvent (SPP), but also on its acidity (SA) (see Figure 10.3.9) ... 

A brief review has been presented of the correlation analysis of solvolysis rates 50 years later, i.e. since Grunwald and Winstein proposed their eponymous equation in 1948.111 -pije authors then propose a method of correlation analysis involving multiple regression on solvent scales SPP (polarity-polarizability), SA (acidity) and SB (basicity). These scales are based on the solvatochromism of suitable probes and were initially for pure (i.e. one-component) solvents, but have now been extended to binary solvent mixtures. This enabled the authors to present a correlation for the solvolysis rate constants of r-butyl chloride in 27 pure solvents and 147 binary solvent mixtures, having a correlation coefficient r = 0.990 and a standard error of the estimate s = 0.40. The most important term in the equation is that involving SPP next comes... 

Catalan J (2009) Toward a generalized treatment of the solvent effect based on four empirical scales dipolarity (SdP, anew scale), polarizability (SP), acidity (SA), and basicity (SB) of the medium. J Phys Chem B 113 5951-5960... 

---