The absorption spectra of nitroxides in the visible region are highly solvatochromic. In this paper we report some data on solvent effects on the band energy, ET, of 2,2,6,6-tetramethylpiperidinyl-1 -oxy (TEMPO) and 4-oxo-2,2,6,6-tetramethylpiperidinyl-l-oxy (OTEMPO). Special attention has been given to hydroxylic solvents and particularly to a series of reference solvents of graded hydrophilicity, dodecane, primary alcohols, ethanol-water and methanol-water mixtures, and water. The polarity index H, defined as the ratio of molar concentration of OH dipoles in these solvents to that in water (55.4 M), has been shown to exhibit a good linear correlation with their bulk dielectric constant, D. This relation has been examined in terms of theories of dielectric constants of associated liquids. ET values of both TEMPO and OTEMPO show excellent linear correlations with D and H in hydroxylic solvents including methanol-water and ethanol-water mixtures. The values also show good linear correlations with the nitrogen hyperfine splitting constants (AN) of di-tert-butyl nitroxide (DTBNO) and other similar molecules for both aprotic and hydroxylic solvents. A general theoretical approach for describing aprotic solvent effects has been explored. It is found that a reaction field treatment for estimating the effects of the solvent dielectric constant on the energy of the nitroxide dipole gives a reasonable description of An and ET data in many aprotic solvents. The role of the effective polarizability of the dipole has been investigated. It has been shown that for such systems the refractive index of the solvent plays a relatively minor role, and reasons have been adduced for this in terms of a general theory which indicates that in some other cases refractive index effects can be significant. For hydroxylic solvents equations applicable to aprotic solvents are unsatisfactory because of very pronounced hydrogen-bonding effects on ET and AN. It has been shown that corrections for general medium effects do not materially alter the linear relation of hydrogen-bonding effects to D and H. The hydrogen-bonding effects can be expressed in terms of proportionality to H. The general expressions combining equations for aprotic solvents with a term , where is a proportionality constant, are found to give good descriptions of AN and ET data in both aprotic and hydroxylic solvents. This approach to solvatochromic effects using nonempirical solvent parameters gives as good correlations as empirical polarity scales in representing such data. Preliminary studies indicate the usefulness of such treatments for other systems as well. The relationship of hydrogen-bonding effects to H and D have been analyzed by focusing attention on the electrostatic component of hydrogen-bonding interactions which is expected to be of primary importance. Such analyses have been shown to be consistent with some formal theoretical treatments. Their physical significance has been examined. A possibly important role of a solute-solvent correlation parameter similar ...

No abstract

as spectroscopic "probes" of the hydrocarbon interiors of micelles and other lipid assemblies. Many common simplifying assumptions about the behavior of such spectroscopic probes in lipid assemblies can be grossly misleading. The implications of some of the results obtained for spectroscopic probe studies are pointed out.

Micelle-water distribution coefficients have been determined for 2,2,6,6-tetramethylpiperidinyl-l-oxy (TEMPO) and 4-oxo-2,2,6,6-tetramethylpiperidinyl-l-oxy (OTEMPO) in sodium dodecyl sulfate (SDS) and cetyltrimethylammonium chloride (CTAC) and in several other micellar systems of different charge types, including the phospholipid palmitoyl lysolecithin, for TEMPO. In order to interpret the distribution coefficients and the overall solubilizing powers of the micellar systems, we used a recently proposed two-state model which invokes an equilibrium distribution between a "dissolved state", associated with the hydrocarbon core, and an "adsorbed state", associated with the interface region. According to this model, solublization is due in part to solubility in the hydrocarbon core and in part to adsorption to the interface arising out of the interfacial activity of the solubilizate. Dodecane-water bulk partition coefficients have been determined for TEMPO and OTEMPO. On a comparative basis, the overall micelle-water distribution coefficients of TEMPO and OTEMPO in SDS are 6 and 100 times higher than the corresponding dodecane-water partition coefficients. When corrections for solubility in the hydrocarbon core are made for the high Laplace pressure in SDS micelles calculated from the estimated interfacial tension and the radius, the micelle-water distribution coefficients for SDS are found to be higher by factors of 71 and 1600 for TEMPO and OTEMPO. This calculation indicates that less than 2% of the total solubility in the micelles should be ascribed to the hydrocarbon core and more than 98% of the solubilized molecules are in the adsorbed state. Evidence confirming the interfacial activity of the nitroxides has been obtained from the lowering of the tension of the dodecane-water interface by both TEMPO and OTEMPO. These data also indicate that about 99% of solubilized TEMP and OTEMPO molecules are in the adsorbed state, the large fraction arising from the very high surface-to-volume ratios of micelles. The interfacial tension data indicate furthermore a higher adsorbability of OTEMPO as compared to TEMPO, in agreement with deductions from micelle-water distribution equilibria. From the dodecane-water partition coefficients and the use of model hydrocarbon partition data, it is shown that the contribution of the nitroxide group to the free energy of transfer from dodecane to water is approximately -7.5 kcal/mol whereas that of the C=0 group of OTEMPO is approximately -2 kcal/mol. The hydrophilicity of the nitroxide group, arising mainly from hydrogen-bonding interactions, provides a molecular basis for the effects observed. The hydrophilicities of the nitroxide group and the 0=0 group, and the predominance of the adsorbed state in solubilization inferred here, are of importance in interpreting the microenvironmental effects experienced by TEMPO, OTEMPO, and other nitroxides in micellar systems, as reported in the previous paper. These were investigated by using solvatochromic visible absorption bands of the nitrox...

No abstract