Addition reactions with solvent molecules

The chain unit in the thermal and photochemical oxidation of aldehydes by molecular dioxygen consists of two consecutive reactions addition of dioxygen to the acyl radical and abstraction reaction of the acylperoxyl radical with aldehyde. Experiments confirmed that the primary product of the oxidation of aldehyde is the corresponding peroxyacid. Thus, in the oxidation of n-heptaldehyde [10,16,17], acetaldehyde [4,18], benzaldehyde [13,14,18], p-tolualdehyde [19], and other aldehydes, up to 90-95% of the corresponding peroxyacid were detected in the initial stages. In the oxidation of acetaldehyde in acetic acid [20], chain propagation includes not only the reactions of RC (0) with 02 and RC(0)00 with RC(0)H, but also the exchange of radicals with solvent molecules (R = CH3). 

Certain metal salts can react with unsaturated compounds with the concomitant addition of a solvent molecule or other nucleophile to yield p-substituted metal derivatives. These reactions, called solvometallations or oxymetallations, have a number of applications in organic synthesis. A short treatment of certain classic and important reactions is given here. 

When reactions take place at the gas-soUd interphase no additional interactions with medium molecules, as solvent molecules present at the solid-liquid interphase, can take place. Solvation effects are absent. Therefore reactions that require charge-separation are rare and if they occur need charge stabilization by strong electrostatic interactions with the surface. Examples of such reactions are acid-base reactions that occur on oxidic surfaces. 

Many of the C-F activation reactions discussed in this account are complicated by multi-step reaction sequences and secondary reactions with solvent or fluoride ion generated in the course of the transformation. In catalytic systems, the additional reagents needed to achieve catalytic turnover may cause undesired side reactions to take place. Gas-phase ion molecule reactions provide one way to study model systems in the absence of these complicating factors to obtain fundamental information on the C-F activation process [84]. In addition, theoretical treatments of C-F bond activation are beginning to provide insight into these transformations. 

The stereoisomers of olefin saturation are often those derived by cis addition of hydrogen to the least hindered side of the molecule (99). But there are many exceptions and complications (97), among which is the difficulty of determining which side of the molecule is the least hindered. Double-bond isomerization frequently occurs, and the hydrogenation product is the resultant of a number of competing reactions. Experimentally, stereochemistry has been found to vary, sometimes to a marked degree, with olefin purity, reaction parameters, solvent, and catalyst 30,100). Generalizing, it is expedient, when unwanted products arise as a result of prior isomerization, to avoid those catalysts and conditions that are known to favor isomerization. 

For this specific task, ionic liquids containing allcylaluminiums proved unsuitable, due to their strong isomerization activity [102]. Since, mechanistically, only the linkage of two 1-butene molecules can give rise to the formation of linear octenes, isomerization activity in the solvent inhibits the formation of the desired product. Therefore, slightly acidic chloroaluminate melts that would enable selective nickel catalysis without the addition of alkylaluminiums were developed [104]. It was found that an acidic chloroaluminate ionic liquid buffered with small amounts of weak organic bases provided a solvent that allowed a selective, biphasic reaction with [(H-COD)Ni(hfacac)]. 

The differences in rate for the two positions of naphthalene show clearly that an additional-elimination mechanism may be ruled out. On the other hand, the magnitude of the above isotope effect is smaller than would be expected for a reaction involving rate-determining abstraction of hydrogen, so a mechanism involving significant internal return had been proposed, equilibria (239) and (240), p. 266. In this base-catalysed (B-SE2) reaction both k and k 2 must be fast in view of the reaction path symmetry. If diffusion away of the labelled solvent molecule BH is not rapid compared with the return reaction kLt a considerable fraction of ArLi reacts with BH rather than BH, the former possibility leading to no nett isotope effect. Since the diffusion process is unlikely to have an isotope effect then the overall observed effect will be less than that for the step k. ... 

---