Regioselectivity solvent-dependent

A NMR study of reactions of methyl 2-(bromomethyl)-but-2-enoate with the sodium enolate of methyl 2-methyl-3-oxobutanoate has been carried out to rationalize the observed solvent-dependent regioselectivity in terms of addition-eiimination sequences. ... 

Systematic studies on 3-fluoropyridine, the first mono halopyridine to be shown to undergo the DoM process [72CR(C)(275)1535], showed that metalation regioselectivity was dependent on reaction conditions (solvent, temperature, time, metalating agent). 

M, Ek, P. J. Garegg, H. Hultberg, and S. Oscarson, Reductive ring openings of carbohydrate benzylidene acetals using borane-trimethylamine and aluminium chloride. RegioSelectivity and solvent dependence, J. Carbokydr. Chem. 2 305 (1983). 

Additionally, the regioselectivity is strongly solvent-dependent. The most common solvents used in the AA process are alcohols or acetonitrile with high water content (up to 50%). A large amount of water is necessary to ensure high catalytic turnover by enhancing the hydrolysis step in the catalytic cycle. However, in the Boc-AA the opposite was true. Lower water content was favorable by reducing the competitive hydrolysis of the imido osmium species [20], which... 

Octalone dienamines, e.g. 152, have shown a remarkable solvent dependence in the reaction with methyl vinyl ketone (equation 31)86. In methanol, reaction occurs primarily at the less reactive -position of the dienamine and the mechanism probably involves a prototropic shift in the initially formed enolate anion to give 153, and subsequent cyclization onto C-8a of the enimmonium salt to give 154. In toluene, the change in regioselectivity is complete and the product obtained is 157, which must arise from 156, formed in turn from cycloaddition to cross-conjugated dienamine 155. 

The lithiation of /ra .s-A-alkyl-2,3-diphenylaziridines such as 51 was reported to be completely a-regioselective while the stereochemical course of the lithiation-trapping sequence was found to be solvent dependent <07OL1263>. Retention of configuration was observed in hexane, ether, or toluene, while coordinating solvents such as THF or toluene/crown ether produced inversion. Related isomerization and dimerization reactions of a-lithiated terminal aziridines were also reported <07JOC10009>. 

An example of a solvent-dependent change in regioselectivity is the lipase-catalysed partial transesterification of the diester derived from 2-octyl-1,4-... 

The observed regioselectivity can be perturbed to varying degrees by choice of reaction parameters. Solvent polarity can play a role in the control of regioselectivity, as would be predicted by the polar exci-plex model. The regioselectivity of the dimerization of cyclopentenone (equation 11) produces a larger proportion of the head-to-head adduct in more polar solvents. The photoaddition of enone (11) to al-kene (12) also displays a pronounced solvent dependence (equation 12). A consequence of the solvent effect is that nonpolar solvents tend to produce products which would be predicted from the polar exci-plex model, while more polar solvents result in somewhat more of the minor product but do not cause complete reversal of the regioselectivity. 

Similar treatment of TMS propargyl iodide affords a mixture of allenic and pro-pargylic adducts (Table 48). The regioselectivity is strongly solvent-dependent. Allenic adducts predominate in MeCN-DMSO whereas in 1,2-dimethoxyethane (DME) the propargylic adducts are the major products. In these additions the structure of the aldehyde plays a minor role. 

Similarly, for alkenes derived from saturated methyl ketones the regioselectivity is determined by starting hydrazone ( ) (Z) ratios in some solvents but not in others. Thus 2-octanone trisylhydrazone, which is an inseparable 85 15 mixture of ( )- and (Z)-isomers, gives an 85 15 ratio of l-octene 2-octene if vinyllithium formation is carried out in THF, but a 98 2 ratio of the same products when 10% TMEDA-hexane is the solvent. The implication of this observation is that in THF the regioselectivity is determined by azomethine stereochemistry but that in TMEDA-hexane it is not. Note, however, that in this case a iyn-directing effect does not occur in TMEDA, whereas in the previous example it does. Thus more than 10 years after it was asserted that a detailed explanation of the observed solvent dependencies...await further studies owing to the complexities of the reaction system. , little headway has been made. 

Methylation and benzylation of the pyrrolidine dienamine of 3-methyl-A "-2-octa-lone gives a mixture of N- and jS-alkylated products in protic and aprotic solvents. However, the position of attack by acrylonitrile and methyl acrylate is solvent-dependent. In protic solvents the / -alkylated octalone is obtained on hydrolysis, whereas in aprotic solvents the -alkylated product is produced (Scheme 7). This change in the regioselectivity arises from the C-3 methyl group being forced into a quasi-axial orientation because of allylic strain in the equatorial orientation. As a consequence the carbanionic centre in the initially formed zwitterion 6 cannot be neutralized by an internal proton transfer of an axial proton at C-3. In protic solvents intermolecular protonation renders the reaction irreversible, but in aprotic solvent reversion to starting material occurs. This allows 5- or -alkylation to occur and this is rendered irreversible by internal proton transfer from the 6 - or 5-position, respectively, to the carbanionic centre in the resulting zwitterion (Scheme 7). 

When more than one reactive position is available in a heterocyclic substrate, as is often the case for pyridines for example, there are potential problems with regioselectivity or/and disubstitution (since the product of the first substitution is often as reactive as the starting material). Regioselectivity is dependent to a certain extent on the nature of the attacking radical and the solvent, but may be difficult to control satisfactorily. ... 

---