Rearrangement soft carbon nucleophiles

The first step is just the SN2 displacement of Cl- by RS that you have already seen. The second step actually involves chlorination at sulfur (you have also seen that sulfides are good soft nucleophiles for halogens) to form a sulfonium salt. Now a remarkable thing happens. The chlorine atom is transferred from the sulfur atom to the adjacent carbon atom by the Pummerer rearrangement. 

Hydroxide and alkoxide ions, both hard nucleophiles, react with ethyl acrylate 93, an a,p-unsaturated ester, by direct attack at the carbonyl carbon to bring about ester hydrolysis and ester exchange, respectively. However, the enolate 94, a soft nucleophile, reacts in conjugate manner to form 95 predominantly. In an alternate pathway, it is also likely that the enolate 94 reacts through the oxy anion (hard nucleophile) directiy at the carbonyl carbon (hard electrophile) to generate the species 96, which rearranges in an oxy anion accelerated [3.3] sigmatropic shift manner, as shown, to form 97 and, thus, the above product 95. However, it is not certain that such a direct attack by the enolate is not more rapid and reversible. 

Ionization of substrates 1 and 2 leading to the symmetrically 1,1-disubstituted diastereomeric Ti-allyl complexes 3 and 4 also allows complete conversion to one product enantiomer, provided that nucleophilic attack occurs at the carbon bearing substituent R2. High enantiomeric excess may be achieved if a rapid equilibration between the two intermediate re-allyl species is established and the soft carbanion preferentially attacks one of them. Interconversion of the reactive complexes is possible via epimerization by nucleophilic attack of free palladium(O) anti to the jr-allyl complex or by n-a-n rearrangement involving the formation of a Pd-C c-bond at the symmetrically substituted allyl terminus. This process is only fast for R1 = H, due to a low degree of steric congestion or for R1 = aryl because of rc-benzyl participation. 

---