The condensation of certain enolates with trichloroethylene to yield a-dichlorovinyl ketones has been shown to proceed by way of dichloroacetylene as an obliqatory intermediate. In a recent communication we described a novel condensation of certain enolates with CI2C= CHCI to yield cl-dichlorovinyl ketones, illustrated in eqn 1.' 1. LDA IHMPA (1) 2. C12C=CHCI 1 2 -Although some limited synthetic applications of this reaction were shown, neither its scope nor its mechanism was understood. At least four discrete mechanisms could be proposed to rationalize the initial data, namely (1) transfer alkylation, 2(2) s 4 5 NR1 radical chain, 3 (3) addition-elimination, and (4) elimination-addition. The complete absence of cl-chloroketones (stable to reaction conditions) argued against transfer alkylation. Neither 0 2, nor light, nor AIBN, nor t-Bu2N0 had effect on product yield, makinq a radical chain mechanism untenable. 3a Differentiation between addition-elimination and elimination-addition (Scheme 1.1 was less obvious. The observed SCHEME I Addition-Elimination Elimination-Addition C12C=CHCI addition C12C=CHCI elimination Clcrccl CICZccl addition s 2369 formation of 1,2-dichlorovinyl rather than 2,2-dichlorovinyl products seemed to argue against addition-elimination since the 2,2_regiochemistry would be anticipated. 6 In addition, it has been shown that the reaction between C12C=CHCI and arylthiolate' or phenolate* nucleophiles proceeds through initial formation of CIC:CCI. On the other hand, our limited knowledqe of the substitution chemistry of CIC-CCI with carbon nucleophiles, based on the work of Ott with sodio diethyl ethylmalonate, would predict formation of the a-( 2-chloroethynyl) derivatives from a dichloroacetylene intermediate rather than the observed dichlorovinyl products. 9 To establish the possible role of CIC:CCI in our dichlorovinylations we undertook to generate CIC:CCI free of C12C=CHCI and to examine its reactions with enolates. None of the several published procedures for preparation of CIC?CCI were suitable in our hands for this purpose. That of Ott 10 gave unacceptable mixtures, and that of Kloster-Jenson 11 qave unreproducible results. We ultimately found that addition of a solution of C12C=CHCI in Et20 at -78'C to a suspension of LiN(SiMe3)2 in hexanes, followed by warming to rt over 3 h and direct distillation at 30-36OC through a 12 cm Vigreux column reproducibly qave a distillate free of C12C=CHCI that contained (glc) ca 50% CIC:CCI, 40% diethyl ether and 10% hexanes. 12 When the kinetic enolate of 1 (from 1. and one equiv each of LDA and HMPA) was reacted in THF with freshly distilled ClCtCCI'Et20, prepared as above, (-78OC + rt, 3 h), no dichlorovinyl ketone 2 was formed. We obtained 64% of the cl-chloroethynyl ketone 3; excess ClCrCCl or Kende, A.