Control of alkyne coupling to obtain cyclooctadecanonaene

Question

How is the resulting annulene's ring size regulated? We get trimerisation, but aren't other reactions very possible?

This question originated with my investigation of the synthesis of cyclooctadecanonaene with a basis in Org. Synth. 1974, 54, 1

Although the diagram in the case study makes sense, I see no reason why more polyyne’s cannot be added to the matrix. In step A, the very poor yield (less than 1%!) seems to me that this is just brute chance of happening.

If that were the case then many more different possible configurations should exist. It made me question if there are any steric or templating effects happening. This applies to any annulene.

Answer 1

How is the resulting annulene's ring size regulated? We get trimerisation, but aren't other reactions very possible?

Answer to your question is "yes." If you read the given reference carefully, you would find there is at least one other cyclic isomer (tetradehydro[24]annulene) exists in the product mixture, which is also isolated. According to submitters' original work (e.g., Ref.1), There is no control to get desired cyclic product from reaction A. The abstrct of Ref.1 states that:

The oxidative coupling of 1,5-hexadiyne (I) with cupric acetate in pyridine yields the cyclic trimer II, tetramer III, pentamer IV, hexamer V and heptamer X. The cyclic trimer is rearranged smoothly by means of potassium t-butoxide in t-butyl alcohol to the fully conjugated eighteen-membered ring monocyclic hexaene-triyne XIII (tridehydro-[18]annulene), a reaction which also produces some triphenylene. Similar rearrangements of III, IV and V give rise to the corresponding conjugated monocyclic polyene-polyynes tetradehydro-[24]annulene, pentadehydro-[30]annulene and hexadehydro-[36]annulene (XIV, XV and XVI, or isomers), respectively, The synthesis of these conjugated substances is carried out most conveniently by the coupling of 1,5-hexadiyne followed by direct rearrangement, without isolation of intermediates. The properties of the novel conjugated polyene-polyynes are discussed briefly, especially in relation to their possible aromatic nature.

In addition to these cyclic products, when subjected to exhaustive column chromatography with alumina (700 fractions), the first compound to be eluted from the column showed to be the linear dimer, 1,5,7,11-dodecatetrayne (9% yield; m.p.: $98$-$\pu{99 ^\circ C}$) as well. This linear dimer was followed by four crystalline substances, which proved to be four cyclic poly-acetylenes, II, III, IV and V in order of elution, respectively. It was estimated that these cyclic poly-acetylenes had been formed in yields of ca. 6, 6, 6, and 2%, respectively, making 1,5,7,11-dodecatetrayne being the major product. Further, it was reported that in one of identical reaction process, heptamer X was obtained in ca. 2% yield, in place of cyclic hexamer V.

There is different approach to make [18]annulene with more control and better yield has been published (Ref.2):

References:

1. Franz. Sondheimer and Reuven. Wolovsky, "Unsaturated Macrocyclic Compounds. XXI.$^1$ The Synthesis of a Series of Fully Conjugated Macrocyclic Polyene-polyynes (Dehydro-annulenes) from 1,5-Hexadiyne," J. Am. Chem. Soc. 1962, 84(2), 260–269 (DOI: https://doi.org/10.1021/ja00861a028).
2. H. P. Figeys and M. Gelbcke, "Olefinic and acetylenic compounds. I. Improved simple synthesis of [18]annulene; new stereospecific synthesis of 1,5-hexadiyne-3-ene cis," Tetrahedron Letters 1970, 11(59), 5139-5142 (DOI: https://doi.org/10.1016/S0040-4039(00)96960-0).