No abstract

1-Phenyl-3,3-biphenyleneallene (2), the base-catalyzed rearrangement product of 9-phenylethynylfluorene (1) yields a yellow, head-to-tail dimer 6 that, upon gentle warming, is converted to the red tail-to-tail isomer trans-3,4-diphenyl-1,2-bis(fluorenylidene)cyclobutane (7), in which the two fluorenylidene moieties severely overlap. The helical sense of the fluorenylidene moieties in 7 matches that of the phenyl substituents, and the interplanar angle between the fluorenylidene moieties is 41 degrees . At 80 degrees C, 6 isomerizes to orange cis-3,4-diphenyl-1,2-bis(fluorenylidene)cyclobutane (8), which at 110 degrees C is converted to orange trans diastereomer 9, whereby the helicity of the overlapping fluorenylidene moieties is reversed from that in 7 such that they are aligned with the ring hydrogen atoms, and the interplanar angle between the fluorenylidene moieties is now 60 degrees . At 180 degrees C, 6 rearranges to dispirodihydrotetracene 3 and blue, electroluminescent diindenotetracene 4, which is readily oxidized to peroxide 5. In the solid state, both 3 and 4 adopt structures with Ci symmetry (only an inversion center) such that the central polycyclic framework is nonplanar. Deprotonation of yellow head-to-tail allene dimer 6 with tBuOK in DMSO and reprotonation with HOAc yields the [1,3]-hydrogen migration product 10, in which the proton originally on the cyclobutane ring is now sited at C9 on the exocyclic fluorenyl substituent. Analogously, deprotonation and reprotonation of orange dimer 9 furnishes [1,3]-hydrogen migration product 11. Side product 17, formed during the synthesis of 1 from 9-phenylethynylfluoren-9-ol, BF3 and Et3SiH, was shown to be a silyl-indene spiro-linked to C9 of fluorene. All products were characterized by NMR spectroscopy and X-ray crystallography, and the mechanisms of these interconversions are discussed.

Caught in the act: The [(dibenzosuberenol)Co2(CO)6] cluster 1 (see scheme) loses a carbonyl ligand to yield 2, the first structurally characterized alkyne–dicobaltpentacarbonyl–alkene complex. This finding provides experimental support for the first step of the proposed mechanism of the Pauson–Khand reaction and yields crucial structural data relevant to computational studies of the process.

The hexacarbonyldicobalt complexes of a range of 5-alkynyl-5H-dibenzo [a,d]cycloheptenes, 11a-e, readily undergo loss of a carbonyl ligand with concomitant formation of pentacarbonyldicobalt clusters, 12a-e, in which the vacant coordination site on cobalt is now occupied by the C(10)-C(11) double bond of the central seven-membered ring. These (η 2 -alkene)(μ-alkyne)pentacarbonyldicobalt complexes provide structural models of the first step of the proposed mechanism of the Pauson-Khand process for the formation of cyclopentenones via the coupling of an alkene, an alkyne, and a source of carbon monoxide. X-ray crystallographic data reveal that the distance between the alkene carbons and the nearest alkyne carbon is approximately 2.85 A ˚, slightly shorter than the theoretically predicted value of ∼2.95 A ˚. VT NMR data for the interconversion of 11b and 12b yielded activation energies for the forward and reverse processes, 29 and 15 kcal mol -1 , respectively, and the enthalpy change for the endothermic process (14 kcal mol -1 ); these match very well the earlier predictions from DFT calculations. Although the ethynyl-hexacarbonyldicobalt complex, 11f, does not suffer facile elimination of CO, it does participate in an intermolecular PKR with norbornadiene. Likewise, attempts to extend the reach of the alkynyl moiety through incorporation of an additional methylene group yield only an intermolecular PKR product. It is suggested that the pentacarbonyl complexes, 12a-e, do not continue along the PKR pathway because of the unfavorable relative orientation of the cobalt-coordinated alkyne and alkene.

Protonation of 9-(ferrocenylethynyl)fluoren-9-ol (9) yields the conjugated enone derived from a Meyer-Schuster rearrangement. However, treatment of 9 with thionyl chloride at -30°C proceeds with elimination of SO 2 to furnish 3,3-biphenylene-1-chloro-1-ferrocenylallene (14), the 13 C NMR data of which indicate that the central carbon does not have markedly carbenic character. Upon warming, this allene readily forms the sterically highly encumbered head-to-head dimer 15, whereby the 1,2-bis(alkylidene)cyclobutane adopts a butterfly conformation with two exocyclic Cl(Fc)Cd moieties and a very long (∼1.65 Å) C(3)-C(4) bond linking the two spiro-bonded fluorenyl fragments. In contrast, attempts to generate the analogous ferrocenylbromoallene 16 instead yielded the hexa-1,2-dien-5-yne 17, the product of allenyl radical head-to-tail coupling. Protonation of ferroceno[2,3]inden-1-ol (25) confirmed a proposal, originally put forward by Caïs in 1965, concerning the possible diradical nature of the resulting cation; the unambiguous characterization of the ferroceno[2,3]inden-1-yl dimer (29) provides experimental verification of Caïs's hypothesis. The X-ray crystal structures of 9-(ferrocenylethynyl)fluoren-9-ol (9) and of its isomer 1-(phenylethynyl)ferroceno[2,3]inden-1-ol (26) are reported along with, among others, the structures of 15, 17, 25, and 29. The possible role of these allenes as precursors to ferrocenylcontaining electroluminescent tetracenes is discussed.