Isolation of 1,4-Li2-C6H4 and its reaction with [(Ph3P)AuCl]

The difficulty in generating 1,4-Li2-C6H4 utilising the lithium halogen exchange reaction on 1,4-Br2-C 6H4, 1,4-I2-C6H4 and 1-Br-4-I-C6H4 is revisited and only on treatment of 1,4-I2-C6H4 with 2 molar equivalents of n-BuLi can 1,4-Li2-C6H41 be isolated in excellent yield. Treatment of 1 with two equivalents of [ClAu(PPh3)] gives [1,4-(Ph3PAu)2-C6H4] 2a in excellent yield. Subsequent treatment of 2a with 2.5 molar equivalents of PPh 2Me, PPhMe2 or PMe3 affords the PPh3 substituted compounds [1,4-(LAu)2-C6H4] (L = PPh2Me 2b, PPhMe22c, PMe32d) in essentially quantitative yields. On treatment of 1,4-Br2-C6H 4 or 1-Br-4-I-C6H4 with 2 molar equivalents of n-BuLi only mono-lithiation takes place to give 1-Br-4-Li-C6H 43 as shown through the isolation of essentially 1:1 molar equivalents of Ph2PC6H4-4-Br and Ph 2PBu on treatment with 2 molar equivalents of ClPPh2. Treatment of 3, prepared by lithium/iodine exchange on 1-Br-4-I-C 6H4, with [ClAu(PPh3)] affords [(Ph 3P)Au(C6H4-4-Br)] 4 as expected and in addition [(Ph3P)Au(n-Bu)(C6H4-4-Br)2] 5, indicating the straightforward chloride/aryl exchange at gold may proceed in competition with oxidative addition of the n-BuI, generated in the initial lithium/iodine exchange reaction, to some 'aurate' complex Li[Au(C 6H4-4-Br)2] 6 formed in situ followed by reductive elimination of Br-C6H4-4-n-Bu in a manner that mimics lithium diorganocuprate chemistry. All of the gold-containing compounds have been spectroscopically characterised by 1H and 31P-{1H} NMR and in addition compounds 2a-d and 5 by single crystal X-ray diffraction studies. The solid state structures observed for 2a-d are dictated by non-conventional hydrogen bonding and the packing requirements of the phosphine ligands. For 2a and 2b there is no close Au⋯Au approach, however for 2c and 2d the reduction in the number of phenyl rings allows the formation of Au⋯Au contacts. For 2c and 2d the extended structures appear to be helical chains with Au⋯Au contact parameters of 3.855(5) Å and C-Au-Au-C 104.1(3)° for 2c and 3.139(4) Å and C-Au-Au-C -92.0(2)° for 2d. The Au⋯Au approach in 2c is longer than is normally accepted for an Au⋯Au contact and is dictated by ligand directed non-conventional hydrogen bonding to the aurated benzene ring and the π-stacking requirements of the phosphine ligand. By comparison of the structures 2a-2d with other structures in the database it is evident that the aurophilic interaction is a poor supramolecular synthon in the presence of non-conventional hydrogen bond donors. Searches of the CCDC database suggest that the observed parameters for the Au⋯Au contact in 2c sit close to the cut-off point for observing this type of contact. In addition to aurophilic contacts and non-conventional hydrogen bonds there are a number of halogenated solvent C-Cl⋯Au contacts observed in the structures of 2a and 2d. The nature of these contacts have implications for the accepted van der Waals radius of gold which should be extended to 2 Å. © 2010 The Royal Society of Chemistry.