C3-symmetric lanthanide tris(alkoxide) complexes formed by preferential complexation and their stereoselective polymerization of rac-lactide.

A great number and variety of homogeneous lanthanide catalysts are now used in industry and academia as low-cost, low-toxicity, Lewis acidic, and coordination catalysts. In the search for new asymmetric catalysts, chiral C3-symmetric complexes are emerging as interesting competitor systems to the ubiquitous C2-symmetric systems. [2] Trivalent lanthanide cations are obvious candidates for the development of catalysts with threefold symmetry. However, the lability and the weak coordination geometry preferences of Ln centers make the synthesis of enantiopure lanthanide coordination compounds a difficult goal. One approach is to use preresolved, chiral polydentate ligands such as C3-symmetric tris(oxazoline) adducts. Another involves the coordination of three enantiopure, C2-symmetric, biaryl ligands to make a homochiral [LnL3] complex. [4] The most successful asymmetric and bifunctional lanthanide catalysts are based on Li3[Ln(L)3] (L= chiral enantiopure 1,1’-binaphtholate, binolate). This latter class includes, to the best of our knowledge, the only example of spontaneous resolution of three molecules of a racemic ligand around a lanthanide center to date: the reaction of [Y{N(SiMe3)2}3] with rac-NaHbinol affords an equal mixture of the heterobimetallic (RRR)and (SSS)Na3[Y(binol)3] complexes A (binol= 1,1’-bi-2-naphthyl). However, all other lanthanide/alkali-metal combinations give different combinations of (RRS)and (SSR)complexes. The polymerization of the biorenewable monomer raclactide (B) into the biodegradable polymer polylactide (PLA) provides an interesting challenge for new chiral catalysts, since the physical properties of the lactide polymers are highly dependent on the polymer stereochemistry. It has been shown that when independently synthesized poly[(R)-(lactic acid)] and poly[(S)-(lactic acid)] are mixed, the chains cocrystallize to form a stereocomplex with a melting point of up to 230 8C, which is 50 8C higher than that of poly[(S)(lactic acid)] (PLLA or poly[l-(lactic acid)]) alone. However, the enantiopure R,Rmonomer is much more expensive than the rac-lactide. Thus, a one-pot catalytic process in which a racemic mixture of a chiral catalyst polymerizes (R,R)-(d-lactide) and (S,S)-(l-lactide) monomers separately into two, enantiomerically pure, isotactic polymer chains, which can then mix to form a stereocomplex, is an important goal. Currently, highly selective, single-site catalysts based on the Al-salen framework C have been shown to promote the formation of polylactide with a high stereoblock content, in which the long, alternating chains of (RRRRRR)nand (SSSSSS)nPLA are formed, which subsequently form a stereocomplex polylactide with melting points up to 196 8C. Both single-site [LLn(OR)] and homoleptic [M(OR)3] lanthanide alkoxides have been shown to be excellent initiators for the synthesis of PLA, but the products are normally heterotactic. Herein we show how a racemic mixture of a bidentate ligand L, which contains a single chiral carbon center, is resolved into a mixture of two homochiral C3-symmetric complexes upon lanthanide complexation to form (RRR)[LnL3] and (SSS)-[LnL3]. The utility of this spontaneous resolution is demonstrated by the use of these chiral complexes as initiators for the formation of high-melting, stereoregular polylactides. HL is a chiral alcohol that is derived from a cheap, commercially available chiral (racemic or enantiopure) epoxide. The colorless product 1 (HL, (tBu)2P(O)CH2CH(tBu)OH) was obtained by treatment of a mixture of racemic 3,3-dimethylepoxybutane and n-butyllithium with di-tert[*] Dr. P. L. Arnold, J.-C. Buffet School of Chemistry Joseph Black Building, University of Edinburgh West Mains Road, Edinburgh, EH93JJ (UK) Fax: (+44)131-650-6453 E-mail: [email protected]