Synthesis and characterization of k-2-bis-N-heterocyclic carbene rhodium(I) catalysts: Application in enantioselective arylboronic acid addition to cyclohex-2-enones Matthew S. Jeletic, Roxy J. Lowry, Jason M. Swails, Ion Ghiviriga, Adam S. Veige * Department of Chemistry, Center for Catalysis, University of Florida, P.O. Box 117200, Gainesville FL 32611, USA article info Article history: Received 8 May 2011 Received in revised form 22 May 2011 Accepted 26 May 2011 Keywords: Carbene diNHC Chiral C 2 -symmetric Rhodium Enantioselective catalysis Arylboronic acid abstract This manuscript describes the synthesis and characterization of new derivatives of di-N-heterocyclic carbene (diNHC) ligands derived from trans-9,10-dihydro-9,10-ethanoanthracene-11,12-diylmethanediyl (DEAM) and trans-9,10-dihydro-9,10-ethanoanthracene-11,12-diyl (DEA). Synthesized as the diazolium salts, the new ligands are employed, in conjunction with known derivatives, to examine specific prop- erties than influence the chiral induction in rhodium-catalyzed arylboronic acid addition to cyclohex-2- enones. Three properties of the diNHC ligands are modified and include: 1) the size and composition of the N-heterocyclic substituent (Me, i Pr, Bn, CHPh 2 , o-MeBn, and R-CHMePh), 2) the type of N-heterocycle (benzimidazole versus imidazole), and 3) the size of the chiral pocket (DEAM versus DEA). Results from the catalytic studies indicate the DEAM ligand, which contains an extra CH 2 group compared to DEA, is too flexible to induce enantiomeric excess. The DEA derivatives containing an imidazole or benzimidazole provide distinctly different results despite relatively small differences between the heterocyclic carbene donors. An unexpected source of chiral induction is rationalized using results from catalytic data (% e.e.) and complemented by X-ray structural data and DFT calculations. Published by Elsevier B.V. 1. Introduction Enantiopure di-N-heterocyclic carbene ligands (diNHCs) are emerging as effective chiral auxiliaries in metal catalyzed asym- metric catalysis [1e25]. However, few design principles exist to guide chiral diNHC catalyst engineering. Representations of a chiral metal ion environment, such as Knowles’ quadrant model [26,27] and Trost’s version [28], serve as good approximations, especially for asymmetric alkene hydrogenation [29] and allylic alkylation [30,31]. These models fit well for chiral diphosphine ligands, but in some cases they do not, resulting in some modifications to the model [32e35]. Advances in computational techniques [36e40] may ultimately permit catalyst designs that correctly select for a specific product enantiomer prior to any laboratory investigation. The problem is that small energetic differences of 1e2 kcal/mol in the diastereo- or enantio-determining transition states lead to large changes in % enantiomeric excess (% e.e.) [41]. Predicting or designing ligand and catalyst features, that discriminate within such a small energy range is difficult. After factoring in the addi- tional variables of temperature, additives, and solvent choice, it is not surprising that trial and error remains the common method for discovery. Providing some guidance regarding NHC [42] ligand steric influence on the metal ion coordination sphere is the %V buried model, but this model applies to monodentate ligands [43,44]. In addition, the relative s-donation strength of NHCs, and thus the M- NHC bond strength, is becoming clearer from both experiment and computational results [45e53]. Despite possessing a strong M-NHC bond, NHCs can reductively eliminate from corresponding metal- hydrido and alkyl species to give imidazolium salts [18,54e63].A systematic study by Hermann et al. provides some insight into the effects of achiral diNHC ligand bridge length, N-substituent, and heterocycle composition on Rh(I) catalyzed hydrosilylation of ketones [64]. Considering the limited models available for designing effective chiral catalysts using diNHCs, it is important that a chosen ligand platform permit rapid and facile modification. Our previously reported DEAM (trans-9,10- dihydro-9,10- etha- no anthracene-11,12-diyl methanediyl) and DEA (trans-9,10- dihy- dro-9,10- ethano anthracene-11,12-diyl) diNHC ligands provide opportunities to easily modify steric and electron-donating features (Fig. 1) [17,18,65e67]. In addition, the DEAM ligand provides an opportunity to link the two NHC heterocycles through the R groups to create chiral dicarbene cyclophane ligands [17]. * Corresponding author. Tel.: þ1 352 392 9844; fax: þ1 352 392 3255. E-mail address: veige@chem.ufl.edu (A.S. Veige). Contents lists available at ScienceDirect Journal of Organometallic Chemistry journal homepage: www.elsevier.com/locate/jorganchem 0022-328X/$ e see front matter Published by Elsevier B.V. doi:10.1016/j.jorganchem.2011.05.015 Journal of Organometallic Chemistry 696 (2011) 3127e3134