Picolyllithium DOI: 10.1002/anie.200806221 Carbanion or Amide? First Charge Density Study of Parent 2-Picolyllithium** Holger Ott, Ursula Pieper, Dirk Leusser, Ulrike Flierler, Julian Henn, and Dietmar Stalke* Dedicated to Professor Helmut Werner on the occasion of his 75th birthday Organolithium compounds have gained in importance ever since their discovery by Schlenk and Holtz in 1917. [1] Today, they are well-established reagents in organic and inorganic synthesis and are readily applied to suit various preparative protocols. [2] These range from deprotonation of weakly acidic reagents to bond formation (transfer of organic groups) and anionic polymerization reactions. The introduction of coordinating pyridyl side chains (i.e. methylpyridyl) in ligands is one example of a C ÀC bond formation reaction conducted using organolithium com- pounds. The design of pyridyl-substituted ligands [3] usually starts with the deprotonation of 2-picoline (2-methylpyridine) with commercially available n-butyllithium. [4] The reactivity determines the yields and is mainly based on the basicity, steric demand, and the Pearson hardness of the nucleophile. Moreover, the aggregation state of the lithium compound in solution, which can be deduced from single crystal structure determination, influences the behavior of the nucleophile. [5] More detailed information on the reactivity of the molecule is available from diffraction experi- ments. [6] High-resolution X-ray diffraction experiments enable the accurate determination of the molecular electron density distribution in the crystal. The experimental results can be compared to densities derived from gas-phase calculations under the provisions of Baders quantum theory of atoms in molecules (QTAIM). [7] Thus we synthesized and grew single crystals of two unsubstituted 2-picolyllithium compounds differing only in the donor bases. This approach eliminates effects on the anion that derive from additional side-arm groups. The structural analyses should provide undisguised insight into the electron distribution in the aromatic heterocyclic carbanion as a whole, [8] which has been a matter of discussion ever since the first crystal structural analysis of a substituted picolyl- lithium compound. [9] In particular, the controversially dis- cussed properties of Li ÀX (X = C, N, O) bonds should be elucidated. [10] Scherer et al. even chose a derivative of 2- picolyllithium in their pioneering experimental charge density study on Li···H agostic interactions. [11] 2-Picolyllithium (PicLi) was prepared by slowly adding n- butyllithium to an equimolar amount (to give 1) or a 2.5-fold excess (to give 2) of 2-picoline in diethyl ether at À20 8C. Storage in the refrigerator yielded single crystals suitable for crystal structure analysis. The crystals consisted of the dimers [2-PicLi·OEt 2 ] 2 (1) and [2-PicLi·PicH] 2 (2). The compounds crystallize in centrosymmetric space groups (1: P1 ¯ ; 2 : C2/c) with half of each dimer in the asymmetric unit. Since 1 and 2 show similar structural features, a joint discussion of the PicLi motif will be presented (Figure 1). [12] The PicLi dimer is linked by two different lithium–anion interactions: a Li À N bond with the lithium atom located almost ideally in the pyridyl ring plane (deviation: 0.26 ; angle between Li ÀN and the plane: 88) and a h 3 -aza-allylic contact in which the lithium cation is coordinated by the p system of the methylene group (C6), the ipso-carbon atom (C1), and the ring nitrogen atom (N1). The coordination sphere of the lithium cation is completed with one donor molecule per metal atom (1: Et 2 O; 2 : 2-picoline). The in- plane Li ÀN bonds (2.031(2) (1), 2.021(1)  (2)) are about 0.1  shorter than the contacts to the respective aza-allyl nitrogen atoms (2.133(2) (1), 2.136(1)  (2)). The lithium– carbon distances are 2.29  to the ipso-carbon atom (2.297(2) (1), 2.284(1)  (2)) and slightly longer to the methylene carbon atom (2.321(3) (1), 2.328(1)  (2)). Similar bonding situations are reported in the 2-(tri methylsilylmethyl)pyridyllithium-diethyl ether adduct [13] (3) and the 2-(bis(trimethylsilyl)methyl)pyridine adduct [14] (4). There, the aza-allyl nitrogen–lithium bonds (2.19(1) (3), Figure 1. Molecular structures of [2-PicLi·OEt 2 ] 2 (1, left) and [2-PicLi·- PicH] 2 (2, right). [*] H. Ott, Dr. U. Pieper, [+] Dr. D. Leusser, U. Flierler, Dr. J. Henn, Prof. Dr. D. Stalke Institut für Anorganische Chemie der Universität Göttingen Tammannstrasse 4, 37077 Göttingen (Germany) Fax: (+ 49) 551-39-3459 E-mail: dstalke@chemie.uni-goettingen.de [ + ] Present address: Department of Biopharmaceutical Sciences University of California, San Francisco (USA) [**] This work was supported by the Deutsche Forschungsgemeinschaft within the priority program 1178 Experimental charge density as the key to understand chemical interactions, the Volkswagenstiftung, CHEMETALL GmbH Frankfurt, and the Fonds der chemischen Industrie (H.O.). Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/anie.200806221. Communications 2978  2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2009, 48, 2978 –2982