Free-Radical Chemistry DOI: 10.1002/anie.200600831 Water: The Ideal Hydrogen-Atom Source in Free- Radical Chemistry Mediated by Ti III and Other Single-Electron-Transfer Metals?** Juan M. Cuerva,* Araceli G. Campaæa, JosØ Justicia, Antonio Rosales, JuanL. Oller-López, Rafael Robles, Diego J. Cµrdenas,* Elena Buæuel, and J. Enrique Oltra* The reactivity of water with both carbanion and carbocation intermediates is well known, but until now it has generally been believed that water is inert towards free radicals. [1] This hypothetical passivity has been attributed to the strong HÀ OH bond, which, with a bond-dissociation energy of 117.59 Æ 0.07 kcalmol À1 , [2] would impede any potential hydrogen-atom transfer from water. Some years ago, however, we chanced to observe that tertiary radicals were reduced effectively in the presence of bis(cyclopentadienyl)titanium(III) chloride [3] ([Cp 2 TiCl]) and water. [4] This observation further facilitated the control of the final step in titanocene-catalyzed radical cyclizations, which are useful for the straightforward synthesis of complex polycyclic terpenoids. [5] However, as, at the time, the idea of water acting as a hydrogen-atom source seemed to be counterintuitive, this phenomenon was rationalized by invoking either the formation and subsequent hydrolysis of alkyl–Ti IV complexes or a virtually intramolecular hydrogen transfer via a quite sophisticated cyclic transition state. [4,5] We now have solid evidence to show that water really can act as a complete hydrogen-atom source rather than as a simple proton donor for radical reductions mediated by Ti III and, presumably, other metals that react by single-electron trans- fer. Many highly selective free-radical reactions have been developed and have proved to be very useful in the total synthesis of complex organic compounds owing to the mild conditions required and their compatibility with many func- tional groups. [6] Within this context, RajanBabu and Nugent introduced a novel concept: homolytic oxirane opening induced by [Cp 2 TiCl]. [7] This reaction generates the most substituted (i.e., most stable) b-titanoxy radical, which, among other transformations, could be either trapped by a second [Cp 2 TiCl] species to provide an alkene (epoxide deoxygenation) by “TiO” elimination or, in the presence of a hydrogen-atom donor such as cyclohexa-1,4-diene (1,4- CHD), reduced to an alcohol with the opposite regiochem- istry to that expected from the reduction with metal hydrides (Scheme 1). [7] Since the development of the first titanocene-catalyzed reaction, [8] [Cp 2 TiCl]-mediated processes have become for- midable tools in organic synthesis. [9] However, conventional hydrogen-atom donors used in radical chemistry, such as 1,4- CHD, Bu 3 SnH, and 2-methylpropan-2-thiol, are toxic, expen- sive, and/or foul smelling, thus seriously limiting the applica- tion of the reaction described by RajanBabu and Nugent and other radical reactions for large-scale preparations. [6] There- fore, the discovery of novel and more convenient hydrogen- atom sources seemed desirable. [10] In this context, we decided to explore the possibility of employing safe and inexpensive water as an ideal hydrogen source in Ti III -mediated radical chemistry; titanocene(III)-promoted homolytic epoxide opening was used as a model process for this study. We first examined the reaction between 6,7-epoxyneryl acetate (1) and an excess of [Cp 2 TiCl] (2.5 equiv) in strictly dry THF, that is, under the conditions described by Rajan- Babu and Nugent for epoxide deoxygenation. [7] In this manner we obtained the allylic alcohol 2 (35% yield of isolated 2) together with 3 (29%), the product of 5-exo cyclization, and minor amounts of the reduction product 4 (6%); neryl acetate (7), the expected deoxygenation product, was not detected (Scheme 2). [11] The formation of the major product 2 suggested that a mixed disproportionation had occurred that was slightly Scheme 1. [Cp 2 TiCl]-induced deoxygenation and reductive opening of epoxides by radical chemistry. [7] Scheme 2. [Cp 2 TiCl]-promoted transformation of 1 in strictly dry THF. [*] Dr. J. M. Cuerva, A. G. Campaæa, Dr. J. Justicia, Dr. A. Rosales, Dr. J. L. Oller-López, Dr. R. Robles, Dr. J. E. Oltra Departamento de Química Orgµnica de la Facultad de Ciencias Campus Fuentenueva s/n Universidad de Granada 18071 Granada (Spain) Fax: (+ 34) 958-248-437 E-mail: jmcuerva@platon.ugr.es joltra@ugr.es Dr. D. J. Cµrdenas, Dr. E. Buæuel Departamento de Química Orgµnica C-I. Universidad Autónoma de Madrid Cantoblanco, 28049 Madrid (Spain) Fax: (+ 34) 913-973-966 E-mail: diego.cardenas@uam.es [**] This research was supported by the “Junta de Andalucía” (PAI group FQM339) and the Spanish Ministry of Education and Science (Projects CTQ2005-08402/BQU and CTQ2004-02040/BQU). We also thank our English colleague Dr. J. Trout for revising our text and the “Centro de Computación Científica-UAM” for computation time. Supporting information for this article is available on the WWW under http://www.angewandte.org or from the author. Communications 5522 # 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2006, 45, 5522 –5526