DOI: 10.1002/chem.201301970 One Site Is Enough: A Theoretical Investigation of Iron-Catalyzed Dehydrogenation of Formic Acid Rocío Sµnchez-de-Armas, Liqin Xue, and Mårten S. G. Ahlquist* [a] We present a study of the reaction mechanism of an iron- based catalyst for formic acid dehydrogenation. Based on these results we propose a reaction mechanism with a neu- tral iron complex, where all the reaction steps take place at one reaction site of the catalyst. One of the hydrides acts as a spectator ligand and it should be possible to replace it to alter and improve the reactivity of the catalyst. Storage of renewable energy, for example, solar energy, in chemical bonds could allow for the use of renewable energy at any place and time. As a clean energy carrier, hydrogen has attracted much attention. However, hydrogen is a very light gas and storing it and transporting it could be hazard- ous. Formic acid (FA) has therefore been proposed as an al- ternative sustainable material for hydrogen storage. [1] The decomposition of FA can occur through two main reaction pathways, namely, dehydrogenation [Eq. (1)] generating H 2 and CO 2 , and dehydration [Eq. (2)] forming CO and H 2 O. Both dehydrogenation and dehydration reactions are ther- modynamically favorable by 7.8 and 3.0 kcal mol 1 , re- spectively. [2] In terms of hydrogen production, selective dehydrogena- tion without dehydration is essential, because CO, which is generated in the reaction, is a poison for most fuel cells. [3] In recent years, there have been a number of reports on the se- lective decomposition of FA by using both homogeneous and heterogeneous catalysts to produce H 2 . However, many of these catalysts are based on expensive low-abundant noble metals, such as ruthenium, [4, 5] rhodium, [5] heterodinu- clear iridium–ruthenium, [6] or platinum centers. [7] For exam- ple, Ru-catalyzed conversion of a MeOH–H 2 O mixture into H 2 and CO 2 , a process that also includes a stepwise conver- sion from FA to H 2 and CO 2 , has been reported recently. [8] Furthermore, the reverse reaction, that is, hydrogenation of CO 2 to FA, catalyzed by Ru [9] or Ir [10] complexes has also been investigated by theoretical methods. Exploitation of catalysts based on non-noble metals with high efficiency for FA decomposition is still required. In nature, there are enzymes containing abundant first-row transition metals (iron or nickel) in their active sites; these enzymes are able to catalyze the production or oxidation of hydrogen in many organisms. [11–13] Inspired by nature, simple iron carbonyl phosphine complexes for dehydrogenation of FA were reported by the research group of Beller in 2010. [14] Very recently, a well-defined iron–phosphine catalyst system capable of dehydrogenating FA to H 2 and CO 2 with unex- pectedly high activity has been reported [Eq. (3)]. [15] The catalytic system contains an iron (Fe II ) cation coordi- nated by a tetradentate phosphine ligand, a complex that is prepared by adding 0.005 mole percent of FeACHTUNGTRENNUNG(BF 4 ) 2 ·6H 2 O and tris[(2-diphenylphosphino)ethyl]phosphine (PP 3 ) to a solution of FA in an environmentally benign propylene car- bonate medium. In addition, this well-defined iron catalyst is also capable of reducing carbon dioxide to formates, that is, the reverse process of dehydrogenation of FA, a process that is of importance in the benign use of carbon dioxide as a hydrogen-storage material. [16] Based on the results of in situ nuclear magnetic resonance spectroscopy, kinetic studies, and DFT calculations, two dif- ferent catalytic cycles (Figure 1) for the dehydrogenation of FA [Eq. (3)] have been proposed by Beller et al. [15] Both competing cycles I and II start with the same precursor com- plex [FeHACHTUNGTRENNUNG(PP 3 )] + (1). For cycle I, the Fe-hydride complex 1 reacts with a proton from FA to form a hydrogen molecule followed by binding of the formate anion to the Fe metal center to generate complex 2 ; then b-hydride elimination from complex 2 yields complex 3, which eliminates CO 2 to regenerate precursor complex 1 to complete the catalytic cycle. On the other hand, cycle II involves coordination of a formate anion to complex 1 to give complex 4, subsequent b-hydride elimination followed by protonation of the hy- dride and release of CO 2 to form Bianchini)s complex [FeH- ACHTUNGTRENNUNG(h 2 -H 2 )ACHTUNGTRENNUNG(PP 3 )] + (5), [17] and elimination of H 2 to complete the catalytic cycle. Based on their calculated results, Beller et al. [a] Dr. R. Sµnchez-de-Armas, + Dr. L. Xue, + Dr. M. S. G. Ahlquist Division of Theoretical Chemistry & Biology School of Biotechnology, KTH Royal Institute of Technology 10691 Stockholm (Sweden) Fax: (+ 46) 8-5537-8590 E-mail : mahlquist@theochem.kth.se [ + ] These authors made equal contributions to this work. Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201301970. Chem. Eur. J. 2013, 19, 11869 – 11873 # 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 11869 COMMUNICATION