This journal is © the Owner Societies 2014 Phys. Chem. Chem. Phys. Cite this: DOI: 10.1039/c4cp01527d CO dissociation on magnetic Fe n clusters Abdesslem Jedidi, ab Alexis Markovits,* b Christian Minot, b Manef Abderrabba c and Michel A. Van Hove d This work theoretically investigates the CO dissociation on Fe n nanoparticles, for n in the range of 1–65, focusing on size dependence in the context of the initial step of the Fischer–Tropsch reaction. CO adsorbs molecularly through its C-end on a triangular facet of the nanoparticle. Dissociation becomes easier when the cluster size increases. Then, the C atom is bonded to a square facet that is generated as a result of the adsorption if it does not yet exist in the bare cluster, while the O atom is adsorbed on a triangular facet. In the most stable situation, the two adsorbed atoms remain close together, both having in common one shared first-neighbor iron atom. There is a partial spin quenching of the neighboring Fe atoms, which become more positively charged than the other Fe atoms. The shared surface iron atom resembles a metal-cation from a complex. Despite the small size of the iron cluster considered, fluctuations due to specific configurations do not influence properties for n 4 25 and global trends seem significant. 1. Introduction Since its discovery in 1926, the Fischer–Tropsch synthesis (FT) 1 has been an icon for catalysis. It produces hydrocarbons (synthetic fuel) from coal, natural gas, or biomass. In addition to being of fundamental interest as a typical catalytic reaction, FT nowadays has also gained renewed industrial interest. In periods where crude oil was abundant and cheap, the practical interest in FT was limited to countries suffering from an oil blockade; nowadays, the growing oil prices have renewed interest in the usefulness of FT. Since it produces very clean fuels with high-quality heavy hydrocarbons, 2–4 it has special importance for the aerospace and high technology industries. 5–7 Motivations for studying the mechanism of FT are the under- standing of a typical catalytic reaction involving a transition metal and the possibility of improving an industrial process of primary importance. FT takes place on transition metals by forming surface- bound metal carbonyls. The proposed mechanism starts with CO dissociation forming carbon monomers and possibly oxide and carbide ligands. 8,9 The complete conversion of CO involves hydrogenolysis of carbon fragments, formation of C–C bonds and water, and desorption. There is a controversy regarding whether CO dissociates before or after the addition of one or more hydrogen atoms (hydrogenolysis). 10,11 Mohammad Reza Elahifard et al. 12 showed that direct CO dissociation is preferred on the Fe(100) surface but not on the Fe(110) surface. Recently, FT reaction starting from CO insertion to long chain hydrocarbon formation was extensively studied from the kinetic and micro- kinetic point of view by R. A. van Santen et al. 8,13 A systematic study on a series of transition metals supported on TiO 2 has shown different mechanisms depending on the nature of the metal; however, on the most commonly used transition metal clusters, Co, Fe and Ru, the mechanism initiated by CO dissocia- tion is likely. 14 Niemantsverdriet et al. 15 have experimentally studied a spherical iron nanoparticle model of well-defined size, showing that the model exhibited higher activity than other techniques. This led to controlled size studies of iron-based nano-catalysts in FT synthesis. The main goal of this study is to unravel the general features of CO dissociation on Fe n nanoparticles, with n between 1 and 65, focusing on the size dependence 16 of cluster properties and reactivity. 17 This nanoparticle size remains within the nano- scale domain. However, results may link the quantum domain of few atoms to the infinite flat crystal surface that has been studied by many authors. 18–24 According to Woodruff, ‘‘the non-scalable size range’’ extends to about N = 65, while the ‘‘scalable size range’’ of nanoparticles starts beyond this size. 25 In a previous study devoted to Fe 13 , 26 our group has compared several methods of calculation and concluded that DFT + U, a Division of Physical Sciences and Engineering, KAUST Catalysis Center, Modeling group, King Abdullah University of Science and Technology (KAUST), 4700 KAUST, Thuwal 23955-6900, Saudi Arabia b Sorbonne Universite ´s, UPMC Univ. Paris 06, UMR 7616, Laboratoire de Chimie The ´orique, case 137, 4 place Jussieu, F-75005, Paris, France. E-mail: alexis.markovits@upmc.fr; Fax: +33(0)144274117; Tel: +33(0)144272682 c Laboratoire de Mate ´riaux, Mole ´cules et Applications (LMMA), Universite ´ de Carthage, Boite postale BP51, 2070 La Marsa, Tunisia d Institute of Computational and Theoretical Studies, and Department of Physics, Hong Kong Baptist University, 224 Waterloo Road, Kowloon, Hong Kong Received 8th April 2014, Accepted 4th August 2014 DOI: 10.1039/c4cp01527d www.rsc.org/pccp PCCP PAPER Published on 07 August 2014. Downloaded by King Abdullah Univ of Science and Technology on 28/08/2014 07:32:48. View Article Online View Journal