1
New Trends in the Design of Metal Nanoparticles and Derived Nanomaterials for Catalysis

Alain Roucoux1and Karine Philippot2

1Université de Toulouse, UPS, INPT, CNRS, LCC (Laboratoire de Chimie de Coordination), UPR 8241, Toulouse Cedex 4, F-31077, France

2Univ Rennes, Ecole Nationale Supérieure de Chimie de Rennes, CNRS, ISCR - UMR 6226, 11, allée de Beaulieu, Rennes, F-35000, France

1.1 Nanocatalysis: Position, Interests, and Perspectives

Being part of heterogeneous catalysts, metal nanoparticles (NPs) have been known for a long time (see for instance the pioneering works of P. Sabatier [1] and L.D. Rampino and F.F. Nord [2]), but a renewed interest has emerged in the past three decades for the design of better-defined systems [3]. Interestingly, a great part of the former heterogeneous catalysis community is now merging into the nanoparticle one [4]. Thus, as it can be observed in the literature through the overincreasing number of papers and patents published by both academic and industrial institutes, huge research efforts are devoted to the synthesis of more precisely defined metal nanospecies and even more recently at an atomic precision level [5-7], as well as the study of their characteristics. This keen interest for nanoscale metal-based species derives from their particular matter state (finely divided metals) and their electronic parameters, thus influencing the physical and chemical properties that these entities possess in comparison with bulk metals and molecular complexes. Small nanoparticles are usually called nanoclusters, but there is a continuum of situations from molecules to solid state between small clusters defined by molecular orbitals and larger nanoparticles defined by energy band structures. These various types differ by the number of metal atoms, the nature of the stabilizing ligands, and the dispersity. Rigorously, the term cluster or nanocluster only concerns molecularly polymetallic assemblies with ligands for which the X-ray crystal structure is known, whereas the term nanoparticles is used for mixtures of more or less polydisperse large nanoclusters defined by the histogram disclosed by transmission electron microscopy (TEM) measurement [4]. Besides the fundamental aspects of the research around metal nanoparticles, the interest is also governed by the potential applications that they may find. The unique properties of nanosized metal particles make them very attractive materials for various domains of applications including optoelectronics, sensing, biomedicine, energy conversion, storage, and catalysis, as nonexhaustive examples [8-11]. Concerning this last field, several books dedicated to nanocatalysis have been edited in the past 15?years [1, 12]. Indeed, metal nanoparticles are particularly interesting catalytic species because of the high surface-to-volume ratio they display. This ratio is promising when metal nanoparticles are at a size as close as 1?nm, or even below (subnanoparticles), because the number of surface atoms can be >90%, thus providing a great number of potential active sites. Remarkably, the subnanometric particles of late transition metals have been reported as very active, although the optimum catalytic activity is believed to be attained with particles containing between 12 and 20 atoms (i.e. sizes close to 1?nm or slightly below) [13]. The atomic sites exposed in metal nanoparticles may have various reactivities owing to their different coordination numbers according to their situation on the surface (corner, edge, and face atoms). By the way, the development of synthesis tools that enable to produce ultrasmall nanoparticles remains of prime importance in order to promote high surface area combined with an efficient distribution of surface atoms. Indeed, the proportion of edge and corner surface atoms possessing lower coordination numbers increases with the decrease in the particle size. Besides the size, other key morphological parameters need to be controlled. Thus, the crystalline structure is also of great importance because according to the types of crystalline plans that can be exposed at the surface, different catalytic properties could be achieved. Controlling the particles shape also constitutes another route to orientate the crystalline plans exposed [14-16]. The last but not the least key parameter is the composition of the metal nanoparticles that has to be adjusted depending on the targeted catalytic reaction. Apart from the nature of the metallic core that may govern the reactivity (some metals are well known for certain catalytic applications but not for others), the surrounding stabilizer for colloidal catalysis (ionic liquids [ILs], polymers, dendrimers, surfactants, polyols, ligands, etc.) or the support for supported catalysis (metal oxides such as silica, metal organic frameworks [MOFs], carbon derivatives, nanocelluloses, etc.) may also influence or even orientate the catalytic performances. While calcination is typically applied in traditional heterogeneous catalysis in order to suppress any organics and liberate the active sites, such treatment on small nanoparticles can be critical, thus potentially leading to their sintering. Moreover, naked nanospecies are not always optimal catalysts. In modern nanocatalysis, the presence of organic ligands onto the particle surface is not considered as detrimental for catalytic applications but could be a way to improve or even modify the chemoselectivity [17]. Aiming at catalysis by appropriate design, current developments in nanochemistry dedicated to catalytic applications often rely with multifunctionality [18]. This can be achieved by the proper design of nanohybrids, the term hybrid referring to the appropriate association between a metal core and a stabilizing shell or a support. When using typical ligands from coordination chemistry as capping agents, a parallel can be performed with molecular catalysis. Indeed, the interaction of the ligands with the metal atoms on the particle surface can be compared to ligand interactions with the metal centers in homogeneous complexes and is a parameter of paramount importance for stability and catalytic performances (activity and selectivity properties). Thus, ligands can be chosen in order to tune the surface properties of metal nanoparticles through steric and/or electronic effects [19, 20]. The challenge remains to find protective agents that are able to stabilize well-defined nanoparticles while controlling accessibility at the metal surface and reactivity [13, 21]. Strongly bound capping ligands (such as thiols or phosphines) can result in the poisoning of a nanocatalyst at high surface coverage. However, the presence of a limited amount of ligand can be beneficial in terms of catalytic performances (suitable and replicable activities and/or selectivities). The strong coordination of a ligand at a metal surface can also be a way to block selectively some active sites in order to orientate the catalysis evolution. In comparison with the investigation of facet dependency [12, 22], the ligand influence on the catalytic activity has been less intensively studied, but recent results well illustrate the interest to do so [23-27]. Capping agent-stabilized metal nanoparticles can be applied to catalysis as stable colloidal suspensions in various media (water, polyols, and organic solvents) but also in heterogeneous conditions after their deposition on the surface or confinement in the pores of solid supports [28]. Ionic liquids [29] are also very efficient to stabilize metal nanoparticles, and their colloidal suspensions can even be deposited onto inorganic materials [30]. When employing a support, it not only prevents nanoparticle aggregation but may also act in synergy with nanoparticle surface and favor the activation of substrates in a manner comparable to the positive synergy observed between two transition metal atoms in alloys or between a transition metal and a main group element such as nitrogen. Therefore, it was found that N-doped carbon supports were superior to undoped analogs because of such synergy effects [4].

Thus, having in hand synthesis strategies that allow access, in a reproducible manner, to well-defined metal nanoparticles in terms of size, crystalline structure, composition (metal cores and stabilizing agents), chemical order (bimetallic or ternary systems), shape, and dispersion constitutes a beneficial condition to finely investigate the catalytic properties of metal nanoparticles and define structure/properties relationships. Taking advantage of recent developments in nanochemistry that offer efficient tools to reach these objectives, nanocatalysis is now well established as a borderline domain between homogeneous and heterogeneous catalysis. With a molecular approach, nanocatalysts can be considered as assemblies of individual active sites where metal-metal bonds will also have influence [31]. Precisely designed nanohybrids (including the choice of an adequate and noninnocent stabilizer or support) are expected to present benefits from both homogeneous and heterogeneous catalysts, namely, high reactivity and better selectivity [32]. One aim lies in the design of more performant nanocatalysts in order to develop more efficient and eco-compatible chemical production...