Role of Interfacial Energy and Crystallographic Orientation on the Mechanism of the ZnO + Al 2 O 3 → ZnAl 2 O 4 Solid-State Reaction: I. Reactivity of Films Deposited onto the Sapphire (110) and (012) Faces Sonia Pin, †,‡ Marco Suardelli, ‡ Francesco D’Acapito, ∥ Giorgio Spinolo, ‡ Michele Zema, § Serena C. Tarantino, § and Paolo Ghigna* ,‡ † General Energy Research (ENE), Laboratory for Bionenergy and Catalysis, Paul Scherrer Institut, 5232 Villigen PSI, Switzerland ‡ INSTM, Department of Chemistry and § Department of Earth and Environment Sciences, University of Pavia, I27100 Pavia, Italy ∥ CNR-IOM-OGG c/o ESRF, GILDA-CRG, BP 220, F38043 Grenoble Cedex, France * S Supporting Information ABSTRACT: The initial steps of the reaction between ZnO and Al 2 O 3 have been investigated with X-ray diffraction, atomic force microscopy, and X-ray absorption spectroscopy at the Zn−K edge starting from 45 nm thick zincite films deposited onto (110)- and (102)-oriented sapphire single crystals. The formation of nonequilibrium phase(s) has been detected for both orientations. For the (001) zincite ∥ (110) sapphire interface, the rate-determining step is the motion of the interface(s); the growth of the spinel layer is linear with time, with a rate constant k = 1.1(2) × 10 −9 cms −1 at 1000 °C. At the (110) zincite ∥ (012) sapphire interface, the reaction shows dumped oscillations. The results are discussed along with a comparison with previous results on thinner films to clarify the role of interfacial free energy and crystallographic orientation. ■ INTRODUCTION This work aims at contributing to the understanding of the initial steps of solid−solid heterogeneous reactions. For these reactions, the well-established approach 1−6 provides a sound description of the growth of a product phase spatially placed between two regions occupied by the reagent phases, that is, the processes occurring when some amount of the product phase already exists. Much less known are the initial steps, when the product starts to form and the single interface between the two reactants turns into a couple of interfaces between the product layer and each of the two reagent phases. These steps are obviously controlled by more factors than free nucleation in a fluid medium: a favorable relation of translational lattices in addition to similarity in the atomic arrangements is expected, for instance, to favor nucleation by lowering the reactant/product interfacial energy. For exploring the initial steps of a solid-state reaction, we have recently suggested and used 7−12 an experimental protocol that essentially investigates with a multi-technique approach the time evolution of model reactive systems made of a thin layer of one reagent deposited onto a single crystal slab of the other reagent and compares the results obtained with films deposited onto different crystal orientations and with different thick- nesses. Several experimental setups have indeed devised for exploring mechanistic aspects of solid−solid heterogeneous reactions. 13,14 A very effective example is the reactive system based on the single crystal of one reagent with vapors of the other reagent, 15−20 but also the reactive system based on the combination of a single crystal and one or more deposited films is not new in the scientific literature on solid-state reactions. Typically, this setup has been used to characterize the growth of the product layer under control by interface mobility or the transition to diffusion control, less frequently to investigate product nucleation. 21−29 Considering in addition various film thicknesses appears very challenging because it allows us to control the final amount and thickness of the product phase and to tune the contribution of interfacial free enthalpy to the overall driving force of the reaction. Then, the growth of the product can be stopped at values corresponding to the initial steps of a massive reaction, and it is at least in principle possible to explore various reaction regimes and to enhance or depress the role of di fferent mechanistic steps. In particular, intermediate compounds are possibly stabilized in some experimental setups by the surface energy, and it becomes possible to detect and clarify the role of intermediate phases that are otherwise impossible to study or do not occur at all. Received: December 19, 2012 Revised: January 28, 2013 Published: January 30, 2013 Article pubs.acs.org/JPCC © 2013 American Chemical Society 6105 dx.doi.org/10.1021/jp3124956 | J. Phys. Chem. C 2013, 117, 6105−6112