Ammonia Synthesis on Proton-enriched Palladium Substrate

Production of Ammonia is one of the most important in the chemical industry; most of the Ammonia produced is then converted to other Nitrogen-containing molecules, such as anhydrous ammonium nitrate or urea, which are typically used for inhorganic fertilizers. The standard industrial reaction for the production of Ammonia is the Haber-Bosch Process (sec. 1.1); an high-pressure and high-temperature process known and developed since the early twentieth century [31, 11]. The Haber-Bosch process produces Ammonia directly from its components, Nitrogen and Hydrogen, supplied in gas form. Although successful, this industrial process, consumes a large amount of energy which, given the huge production volume, employs a significant percentage of total world energy production and natural gas consuption [70]. Because of its economical importance, improving the Ammonia production process is still an open challenge and a rich field of research. A search in related journal archives for papers regarding Ammonia production typically returns hundreds or thousands of results; nevertheless there is still room for improvement and better understanding of the reaction. A part of the studies have been focused on improving the Haber-Bosch process itself; e.g. replacing the original Iron catalyst with more efficient metal alloys. However other studies have tried to find novel reaction mechanisms, that could possible overcome the high-pressure and high-temperature requirements. Some studies have been based on certain biological process, occuring in bacteria, that can produce Ammonia at room temperature and pressure [77, 68, 24]. In the biological process the energy required by the reaction is not supplied thermally but by reduction of adenosine triphosphate (ATP) molecules, which makes them dicult to reproduce in vitro. Another possible approach is to supply the required energy electrochemically, by applying an external electric eld to a solid-state catalyst, or having a certain amount of current circulating through it, it is possible to transfer to the reactants the necessary amount of energy. The electrical source can also act in an indirect way, modifying the nano-scale properties of the catalyst in a way that improves its properties; this eect is called non- Faradayc Electrochemical Modification of Catalytic Reaction (NEMCA) [56, 80]. In 1998 Stoukides and coworkers [50, 49] (sec. 1.2)demonstrated the possibility of producing Ammonia at atmospheric pressure by supplying the required amount of Hydrogen via a proton-conducting perovskite. The perovskite, in the form of a little brick or a pipe, is coated on two sides with a Palladium paste, obtaining two catalytic surfaces separated by a proton conductor. The two catalysts are then connected through an electrical circuit and a certain bias is applied, making them act as an anode and a cathode. When the anode is exposed to molecular Hydrogen, H2 dissociates spontaneously. Because of the bias, the Hydrogen atoms are stripped of their electron and forced to cross the proton conductor; the electron will instead travel through the external electrical circuit; recombination of protons and hydrogens happens in the cathode. The cathode itself is exposed to Nitrogen gas. Combining this electrochemical mechanism with high temperature (between 500 C and 750 C) the author were able to induce a steady ux of Ammonia, with a high conversion efficiency. They also proposed the presence of a NEMCA effect, although weak. In the present work we will re-examine the original experiment and tackle its characteristic and inner mechanics at the nano-scale level. We will use ab-initio methods (chap. 2) to construct a model of the catalytic process and simulate its intermediate steps. In our study we have used the computational tools provided by the Quantum-espresso distribution [28]. In particular we will use Density Functional Theory (sec. 2.2) and the Projector-Augmented Wave method (PAW, sec. 2.3) to reproduce the electronic structure of the system; the Born-Oppenheimer approximation, together with the Hellmann-Feynman theorem, will be implicitly used to optimize the nanoscopic structure, and find intermediate steps of the reactions. We will also used the Transition State Theory (sec. 3.3) and the Nudged- Elastic Band method (sec. 3.4) to estimate the energy barriers involved in the reaction and, consequently, the reaction rate. In chapter 4 we will examine the catalyst structure in detail; in particular we will focus on the effect of active Hydrogen pumping by mean of the applied cell potential. In section 4.1 we will see how a cell potential of realistic amplitude can force a very large amount of Hydrogen in the Palladium bulk. The resulting system, called Palladium hydride, can undergo a phase transition that changes its unit-cell volume up to 10%; we will see the details of this phase change and the possibly resulting structures. In section 4.2 we will move our focus to the catalyst surface. We will tackle the problem of determining the adsorbed Hydrogen population, in the case of normal Palaldium and Palladium hydride. This complex problem involves a three-phase equilibrium, where the chemical potentials of Hydrogen in the gas and Hydrogen adsorbed in the bulk or on the surface have to be equal. In order to estimate the chemical potential we will use a Monte Carlo simulation built on top of a simplified model where the total energy is computed as a sum of adsorption on-site energies and neighbour-site interactions. Finally, in chapter 5, we will tackle the core of the problem, trying to find a suitable reaction path for the Ammonia production. We will examine the possibility of Nitrogen dissociative adsorption as well as the possiblity of Nitrogen hydrogenation prior to its dissociation. The former will be easily proved impossible, at the experimental conditions, so we will focus on the subsequent hydrogenations of the N2 molecules. We will examine the process up to the final breaking of the N{N bond, where the formation of Ammonia can proceed without further barriers. An order of magnitude estimate of the Ammonia production in the system will be made and found to be compatible with the experimental findings.