PhD Oral Exam - Mehdi Shamekhi, Physics

When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.

Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.

Abstract

The reliance of essential chemical production on fossil-driven, high-temperature processes remains a critical barrier to achieving sustainable and low-carbon chemical manufacturing. Ammonia (NH3) is a cornerstone of modern society, a key ingredient for fertilizer production, food security, and emerging applications as a carbon-free energy carrier. However, NH3 conventional industrial synthesis through the Haber–Bosch process remains highly energy-intensive and environmentally unsustainable. The electrochemical nitrogen reduction reaction (NRR) offers a sustainable alternative under ambient conditions but faces two major challenges: low catalytic activity due to weak N2 adsorption and low Faradaic efficiency due to the competing hydrogen evolution reaction (HER).

This thesis develops a comprehensive computational framework that integrates density functional theory (DFT), grand canonical ensemble DFT (GCE-DFT), and machine learning algorithms (MLAs) to accelerate catalyst discovery and elucidate mechanistic pathways of the NRR. The Ag3Sn alloy was first investigated as a promising catalyst, combining DFT calculations with experimental characterization to investigate its catalytic activity. In the next step, a dataset of bimetallic alloys was generated using DFT, and artificial neural networks were trained on descriptors of the transition-metal d-state electronic structure. The resulting ANN achieved accuracy comparable to DFT while enabling rapid screening of hundreds of candidate surfaces. From this screening, Au@Au3Re and Au@Au3Mo emerged as promising catalysts, and their reaction free-energy profiles were calculated with DFT to identify potential-limiting steps and theoretical overpotentials. Charge analysis further revealed substantial electron transfer between constituent elements, providing mechanistic explanations for their enhanced catalytic activity.

To consider one of the most important factors in an electrochemical setup, namely the electrode potential, GCE-DFT was employed. Unlike conventional DFT, which is constrained to constant-charge conditions and therefore cannot directly capture the influence of an applied potential, GCE-DFT enables simulations at constant electrochemical potential by allowing the number of electrons in the system to fluctuate. This constant-potential framework provides a direct mapping between electrode potential and adsorption or reaction energetics, thereby bridging the gap between theory and experimental control via a potentiostat. Using this method, the competition between NRR and the HER was investigated in detail. These findings demonstrate that the electrode potential influences more than just the reaction thermodynamics; it also alters the expected selectivity, challenging the predictions made by conventional DFT calculations.

Altogether, this thesis establishes a rigorous computational approach, combining MLA screening, DFT characterization, and GCE-DFT potential-dependent simulations that advance the rational design of active and selective NRR electrocatalysts. The findings contribute not only to the development of sustainable NH3 synthesis but also provide generalizable strategies for understanding electrochemical reactions where electrode potential and competing pathways play decisive roles.