Salem, Maya
(2025)
Computational Design of Highly Stable Single-Atom Alloys.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
Abstract
Metal alloy catalysts are widely used in the chemical industry for their enhanced properties compared to monometallic catalysts. Adding a second metal alters the physicochemical characteristics of the host metal, improving catalytic performance and reducing susceptibility to chemical intermediates. Recently, single-atom alloys (SAAs) have stood out as single-site catalysts for their well-defined active sites. SAAs typically feature a small amount of active metal dispersed on a host metal surface, combining the benefits of traditional heterogeneous and single-atom catalysts. The performance and stability of SAAs depend on the dopant's ability to segregate to the surface without clustering, quantified by segregation energy (Eseg), aggregation energy (Eagg), and dopant diffusion (atom mobility).
This thesis focuses on investigating the thermodynamic and kinetic stability of SAAs under vacuum conditions (non-ligated systems) and in the presence of adsorbates (ligated systems). Research efforts have predominantly focused on select non-ligated SAAs and the impact of common reaction intermediates, such as CO and H, on their stability. First, we examined the segregation behavior across a wide range of SAA combinations, including platinum group metals as hosts and d8 (Ni, Pd, Pt)- and d9 (Ag, Au, Cu)-based SAAs, and facets under vacuum conditions and in the presence of commonly used ligands (amine and thiol groups) in colloidal nanoparticle (NP) synthesis using Density Functional Theory (DFT) and machine learning (ML). We developed models that predict Eseg in ligated and non-ligated SAAs. Second, we investigated the impact of ligands on the aggregation behavior. Furthermore, we developed a robust approach to identify cases where aggregates will form, based on the thermodynamic stability and surface strain features. Using ML techniques, we built a radial basis function kernel support vector regression model to predict the Eagg in both ligated and non-ligated systems. Third, we gained insight into the mobility of dopants in ligated and non-ligated Pt-based SAAs on (100) and (111) surfaces. Finally, we applied a genetic algorithm coupled with the bond-centric model to determine the most thermodynamically stable chemical ordering in multimetallic NPs composed of Ag, Au, Pd, and Pt. Overall, this thesis deepens our understanding of SAA and multimetallic NP stability and offers crucial insights that can accelerate catalyst design, essential for diverse industrial applications.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
7 January 2025 |
| Date Type: |
Publication |
| Defense Date: |
18 September 2024 |
| Approval Date: |
7 January 2025 |
| Submission Date: |
28 September 2024 |
| Access Restriction: |
2 year -- Restrict access to University of Pittsburgh for a period of 2 years. |
| Number of Pages: |
160 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
Swanson School of Engineering > Chemical and Petroleum Engineering |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
Single-Atom Alloys, Machine Learning, Density Functional Theory |
| Date Deposited: |
07 Jan 2025 21:02 |
| Last Modified: |
07 Jan 2025 21:02 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/46986 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[img]](https://d-scholarship.pitt.edu/style/images/fileicons/application_pdf.png)
![[feed]](/images/feed-icon-32x32.png){kind=icon}