Display Accessibility Tools

Accessibility Tools

Grayscale

Highlight Links

Change Contrast

Increase Text Size

Increase Letter Spacing

Readability Bar

Dyslexia Friendly Font

Increase Cursor Size

James Shepherd

header

Research

The Shepherd Group is focused on designing, creating, and testing new computational approaches with a focus on electron-electron interactions in metals and in high-temperature chemistry. 

Our applications of interest include bond breaking in small molecules, low-gap materials such as metals and semiconductors, surface reactions with heterogeneous catalysts, and photoactive molecules. We have a number of active collaborations with organic and inorganic chemists.  

Quantum methods for materials design 

This research focuses on developing high-accuracy, fully-quantum methods to model the complex electronic structures of novel materials, which are essential for rational materials design. Current methods like coupled cluster theory and full configuration interaction quantum Monte Carlo offer high accuracy but are computationally expensive and limited by finite size and basis set errors. The project aims to overcome these limitations by developing algorithms to efficiently remove these errors, making these methods more accessible for routine use. This will enable chemists and materials scientists to accurately model phase transitions, binding energies, and other phenomena, accelerating the design of new materials such as semiconductors and catalysts. The research also seeks to improve the scalability of wavefunction-based methods to support the modeling of complex interactions in solids, ultimately providing tools that can be used alongside density functional theory. The outcomes are expected to significantly impact the rational design of complex materials by providing reliable, high-accuracy computational tools. 

red and white balls

Tina Mihm, the first graduate student in the group, designed an algorithm that sped up calculations of solids by 100 times (From Nat. Comp. Sci. 2021, 1, 801–808, cover article). 

Quantum density matrix approaches for high temperature electrons 

The project focuses on developing advanced electronic structure methods to model electron-electron interactions at non-zero electronic temperatures, crucial for understanding various applications in basic energy sciences. Traditional methods struggle with accurately predicting behaviors at finite temperatures, especially in systems with strong interactions. While electronic temperature is typically negligible at room temperature, as most systems remain in their ground states, there are specific scenarios where it becomes significant. For instance, temperature plays a crucial role in the conductivity of semiconductors and in phase transitions, such as melting. A more exotic and challenging area where electronic temperature matters is in the field of warm dense matter (WDM). WDM is a state of matter characterized by temperatures ranging from several thousand to hundreds of thousands of kelvins, existing between everyday condensed states and plasmas. This state is found in giant planets, small stars, inertial confinement fusion energy experiments, and laser-induced processes. Our research focuses on developing and applying density matrix quantum Monte Carlo (DMQMC) methods to provide highly accurate calculations for these finite-temperature systems. We aim to create source code, establish benchmarks, and perform quantum Monte Carlo simulations on representative systems. By advancing these methods, we aim to provide the scientific community with novel, open-source tools to study and benchmark electronic temperature effects enhancing our understanding of complex phenomena in basic energy sciences. 

 


Contact / Webpage

Area(s) of Interest

Theory (Th)

Inorganic (In)

Chemical Physics (CP)

Material (Ma)

Physical (Ph)

Organometallics (Om)


Publications

 A shortcut to the thermodynamic limit for quantum many-body calculations of metals. Tina N. Mihm, Tobias Schäfer, Sai Kumar Ramadugu, Laura Weiler, Andreas Grüneis & James J. Shepherd. Nature Computational Science, 1, 801 (2021)

Using Density Matrix Quantum Monte Carlo for Calculating Exact-on-Average Energies for ab Initio Hamiltonians in a Finite Basis Set. William Z. Van Benschoten, Hayley R. Petras and James J. Shepherd. J. Phys. Chem. A 127, 32, 6842 (2023)

CV

2024 – Present: Associate Professor, Chemistry, Michigan State University 2023: Associate Professor, Chemistry, University of Iowa
2017 – 2023: Assistant Professor, Chemistry, University of Iowa 
2015 – 2017: Postdoctoral Fellow, Massachusetts Institute of Technology (Troy Van Voorhis)
2013 – 2015: Postdoctoral Fellow, Rice University, Chemistry (Gustavo Scuseria)
2013: PhD, Chemistry, University of Cambridge (Ali Alavi), Thesis: A Quantum Chemical Perspective on the Homogeneous Electron Gas
2009: BA/MSci, Chemistry, University of Cambridge

Awards

Year Award Organization

CAREER Award 

2021

National Science Foundation

Early Career Research Program

2020

Department of Energy