FHI Theory Department

Theory Department of the Fritz Haber Institute of the Max Planck Society @fhi_mpg_de. We focus on quantitative modeling of materials properties and functions, in particular on processes in working catalysts and energy conversion devices. #machinelearning #physics #chemistry #computingsciences #materialsscience #engineering

2025-01-06

3️⃣ What are the implications?

The findings suggest that numerous low-barrier processes involving collective atomic motion enable dynamic restructuring of the Pd step edge, highlighting a fluxionality aspect in oxidation catalysis that could influence the operando evolution of catalytic interfaces.

2025-01-06

2️⃣ What is new?

The paper presents a systematic exploration that generates nearly 3000 unique elementary processes, revealing a level of complexity far exceeding current microkinetic modeling capabilities.

2025-01-06

1️⃣ What is the paper about?

The paper combines automatic process exploration with an iteratively trained machine-learning interatomic potential to identify elementary processes involved in the initial oxidation of a Pd step edge.

2025-01-06

👏 Congratulations to Patricia P., Alex, Francesco, @christophscheurer and Sebastian, whose paper "ML-Accelerated Automatic Process Exploration Reveals Facile O-Induced Pd Step-Edge Restructuring on Catalytic Time Scales" made it into ACS Catalysis. 👏

If you want to know more, have a look at their paper online or read the paper digest below!

pubs.acs.org/doi/10.1021/acsca

2024-12-17

🎄 Last Friday we had our much awaited Xmas Party. It was a great occasion to celebrate the past incredible year and all the achievements we have accomplished together. 🥳

As the holiday season approaches, we wish everyone to relax, recharge, and enjoy the company of loved ones. ☃️ 🎁

Already looking forward to be spending a fun 2025 together, full of great science, exciting challenges and new opportunities!

2024-12-16

👏 Congratulations to Alex D., Elias, and Vanessa, whose work "Oxygen Adsorption at the Electrochemical Metal/Water Interface: Au(111) vs Pt(111)" made it into The Journal of Physical Chemistry C! 👏

The paper is already online, so if you want to know more, have a look at it 👇

pubs.acs.org/doi/10.1021/acs.j

FHI Theory Department boosted:

Herzlichen Glückwunsch zum #LeibnizPreis! 🏆 👏
10 Wissenschaftler*innen gebührt große Ehre. Und sie erhalten je 2,5 Mio. Euro für ihre #Forschung. Die Verleihung feiern wir am 19. März in #Berlin – zusammen mit vielen bisherigen Preisträger*innen aus #40JahreGWL. Das wird ein Fest!

Die Leibniz-Preisträger*innen 2025 im Kurzporträt:
➡️ dfg.de/de/service/presse/press

2024-12-11

3️⃣ What are the implications?

The findings enhance the understanding of degradation mechanisms at the anode side of SOECs, which is crucial for developing strategies to improve the lifetime and performance of these cells. The study emphasizes the need for atomic-scale understanding to design more durable and efficient energy conversion systems.

2024-12-11

2️⃣ What´s new?

The study provides insights into the formation and growth of mixed ion electron conductive (MIEC) phases and secondary structures at the LSM/YSZ interface after prolonged operation at high temperatures. It combines chemical electron microscopy with theoretical modeling to reveal local phenomena that occur within the same sample, highlighting the presence of nano-scale complexions with improved oxygen diffusivity.

2024-12-11

1️⃣ What is the paper about?

The paper investigates the interfacial structures in solid oxide cells (SOCs), particularly focusing on the solid-solid interfaces hidden between two solids. It explores how these interfaces evolve under high temperatures and electrochemical conditions, affecting the performance and degradation of solid oxide fuel and electrolysis cells (SOFCs, SOECs).

2024-12-11

👏 Congratulations to Hanna, Yu-Te, and @christophscheurer whose work "Boon and Bane of Local Solid State Chemistry on the Performance of LSM-based Solid Oxide Electrolysis Cells" made it into Advanced Energy Materials! 👏

Shout-out also to our collaborators from the Inorganic Chemistry Department at the @fhi_mpg_de and from @fzj! The project was carried out within the @dfg_public Priority Program SPP 2080.

While we wait for their paper to be online, have a look at the paper digest below 👇

2024-12-10

3️⃣ What are the implications?

The implications of this research include improved accuracy in segmenting ESEM data, leading to better insights into phase transitions during catalytic reactions.

This approach reduces false positives compared to random data generation methods and enhances the understanding of catalyst behavior, potentially improving catalyst design and performance.

2024-12-10

2️⃣ What´s new?

The novel aspect of this research is the substitution of expert-annotated data with a physics-based sequential synthetic data model. This model generates synthetic ESEM data by simulating crack formation on the catalyst surface, adhering to physical principles, which helps in training neural networks for semantic segmentation.

2024-12-10

1️⃣ What is the paper about?

The paper discusses the development of #machinelearning models for segmenting microscopy data in catalysis research. It focuses on using a physics-based synthetic data model to generate training data for neural networks, specifically for analyzing environmental scanning electron microscopy (ESEM) data related to isopropanol oxidation over cobalt oxide.

2024-12-10

👏 Congratulations to Maurits, Gianmarco, and @christophscheurer whose paper "Physics-Based Synthetic Data Model for Automated Segmentation in Catalysis Microscopy" made it into Microscopy and Microanalysis.👏

Shout-out also to our collaborators from the Inorganic Chemistry Department at the @fhi_mpg_de Luis and Thomas, and Markus from @tu_muenchen!

While we wait for their paper to be online, have a look at their preprint on @chemrxiv chemrxiv.org/engage/chemrxiv/a or read the paper digest below 👇

2024-12-09

3️⃣ What are the implications?

The findings provide a new atomistic perspective on the relationship between OER activity and the durability of precious metal oxide catalysts. This research could guide the development of more efficient and stable anode materials, advancing PEM water electrolysis technology and supporting the transition to a low-carbon economy by improving hydrogen production efficiency.

2024-12-09

2️⃣ What´s new?

The study introduces a novel surface H-terminated nanosheet model that better represents the short-range structure of am-hydr-IrOx, revealing elongated Ir-O bond lengths compared to traditional crystalline models. It identifies iridium dissolution as a spontaneous, thermodynamically driven process occurring at lower potentials than previously thought. This challenges the conventional understanding and suggests a new dual-mechanistic framework for OER and iridium dissolution.

2024-12-09

1️⃣ What is the paper about?

The paper investigates the mechanisms of catalyst activation and degradation during the Oxygen Evolution Reaction (OER) in hydrous iridium oxides, specifically focusing on amorphous hydrous iridium oxide (am-hydr-IrOx). It aims to enhance the understanding of these processes to improve the efficiency and stability of anode materials in Proton Exchange Membrane (PEM) water electrolysis, a key technology for sustainable hydrogen production.

2024-12-09

📣 Collaborative research by the @hzbde and our Institute has provided insights into the mechanisms of OER performance and iridium dissolution for amorphous hydrous iridium oxides, advancing the understanding of this critical process. 📣

If you are curios to know more, have a look at our press here: release fhi.mpg.de/1654372/2024-12-09-, or read the paper digest below 👇

2024-12-06

📣 We are ready to start the ELLIS workshop "ML for molecules and materials in the era of LLMs", organized by @margraf.

A full day of talks and discussions awaits us, ranging from ML and LLMs to molecular sciences!

Have a look at the program here: moleculediscovery.github.io/wo 📣

Client Info

Server: https://mastodon.social
Version: 2025.04
Repository: https://github.com/cyevgeniy/lmst