From ba9ed4c3889245af8459375a113e66686a32515b Mon Sep 17 00:00:00 2001 From: "J. George" Date: Tue, 4 Aug 2026 15:31:11 +0200 Subject: [PATCH 1/7] Revise research.md for clarity and detail Updated the research page to reflect a more comprehensive overview of materials design efforts, including new themes and contributions to open-source software. Added sections on machine learning potentials and open-source research infrastructures. --- _pages/research.md | 36 +++++++++++++++++++++++++++--------- 1 file changed, 27 insertions(+), 9 deletions(-) diff --git a/_pages/research.md b/_pages/research.md index ab2b831..4c779cc 100644 --- a/_pages/research.md +++ b/_pages/research.md @@ -7,23 +7,41 @@ nav: true nav_order: 3 --- -We work on research projects that aim at an accelerated materials design at BAM and the University of Jena. We primarily focus on the safety and sustainability of materials. Together with our colleagues at BAM, we aim at combining our efforts with [Materials Acceleration Platforms](https://www.bam.de/Navigation/EN/Topics/Materials/Materials-design/materials-design.html) (i.e., automated platforms for materials searches/optimizations). Additionally, [chemically complex materials](https://www.bam.de/Navigation/EN/Topics/Materials/Materials-design/materials-design.html) are one of our joint interests with colleagues at BAM. -#### New chemical heuristics +We accelerate materials design with computational and data-driven approaches at BAM and the University of Jena. In particular, we focus on the design of safe, sustainable, and reliable materials. To achieve this, we combine first-principles simulations, condensed matter physics, materials informatics, machine learning, and high-throughput computational methods. -The group works to test and develop chemical heuristics (intuitive rules) to advance knowledge and understanding in solid-state chemistry and physics, using geometric and quantum chemical descriptors for chemical environments. For the latter, the group is developing open-source tools for automated bond analysis ([LobsterPy](https://github.com/JaGeo/LobsterPy), [atomate2](https://github.com/materialsproject/atomate2)) +Our research is organized around two complementary themes: (i) materials discovery with Materials Acceleration Platforms and (ii) machine-learned models for realistic materials simulations of advanced and chemically complex materials. In particular, we are interested in battery materials. These activities are supported by substantial contributions to open-source software and community research infrastructures. The research connects to the activity fields [Materials Design](https://www.bam.de/Navigation/EN/Topics/Materials/Materials-design/materials-design.html) and [Electrical energy storage and conversion](https://bam.de/en/topics/energy/electrical-energy-storage) at BAM and includes collaborations with researchers across both areas. -#### High-throughput calculations and automations -Large amounts of reliably calculated material data are necessary to develop new chemical heuristics. Here, the group develops workflows and methods for high-throughput calculations to reliably perform such calculations. Among other things, the group regularly contributes its own developments to well-known open-source material informatics codes such as [Pymatgen](https://github.com/materialsproject/pymatgen), and [Atomate2](https://github.com/materialsproject/atomate2) and has extensive expertise with these codes. Because of these activities, Janine is part of the [Materials Project Software Foundation](https://next-gen.materialsproject.org/about/people). +#### Materials Acceleration Platforms and Data-Driven Materials Design +- Inverse design and generative models (e.g., [evaluation of generative models]()) +- Synthesis prediction (e.g., via PU learning) +- Property forecasting (e.g., for thermal conductivity) +- Chemical heuristics and materials descriptors (e.g., for magnetism) +- Automated bonding analysis (e.g., as part of LobsterPy) +- High-throughput calculations and data generation (e.g., for our bonding analysis database) +- Pre-trained and foundation models for materials discovery (e.g., as part of MACE-MP-0) +- Integration of experimental and computational data +- Autonomous optimization workflows -#### Machine Learning Potentials -Together with our collaboration partners (group of [Prof. Volker Deringer](https://www.chem.ox.ac.uk/people/volker-deringer)) at the University of Oxford, we develop the software [autoplex](https://github.com/autoatml/autoplex) to automate the training and evaluation of machine learned interatomic potentials. Together with the group of [Prof. Silvana Botti](https://www.physik.ruhr-uni-bochum.de/Professuren/prof-dr-botti-silvana/) at Ruhr University Bochum, we work on finding new chemically complex oxides. +#### Machine-Learned Models for Realistic Materials Simulations +- Machine-learned interatomic potentials (e.g., [benchmarking of MACE-MP-0]()) +- Automated training and benchmarking (e.g., [development of autoplex]()) +- Vibrational, thermodynamic, and transport properties (e.g., for the computation of thermal conductivity) +- Large-scale atomistic simulations +- Battery materials and interfaces +- Chemically complex materials +- Materials stability, degradation, and safety -#### Vibrational Properties +#### Open-Source Software and Research Infrastructures +- Major contributors to the materials analysis [Pymatgen](https://github.com/materialsproject/pymatgen) +- Maintenance and development of the workflow library [Atomate2](https://github.com/materialsproject/atomate2) +- Maintenance and joint development of software [autoplex](https://github.com/autoatml/autoplex) for automated training of machine learning interatomic potentials (joint development with the group of [Prof. Volker Deringer](https://www.chem.ox.ac.uk/people/volker-deringer) at the University of Oxford) +- Development of [LobsterPy](https://github.com/JaGeo/LobsterPy) for automated bond analysis +- Dataset and software contributions to the [Materials Project](https://next-gen.materialsproject.org/) and contributions to the [Materials Project Software Foundation](https://next-gen.materialsproject.org/about/people) +- Dataset and software contributions to [NOMAD](https://nomad-lab.eu/nomad-lab/) and involvement in FAIRmat 2.0 -Vibration properties play a fundamental role in the stability and heat transport of materials. Both material properties are significant for the safe use of materials. Here, the group is concerned with the ab initio prediction of such data to advance machine learning of such properties. #### 5 most important recent publications with contributions from our lab From 6f237b2ae28cba488f2b0691c0428faf2b6e9c90 Mon Sep 17 00:00:00 2001 From: "J. George" Date: Tue, 4 Aug 2026 15:45:58 +0200 Subject: [PATCH 2/7] Add DOI links to research themes in research.md Updated references in the research section to include DOI links for better accessibility and citation. --- _pages/research.md | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/_pages/research.md b/_pages/research.md index 4c779cc..ac43a98 100644 --- a/_pages/research.md +++ b/_pages/research.md @@ -14,24 +14,24 @@ Our research is organized around two complementary themes: (i) materials discove #### Materials Acceleration Platforms and Data-Driven Materials Design -- Inverse design and generative models (e.g., [evaluation of generative models]()) -- Synthesis prediction (e.g., via PU learning) -- Property forecasting (e.g., for thermal conductivity) -- Chemical heuristics and materials descriptors (e.g., for magnetism) -- Automated bonding analysis (e.g., as part of LobsterPy) -- High-throughput calculations and data generation (e.g., for our bonding analysis database) -- Pre-trained and foundation models for materials discovery (e.g., as part of MACE-MP-0) +- Inverse design and generative models (e.g., [evaluation of generative models](https://doi.org/10.1088/2515-7639/ae869a)) +- Synthesis prediction (e.g., via [co-training and PU learning](https://doi.org/10.1039/D4DD00394B)) +- Property forecasting (e.g., for [thermal conductivity via chemical bonding concepts](https://doi.org/10.1002/aidi.70129)) +- Chemical heuristics and materials descriptors (e.g., for [bond angle-based features for magnetic predictions](https://doi.org/10.26434/chemrxiv-2025-xj84d/v3)) +- Automated bonding analysis (e.g., as part of [LobsterPy](https://doi.org/10.21105/joss.06286) and [atomate2](https://doi.org/10.1039/D5DD00019J)) +- High-throughput calculations and data generation (e.g., for [our bonding analysis database](https://doi.org/10.1038/s41597-023-02477-5)) +- Pre-trained and foundation models for materials discovery (e.g., as part of [MACE-MP-0](https://doi.org/10.1063/5.0297006)) - Integration of experimental and computational data - Autonomous optimization workflows #### Machine-Learned Models for Realistic Materials Simulations -- Machine-learned interatomic potentials (e.g., [benchmarking of MACE-MP-0]()) -- Automated training and benchmarking (e.g., [development of autoplex]()) -- Vibrational, thermodynamic, and transport properties (e.g., for the computation of thermal conductivity) +- Machine-learned interatomic potentials (e.g., [benchmarking of MACE-MP-0](https://doi.org/10.1063/5.0297006)) +- Automated training and benchmarking (e.g., [development of autoplex](https://doi.org/10.1038/s41467-025-62510-6)) +- Vibrational, thermodynamic, and transport properties (e.g., for the [computation of thermal conductivity](https://doi.org/10.1039/d5ta08709k)) +- Chemically complex materials (e.g., see [our recent review](https://doi.org/10.1016/j.cossms.2026.101256)) - Large-scale atomistic simulations - Battery materials and interfaces -- Chemically complex materials - Materials stability, degradation, and safety #### Open-Source Software and Research Infrastructures From a50c792957d9e1e484796e43bc085f136e4de569 Mon Sep 17 00:00:00 2001 From: "J. George" Date: Tue, 4 Aug 2026 15:48:32 +0200 Subject: [PATCH 3/7] Update research page with new content and links --- _pages/research.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/_pages/research.md b/_pages/research.md index ac43a98..847a9d7 100644 --- a/_pages/research.md +++ b/_pages/research.md @@ -7,13 +7,13 @@ nav: true nav_order: 3 --- - We accelerate materials design with computational and data-driven approaches at BAM and the University of Jena. In particular, we focus on the design of safe, sustainable, and reliable materials. To achieve this, we combine first-principles simulations, condensed matter physics, materials informatics, machine learning, and high-throughput computational methods. Our research is organized around two complementary themes: (i) materials discovery with Materials Acceleration Platforms and (ii) machine-learned models for realistic materials simulations of advanced and chemically complex materials. In particular, we are interested in battery materials. These activities are supported by substantial contributions to open-source software and community research infrastructures. The research connects to the activity fields [Materials Design](https://www.bam.de/Navigation/EN/Topics/Materials/Materials-design/materials-design.html) and [Electrical energy storage and conversion](https://bam.de/en/topics/energy/electrical-energy-storage) at BAM and includes collaborations with researchers across both areas. #### Materials Acceleration Platforms and Data-Driven Materials Design + - Inverse design and generative models (e.g., [evaluation of generative models](https://doi.org/10.1088/2515-7639/ae869a)) - Synthesis prediction (e.g., via [co-training and PU learning](https://doi.org/10.1039/D4DD00394B)) - Property forecasting (e.g., for [thermal conductivity via chemical bonding concepts](https://doi.org/10.1002/aidi.70129)) @@ -26,6 +26,7 @@ Our research is organized around two complementary themes: (i) materials discove #### Machine-Learned Models for Realistic Materials Simulations + - Machine-learned interatomic potentials (e.g., [benchmarking of MACE-MP-0](https://doi.org/10.1063/5.0297006)) - Automated training and benchmarking (e.g., [development of autoplex](https://doi.org/10.1038/s41467-025-62510-6)) - Vibrational, thermodynamic, and transport properties (e.g., for the [computation of thermal conductivity](https://doi.org/10.1039/d5ta08709k)) @@ -35,6 +36,7 @@ Our research is organized around two complementary themes: (i) materials discove - Materials stability, degradation, and safety #### Open-Source Software and Research Infrastructures + - Major contributors to the materials analysis [Pymatgen](https://github.com/materialsproject/pymatgen) - Maintenance and development of the workflow library [Atomate2](https://github.com/materialsproject/atomate2) - Maintenance and joint development of software [autoplex](https://github.com/autoatml/autoplex) for automated training of machine learning interatomic potentials (joint development with the group of [Prof. Volker Deringer](https://www.chem.ox.ac.uk/people/volker-deringer) at the University of Oxford) From 4e63671a80035c055cb5bcd3a175b05e31c6a8c9 Mon Sep 17 00:00:00 2001 From: "J. George" Date: Tue, 4 Aug 2026 15:53:33 +0200 Subject: [PATCH 4/7] Add open-source software contributions section --- _pages/research.md | 1 + 1 file changed, 1 insertion(+) diff --git a/_pages/research.md b/_pages/research.md index 847a9d7..a06cb96 100644 --- a/_pages/research.md +++ b/_pages/research.md @@ -35,6 +35,7 @@ Our research is organized around two complementary themes: (i) materials discove - Battery materials and interfaces - Materials stability, degradation, and safety + #### Open-Source Software and Research Infrastructures - Major contributors to the materials analysis [Pymatgen](https://github.com/materialsproject/pymatgen) From 0bbe238f694bd253f6fa8824c151ec1965ab9637 Mon Sep 17 00:00:00 2001 From: "J. George" Date: Tue, 4 Aug 2026 15:56:01 +0200 Subject: [PATCH 5/7] Clean up research.md by removing empty lines Removed empty lines for better formatting and readability. --- _pages/research.md | 4 ---- 1 file changed, 4 deletions(-) diff --git a/_pages/research.md b/_pages/research.md index a06cb96..07e69ad 100644 --- a/_pages/research.md +++ b/_pages/research.md @@ -11,7 +11,6 @@ We accelerate materials design with computational and data-driven approaches at Our research is organized around two complementary themes: (i) materials discovery with Materials Acceleration Platforms and (ii) machine-learned models for realistic materials simulations of advanced and chemically complex materials. In particular, we are interested in battery materials. These activities are supported by substantial contributions to open-source software and community research infrastructures. The research connects to the activity fields [Materials Design](https://www.bam.de/Navigation/EN/Topics/Materials/Materials-design/materials-design.html) and [Electrical energy storage and conversion](https://bam.de/en/topics/energy/electrical-energy-storage) at BAM and includes collaborations with researchers across both areas. - #### Materials Acceleration Platforms and Data-Driven Materials Design - Inverse design and generative models (e.g., [evaluation of generative models](https://doi.org/10.1088/2515-7639/ae869a)) @@ -24,7 +23,6 @@ Our research is organized around two complementary themes: (i) materials discove - Integration of experimental and computational data - Autonomous optimization workflows - #### Machine-Learned Models for Realistic Materials Simulations - Machine-learned interatomic potentials (e.g., [benchmarking of MACE-MP-0](https://doi.org/10.1063/5.0297006)) @@ -35,7 +33,6 @@ Our research is organized around two complementary themes: (i) materials discove - Battery materials and interfaces - Materials stability, degradation, and safety - #### Open-Source Software and Research Infrastructures - Major contributors to the materials analysis [Pymatgen](https://github.com/materialsproject/pymatgen) @@ -45,7 +42,6 @@ Our research is organized around two complementary themes: (i) materials discove - Dataset and software contributions to the [Materials Project](https://next-gen.materialsproject.org/) and contributions to the [Materials Project Software Foundation](https://next-gen.materialsproject.org/about/people) - Dataset and software contributions to [NOMAD](https://nomad-lab.eu/nomad-lab/) and involvement in FAIRmat 2.0 - #### 5 most important recent publications with contributions from our lab - ["Accelerated data-driven materials science with the Materials Project"](https://doi.org/10.1038/s41563-025-02272-0) From 382cc30e8eb41c0e4b7a420954fa9110167ac261 Mon Sep 17 00:00:00 2001 From: "J. George" Date: Tue, 4 Aug 2026 15:58:15 +0200 Subject: [PATCH 6/7] Revise research overview and focus areas Updated the research page to include details on materials design and related themes. --- _pages/research.md | 1 + 1 file changed, 1 insertion(+) diff --git a/_pages/research.md b/_pages/research.md index 07e69ad..c006e2e 100644 --- a/_pages/research.md +++ b/_pages/research.md @@ -7,6 +7,7 @@ nav: true nav_order: 3 --- + We accelerate materials design with computational and data-driven approaches at BAM and the University of Jena. In particular, we focus on the design of safe, sustainable, and reliable materials. To achieve this, we combine first-principles simulations, condensed matter physics, materials informatics, machine learning, and high-throughput computational methods. Our research is organized around two complementary themes: (i) materials discovery with Materials Acceleration Platforms and (ii) machine-learned models for realistic materials simulations of advanced and chemically complex materials. In particular, we are interested in battery materials. These activities are supported by substantial contributions to open-source software and community research infrastructures. The research connects to the activity fields [Materials Design](https://www.bam.de/Navigation/EN/Topics/Materials/Materials-design/materials-design.html) and [Electrical energy storage and conversion](https://bam.de/en/topics/energy/electrical-energy-storage) at BAM and includes collaborations with researchers across both areas. From 075c107c0ca1852b8b423a6c4d644df9b1dee639 Mon Sep 17 00:00:00 2001 From: "J. George" Date: Tue, 4 Aug 2026 17:20:45 +0200 Subject: [PATCH 7/7] Update research.md --- _pages/research.md | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/_pages/research.md b/_pages/research.md index c006e2e..893a2d8 100644 --- a/_pages/research.md +++ b/_pages/research.md @@ -7,8 +7,7 @@ nav: true nav_order: 3 --- - -We accelerate materials design with computational and data-driven approaches at BAM and the University of Jena. In particular, we focus on the design of safe, sustainable, and reliable materials. To achieve this, we combine first-principles simulations, condensed matter physics, materials informatics, machine learning, and high-throughput computational methods. +We accelerate materials design with computational and data-driven approaches at BAM and the University of Jena. In particular, we focus on the design of safe, sustainable, and reliable materials. To achieve this, we combine first-principles simulations, condensed matter physics, materials informatics, machine learning, and high-throughput computational methods. Our research is organized around two complementary themes: (i) materials discovery with Materials Acceleration Platforms and (ii) machine-learned models for realistic materials simulations of advanced and chemically complex materials. In particular, we are interested in battery materials. These activities are supported by substantial contributions to open-source software and community research infrastructures. The research connects to the activity fields [Materials Design](https://www.bam.de/Navigation/EN/Topics/Materials/Materials-design/materials-design.html) and [Electrical energy storage and conversion](https://bam.de/en/topics/energy/electrical-energy-storage) at BAM and includes collaborations with researchers across both areas.