Can a Large Language Model (LLM) really understand chemical reactivity? oMeBench: Towards Robust Benchmarking of LLMs in Organic Mechanism Elucidation and Reasoning
Ruiling Xu, Yifan Zhang, Qingyun Wang, Carl Edwards, Heng Ji
Organic reaction mechanisms are the stepwise elementary reactions by which reactants form intermediates and products, and are fundamental to understanding chemical reactivity and designing new molecules and reactions. Although large language models (LLMs) have shown promise in understanding chemical tasks such as synthesis design, it is unclear to what extent this reflects genuine chemical reasoning capabilities, i.e., the ability to generate valid intermediates, maintain chemical consistency, and follow logically coherent multi-step pathways. We address this by introducing oMeBench, the first large-scale, expert-curated benchmark for organic mechanism reasoning in organic chemistry. It comprises over 10,000 annotated mechanistic steps with intermediates, type labels, and difficulty ratings. Furthermore, to evaluate LLM capability more precisely and enable fine-grained scoring, we propose oMeS, a dynamic evaluation framework that combines step-level logic and chemical similarity. We analyze the performance of state-of-the-art LLMs, and our results show that although current models display promising chemical intuition, they struggle with correct and consistent multi-step reasoning. Notably, we find that using prompting strategy and fine-tuning a specialist model on our proposed dataset increases performance by 50% over the leading closed-source model. We hope that oMeBench will serve as a rigorous foundation for advancing AI systems toward genuine chemical reasoning.
🔗 https://arxiv.org/abs/2510.07731v1
Ruiling Xu, Yifan Zhang, Qingyun Wang, Carl Edwards, Heng Ji
Organic reaction mechanisms are the stepwise elementary reactions by which reactants form intermediates and products, and are fundamental to understanding chemical reactivity and designing new molecules and reactions. Although large language models (LLMs) have shown promise in understanding chemical tasks such as synthesis design, it is unclear to what extent this reflects genuine chemical reasoning capabilities, i.e., the ability to generate valid intermediates, maintain chemical consistency, and follow logically coherent multi-step pathways. We address this by introducing oMeBench, the first large-scale, expert-curated benchmark for organic mechanism reasoning in organic chemistry. It comprises over 10,000 annotated mechanistic steps with intermediates, type labels, and difficulty ratings. Furthermore, to evaluate LLM capability more precisely and enable fine-grained scoring, we propose oMeS, a dynamic evaluation framework that combines step-level logic and chemical similarity. We analyze the performance of state-of-the-art LLMs, and our results show that although current models display promising chemical intuition, they struggle with correct and consistent multi-step reasoning. Notably, we find that using prompting strategy and fine-tuning a specialist model on our proposed dataset increases performance by 50% over the leading closed-source model. We hope that oMeBench will serve as a rigorous foundation for advancing AI systems toward genuine chemical reasoning.
🔗 https://arxiv.org/abs/2510.07731v1
arXiv.org
oMeBench: Towards Robust Benchmarking of LLMs in Organic Mechanism...
Organic reaction mechanisms are the stepwise elementary reactions by which reactants form intermediates and products, and are fundamental to understanding chemical reactivity and designing new...
❤5👍1
We are pleased to announce the release of alvaModel version 3.0.0, a major
update to our software for building, validating, and analyzing QSAR/QSPR
models.
This release introduces several important features aimed at expanding modeling
capabilities, improving interpretability, and streamlining the model
development workflow.
New Modeling Capabilities
Regression and Classification Models
- Decision Tree (Regression and Classification)
- Random Forest (Regression and Classification)
- Consensus Classification Model (weighted on reliability)
Customizable Model Parameters
Users can now define sets of custom parameters for:
- K-Nearest Neighbors (KNN)
- Support Vector Machine (SVM)
- Consensus (Classification)
- Decision Tree
- Random Forest
Applicability Domain
- New method: Bounding Box
Similarity Measures
- Dice and Cosine distances are now supported
Enhanced Evaluation Metrics
New Classification Scores
- AUROC (Area Under the Receiver Operating Characteristic Curve)
- F1 Score
- Matthews Correlation Coefficient (MCC)
- Cohens Kappa
Improved Visualization and User Interface
Model Comparison Features
- New interface for comparing multiple models
- Display of 95% confidence intervals for scores
- Automatic highlighting of best/worst scores
New comparison charts:
- Simultaneous Confidence Interval Plot
- Radar Plot
Classification Model View Enhancements
- New Reliability column
- Color-coded indicators for:
- Correct predictions
- Reliability
- Applicability Domain (A.D.)
- Colored Confusion Matrix with molecule filtering capability
Workflow and Usability Improvements
- Prediction of external datasets without the need for alvaRunner
- New Open Recent Project menu
- Batch model saving in automatic model generation
- Improved dataset management:
- Move or copy molecules between datasets
- Rename external variable columns
- Contextual popup menus for dataset and model tables
Charts and Visualization Enhancements
- Unified chart toolbar for easy selection of chart types and settings
- For regression models: R and RMSE are now displayed in plot legends
- For classification models: ROC Curve and PrecisionRecall Curve available
For an overview of the new features, you can watch the introduction video:
https://youtu.be/5TVcpxTdQAY
More information is available on the alvaModel product page:
https://www.alvascience.com/alvamodel
update to our software for building, validating, and analyzing QSAR/QSPR
models.
This release introduces several important features aimed at expanding modeling
capabilities, improving interpretability, and streamlining the model
development workflow.
New Modeling Capabilities
Regression and Classification Models
- Decision Tree (Regression and Classification)
- Random Forest (Regression and Classification)
- Consensus Classification Model (weighted on reliability)
Customizable Model Parameters
Users can now define sets of custom parameters for:
- K-Nearest Neighbors (KNN)
- Support Vector Machine (SVM)
- Consensus (Classification)
- Decision Tree
- Random Forest
Applicability Domain
- New method: Bounding Box
Similarity Measures
- Dice and Cosine distances are now supported
Enhanced Evaluation Metrics
New Classification Scores
- AUROC (Area Under the Receiver Operating Characteristic Curve)
- F1 Score
- Matthews Correlation Coefficient (MCC)
- Cohens Kappa
Improved Visualization and User Interface
Model Comparison Features
- New interface for comparing multiple models
- Display of 95% confidence intervals for scores
- Automatic highlighting of best/worst scores
New comparison charts:
- Simultaneous Confidence Interval Plot
- Radar Plot
Classification Model View Enhancements
- New Reliability column
- Color-coded indicators for:
- Correct predictions
- Reliability
- Applicability Domain (A.D.)
- Colored Confusion Matrix with molecule filtering capability
Workflow and Usability Improvements
- Prediction of external datasets without the need for alvaRunner
- New Open Recent Project menu
- Batch model saving in automatic model generation
- Improved dataset management:
- Move or copy molecules between datasets
- Rename external variable columns
- Contextual popup menus for dataset and model tables
Charts and Visualization Enhancements
- Unified chart toolbar for easy selection of chart types and settings
- For regression models: R and RMSE are now displayed in plot legends
- For classification models: ROC Curve and PrecisionRecall Curve available
For an overview of the new features, you can watch the introduction video:
https://youtu.be/5TVcpxTdQAY
More information is available on the alvaModel product page:
https://www.alvascience.com/alvamodel
YouTube
alvaModel v3.0 - Introduction
A short video introduction to alvaModel 3.0
alvaModel is a powerful software tool developed for creating, validating, and applying QSAR and QSPR models.
With its intuitive interface and advanced modeling capabilities, alvaModel enables researchers to build…
alvaModel is a powerful software tool developed for creating, validating, and applying QSAR and QSPR models.
With its intuitive interface and advanced modeling capabilities, alvaModel enables researchers to build…
❤9
WE ARE HIRING !
Position Summary: The laboratory of Dr. Donald Weaver, Krembil Research Institute, and Department of Chemistry (University of Toronto) invites applications for one Postdoctoral Researcher Position in Computational/Theoretical Chemistry. Work will involve applications of molecular modelling methods (molecular dynamics/molecular mechanics/quantum mechanics calculations) to understanding neurochemical processes, developing in silico disease models and participating in a multi-disciplinary group focused on drug design for neurodegenerative disease.
Applicants should hold a PhD in Computational/Theoretical Chemistry (must be obtained within the previous 5 years) and have a strong background in the development and use of molecular modelling and electronic structure theory. The successful candidate will work in the Krembil Research Institute, which is part of the Toronto Western Hospital. We expect strong commitment, good communication skills and ability to work in a multidisciplinary environment with highly qualified professionals from international backgrounds.
Major topics of research include: Molecular dynamics simulations;
Molecular mechanics optimizations/protein modelling; DFT and molecular orbital ab initio calculations; Applications to neurochemistry, small molecule structure and hydration/solvation
Duties: Performing molecular modelling simulations; Performing molecular dynamics simulations; Performing molecular mechanics calculations; Performing molecular quantum mechanics calculations; Computer-aided drug design
Apply here:
https://lnkd.in/gmQrhxUz
🔗 https://www.linkedin.com/posts/donald-weaver-067a963_chemistry-computationalchemistry-theoreticalchemistry-activity-7392223121266827264-mJ2U?utm_source=share&utm_medium=member_desktop&rcm=ACoAADByb54BZAu0zfSJLooSdNdx0bXFCOsvoA0
Position Summary: The laboratory of Dr. Donald Weaver, Krembil Research Institute, and Department of Chemistry (University of Toronto) invites applications for one Postdoctoral Researcher Position in Computational/Theoretical Chemistry. Work will involve applications of molecular modelling methods (molecular dynamics/molecular mechanics/quantum mechanics calculations) to understanding neurochemical processes, developing in silico disease models and participating in a multi-disciplinary group focused on drug design for neurodegenerative disease.
Applicants should hold a PhD in Computational/Theoretical Chemistry (must be obtained within the previous 5 years) and have a strong background in the development and use of molecular modelling and electronic structure theory. The successful candidate will work in the Krembil Research Institute, which is part of the Toronto Western Hospital. We expect strong commitment, good communication skills and ability to work in a multidisciplinary environment with highly qualified professionals from international backgrounds.
Major topics of research include: Molecular dynamics simulations;
Molecular mechanics optimizations/protein modelling; DFT and molecular orbital ab initio calculations; Applications to neurochemistry, small molecule structure and hydration/solvation
Duties: Performing molecular modelling simulations; Performing molecular dynamics simulations; Performing molecular mechanics calculations; Performing molecular quantum mechanics calculations; Computer-aided drug design
Apply here:
https://lnkd.in/gmQrhxUz
🔗 https://www.linkedin.com/posts/donald-weaver-067a963_chemistry-computationalchemistry-theoreticalchemistry-activity-7392223121266827264-mJ2U?utm_source=share&utm_medium=member_desktop&rcm=ACoAADByb54BZAu0zfSJLooSdNdx0bXFCOsvoA0
lnkd.in
LinkedIn
This link will take you to a page that’s not on LinkedIn
❤8
ChemTorch is here! 🧪⚗️
Would you like to do reaction property prediction? Use fingerprints, reaction SMILES, reaction graphs, or 3D coordinates interchangeably? Put any architecture on top, with out-of-the-box support for neural networks, D-MPNN, GCN, GatedGCN, GAT, GATv2, GINE, GraphGPS, PNA, SMILES tokenizers, and DimeNet++ (and many more upcoming). Let me introduce ChemTorch, fantastic work by Jasper De Landsheere and Anton Zamyatin, along with Johannes Karwounopoulos
Preprint: https://lnkd.in/dwXMjgSt
Code: https://lnkd.in/dp2RkynK (feel free to open issues to request new features or architectures)
Documentation: https://lnkd.in/dxFj2W7V
ORCA OPI user example: https://lnkd.in/d5pTQrRg
🔗 https://www.linkedin.com/posts/esther-heid-ab20001a6_chemtorch-is-here-would-you-like-to-do-reaction-activity-7391816963485765632-D-GU?utm_source=social_share_send&utm_medium=member_desktop_web&rcm=ACoAADByb54BZAu0zfSJLooSdNdx0bXFCOsvoA0
Would you like to do reaction property prediction? Use fingerprints, reaction SMILES, reaction graphs, or 3D coordinates interchangeably? Put any architecture on top, with out-of-the-box support for neural networks, D-MPNN, GCN, GatedGCN, GAT, GATv2, GINE, GraphGPS, PNA, SMILES tokenizers, and DimeNet++ (and many more upcoming). Let me introduce ChemTorch, fantastic work by Jasper De Landsheere and Anton Zamyatin, along with Johannes Karwounopoulos
Preprint: https://lnkd.in/dwXMjgSt
Code: https://lnkd.in/dp2RkynK (feel free to open issues to request new features or architectures)
Documentation: https://lnkd.in/dxFj2W7V
ORCA OPI user example: https://lnkd.in/d5pTQrRg
🔗 https://www.linkedin.com/posts/esther-heid-ab20001a6_chemtorch-is-here-would-you-like-to-do-reaction-activity-7391816963485765632-D-GU?utm_source=social_share_send&utm_medium=member_desktop_web&rcm=ACoAADByb54BZAu0zfSJLooSdNdx0bXFCOsvoA0
❤9🔥2
Save the date!
The 3rd PySCF Developers Meeting is scheduled for 2026!
📅 August 27-28, 2026
🕘 9:00 AM to 7:00 PM (Aug 27), 9:00 AM to 4:00 PM (Aug 28), all times in GMT-6.
📍 University of Chicago, Illinois.
Stay tuned at PySCF's official webpage 👉 https://pyscf.org/
The 3rd PySCF Developers Meeting is scheduled for 2026!
📅 August 27-28, 2026
🕘 9:00 AM to 7:00 PM (Aug 27), 9:00 AM to 4:00 PM (Aug 28), all times in GMT-6.
📍 University of Chicago, Illinois.
Stay tuned at PySCF's official webpage 👉 https://pyscf.org/
🔥5❤2
📢 Registration Now Open!
Virtual Winter School on Computational Chemistry 2026
📅 *26–30 January 2026*
🌐 http://www.winterschool.cc
💰 Registration is completely free
The Virtual Winter School brings together researchers from around the world to explore a wide range of topics in computational and theoretical chemistry. Our extended lecture format allows speakers to cover both core foundations and cutting-edge developments.
✨ What’s new in 2026
• Two hands-on workshops (Q-Chem and GROMACS)
• Submit a Single Figure Presentation (SFP) and give a short talk
• Two interactive panel sessions with our speakers
🎙 Confirmed Speakers
• Prof. Alexander Sokolov (Ohio State University, USA)
• Prof. Anne B. McCoy (University of Washington, USA)
• Dr. Cristina Trujillo (University of Manchester, UK)
• Prof. Debashree Ghosh (IACS, India)
• Dr. Esther Heid (TU Wien, Austria)
• Prof. Graham Worth (University College London, UK)
• Dr. Jerelle Joseph (Princeton University, USA)
• Dr. Johannes Gierschner (IMDEA Nanociencia, Spain)
• Prof. José Walkimar de M. Carneiro (UFF, Brazil)
• Prof. Raphael C. Bernardi (Auburn University, USA)
• Prof. Remco W. A. Havenith (University of Groningen, Netherlands)
• Dr. Roel Hoefnagels & Dr. Alain Collas (Johnson & Johnson)
🤝 Support the School
A huge thank-you to our sponsors: Q-Chem and the RSC Theoretical Chemistry Interest Group.
The Winter School is entirely community-supported. If you’d like to help sustain it for the coming years, you can donate here:
🔗 https://www.zeffy.com/en-US/donation-form/contribute-to-the-virtual-winter-school-on-computational-chemistry-vwscc
☕️ Fun fact: if 100 attendees donate the price of a fancy coffee, we can keep the School running for another decade.
📬 We can’t wait to see you in the 2026 edition!
— Organising Committee, Virtual Winter School on Computational Chemistry
Virtual Winter School on Computational Chemistry 2026
📅 *26–30 January 2026*
🌐 http://www.winterschool.cc
💰 Registration is completely free
The Virtual Winter School brings together researchers from around the world to explore a wide range of topics in computational and theoretical chemistry. Our extended lecture format allows speakers to cover both core foundations and cutting-edge developments.
✨ What’s new in 2026
• Two hands-on workshops (Q-Chem and GROMACS)
• Submit a Single Figure Presentation (SFP) and give a short talk
• Two interactive panel sessions with our speakers
🎙 Confirmed Speakers
• Prof. Alexander Sokolov (Ohio State University, USA)
• Prof. Anne B. McCoy (University of Washington, USA)
• Dr. Cristina Trujillo (University of Manchester, UK)
• Prof. Debashree Ghosh (IACS, India)
• Dr. Esther Heid (TU Wien, Austria)
• Prof. Graham Worth (University College London, UK)
• Dr. Jerelle Joseph (Princeton University, USA)
• Dr. Johannes Gierschner (IMDEA Nanociencia, Spain)
• Prof. José Walkimar de M. Carneiro (UFF, Brazil)
• Prof. Raphael C. Bernardi (Auburn University, USA)
• Prof. Remco W. A. Havenith (University of Groningen, Netherlands)
• Dr. Roel Hoefnagels & Dr. Alain Collas (Johnson & Johnson)
🤝 Support the School
A huge thank-you to our sponsors: Q-Chem and the RSC Theoretical Chemistry Interest Group.
The Winter School is entirely community-supported. If you’d like to help sustain it for the coming years, you can donate here:
🔗 https://www.zeffy.com/en-US/donation-form/contribute-to-the-virtual-winter-school-on-computational-chemistry-vwscc
☕️ Fun fact: if 100 attendees donate the price of a fancy coffee, we can keep the School running for another decade.
📬 We can’t wait to see you in the 2026 edition!
— Organising Committee, Virtual Winter School on Computational Chemistry
Zeffy
Contribute to the Virtual Winter School on Computational Chemistry (VWSCC)
❤6
What is a conformer? What is chirality? What is quantum chemistry in short? What is gas electron diffraction? What is microwave spectroscopy? What is infrared spectroscopy? What is photoelectron spectroscopy?
How to install and run xTB/CREST/ORCA on Windows/Linux/macOS?
If any of these questions ring a bell for you and you would like to read a short, concise text that elaborates on these topics, then the next short manual is just for you.
"The Hitchhiker's Guide to the Conformational Search"
Tikhonov, D. ; Sun, W. ; Xie, F. ; Berggoetz, F. E. L. ; Singh, H. ; Caliebe, S ; Schnell, M.
doi: 10.3204/PUBDB-2025-04761
https://doi.org/10.3204/PUBDB-2025-04761
There, in 100 pages, you can find a brief theory and practical instructions for installing and using quantum-chemical software and for analyzing the calculation results. Practical tasks (with detailed answers) will help inexperienced users enter the computational chemistry realm, and for experienced quantum chemists, there will be new ways to connect their computations to experimental observables.
How to install and run xTB/CREST/ORCA on Windows/Linux/macOS?
If any of these questions ring a bell for you and you would like to read a short, concise text that elaborates on these topics, then the next short manual is just for you.
"The Hitchhiker's Guide to the Conformational Search"
Tikhonov, D. ; Sun, W. ; Xie, F. ; Berggoetz, F. E. L. ; Singh, H. ; Caliebe, S ; Schnell, M.
doi: 10.3204/PUBDB-2025-04761
https://doi.org/10.3204/PUBDB-2025-04761
There, in 100 pages, you can find a brief theory and practical instructions for installing and using quantum-chemical software and for analyzing the calculation results. Practical tasks (with detailed answers) will help inexperienced users enter the computational chemistry realm, and for experienced quantum chemists, there will be new ways to connect their computations to experimental observables.
❤17🔥5
ORCA 6.1.1 bugfix release is now available!
Dear ORCA community,
It is our great pleasure to release ORCA 6.1.1 to the general public.
While the ORCA teams and their collaborators are constantly working on
new and exciting features, this release is strictly a bug fix release
that irons out some of the glitches that remained in the program
following the release of ORCA 6.1. Thus, in this release, we have taken
care of almost all of the bugs and inconsistencies that were reported to
us between the release of ORCA 6.1 (June 2025) and about late September
2025. At the same time, significant efforts were spent updating and
correcting the manual to more accurately reflect the state of the
program.
We are very grateful to all -- by now nearly 100000 -- academic ORCA
users as well as the FACCTs customers who have spent significant time
and effort documenting the bugs or unexpected program behavior that they
have encountered. It is a truly rewarding experience to be part of such
a large and positive community! Also, it is an absolute pleasure to come
across ORCA users everywhere in the world the ORCA team members are
travelling -- thank you so much for your loyalty and your enthusiasm!
I am also extremely grateful for the very hard and often tedious work
that the ORCA development teams have invested in the last few months,
documenting and cataloguing bugs, tracking down the responsible
programmers and motivating them to fix the problems that the users have
encountered. Thank you very much for your skill, dedication and
patience!
We are absolutely dedicated to make your ORCA experience as pleasant as
possible and we hope that you, the ORCA community, will make excellent
use of ORCA 6.1.1 while we are going back to pushing the envelope some
further for the next ORCA release.
Frank Neese on behalf of the ORCA developers, December 1st, 2025
Full changelog available at 👉 https://orca-manual.mpi-muelheim.mpg.de/contents/appendix/detailedchangelog.html
Dear ORCA community,
It is our great pleasure to release ORCA 6.1.1 to the general public.
While the ORCA teams and their collaborators are constantly working on
new and exciting features, this release is strictly a bug fix release
that irons out some of the glitches that remained in the program
following the release of ORCA 6.1. Thus, in this release, we have taken
care of almost all of the bugs and inconsistencies that were reported to
us between the release of ORCA 6.1 (June 2025) and about late September
2025. At the same time, significant efforts were spent updating and
correcting the manual to more accurately reflect the state of the
program.
We are very grateful to all -- by now nearly 100000 -- academic ORCA
users as well as the FACCTs customers who have spent significant time
and effort documenting the bugs or unexpected program behavior that they
have encountered. It is a truly rewarding experience to be part of such
a large and positive community! Also, it is an absolute pleasure to come
across ORCA users everywhere in the world the ORCA team members are
travelling -- thank you so much for your loyalty and your enthusiasm!
I am also extremely grateful for the very hard and often tedious work
that the ORCA development teams have invested in the last few months,
documenting and cataloguing bugs, tracking down the responsible
programmers and motivating them to fix the problems that the users have
encountered. Thank you very much for your skill, dedication and
patience!
We are absolutely dedicated to make your ORCA experience as pleasant as
possible and we hope that you, the ORCA community, will make excellent
use of ORCA 6.1.1 while we are going back to pushing the envelope some
further for the next ORCA release.
Frank Neese on behalf of the ORCA developers, December 1st, 2025
Full changelog available at 👉 https://orca-manual.mpi-muelheim.mpg.de/contents/appendix/detailedchangelog.html
❤8🔥1
Hiring!
1-year postdoc position in computational chemistry at the University of Copenhagen
The research is focusing on automated reaction prediction in collaboration with two major Pharma companies (see e.g. https://lnkd.in/dBec6qJC)
The ideal candidate has experience in
Quantum chemistry calculations (especially reactivity)
Programming (especially Python)
Cheminformatics software (especially RDKit)
Start date in early 2026
https://www.linkedin.com/posts/jan-h-jensen-8960265_im-hiring-1-year-postdoc-position-in-computational-activity-7403049732929101824-jLGf?utm_source=share&utm_medium=member_desktop&rcm=ACoAADByb54BZAu0zfSJLooSdNdx0bXFCOsvoA0
1-year postdoc position in computational chemistry at the University of Copenhagen
The research is focusing on automated reaction prediction in collaboration with two major Pharma companies (see e.g. https://lnkd.in/dBec6qJC)
The ideal candidate has experience in
Quantum chemistry calculations (especially reactivity)
Programming (especially Python)
Cheminformatics software (especially RDKit)
Start date in early 2026
https://www.linkedin.com/posts/jan-h-jensen-8960265_im-hiring-1-year-postdoc-position-in-computational-activity-7403049732929101824-jLGf?utm_source=share&utm_medium=member_desktop&rcm=ACoAADByb54BZAu0zfSJLooSdNdx0bXFCOsvoA0
lnkd.in
LinkedIn
This link will take you to a page that’s not on LinkedIn
❤5🔥3👍1🤮1
🔬 Postdoc in Computational Chemistry at DTU - Technical University of Denmark 🌍⚛️
Ready to advance molten salt reactor technology? Join DTU Chemistry in a cross-disciplinary project shaping the future of sustainable nuclear energy!
👉 Position: Postdoc in Chemistry
👉 Focus: Atomistic & thermodynamic modeling of molten salts
👉 Start: January 2026
👉 Duration: 2 years
👉 Your mission:
✅ Model molten salts and their complex behavior
✅ Explore solvation of fission, activation & corrosion products
✅ Apply ab initio and classical MD simulations (cp2k, LAMMPS, VASP, MetalWalls)
✅ Use CALPHAD thermodynamics (FactSage, Thermochimica, Thermo-Calc) for multicomponent systems
✅ Investigate machine-learning potentials for molten salt simulations
This is a unique collaboration between DTU Chemistry, DTU Physics, DTU Construct, and DTU Energy – combining theory, modeling, and innovation to tackle key challenges in MSR development.
📌 Requirements: PhD in physics, chemistry, or materials science + experience in computational chemistry and thermodynamics
📅 Apply by: 15 December 2025
🔗 Find details in the job advert at: https://lnkd.in/dyNinmpU
Ready to advance molten salt reactor technology? Join DTU Chemistry in a cross-disciplinary project shaping the future of sustainable nuclear energy!
👉 Position: Postdoc in Chemistry
👉 Focus: Atomistic & thermodynamic modeling of molten salts
👉 Start: January 2026
👉 Duration: 2 years
👉 Your mission:
✅ Model molten salts and their complex behavior
✅ Explore solvation of fission, activation & corrosion products
✅ Apply ab initio and classical MD simulations (cp2k, LAMMPS, VASP, MetalWalls)
✅ Use CALPHAD thermodynamics (FactSage, Thermochimica, Thermo-Calc) for multicomponent systems
✅ Investigate machine-learning potentials for molten salt simulations
This is a unique collaboration between DTU Chemistry, DTU Physics, DTU Construct, and DTU Energy – combining theory, modeling, and innovation to tackle key challenges in MSR development.
📌 Requirements: PhD in physics, chemistry, or materials science + experience in computational chemistry and thermodynamics
📅 Apply by: 15 December 2025
🔗 Find details in the job advert at: https://lnkd.in/dyNinmpU
lnkd.in
LinkedIn
This link will take you to a page that’s not on LinkedIn
🔥5❤4👍1
Unlock Advanced Insights: 🎦 SummerSchool Lectures Now on YouTube!
We’re thrilled to announce that the full lecture recordings from our 2025 Summer School on Spectroscopy and Electronic Structure of Transition Metal Complexes are now available on YouTube!
Held from September 7–12 at the Mülheim Chemistry Campus, this intensive program brought together emerging chemists to explore cutting-edge spectroscopic methods and computational tools, with a special focus on leveraging ORCA software for first-principles quantum chemical calculations.
Why watch these lectures?
Whether you’re a student diving into transition metal chemistry or a researcher refining your computational skills, these sessions offer:
Foundational & advanced knowledge: From ligand field theory to Mössbauer spectroscopy.
Expert insights: Lectures by leading scientists, including Frank Neese, Serena DeBeer, Dimitrios Pantazis, Daniel J SantaLucia, and more.
Featured topics:
🔹 Introduction to Computational Chemistry
🔹 Group Theory and Ligand Field Theory
🔹 Optical, X-ray, and Vibrational Spectroscopy
🔹 Spin Hamiltonians, Magnetochemistry, and EPR
🔹 MCD and Mössbauer Spectroscopy
Perfect for: Chemistry students passionate about both spectroscopic techniques and computational modeling, and how they combine to decode complex molecular behavior!
Find access to all lectures via the link in the comments!
Spread the word, share with peers, and revisit these invaluable resources anytime. Let’s keep the learning going!
Find all lectures here: https://youtube.com/playlist?list=PL4lkhPVEEMVijdazRAOmhQ7gGKBJ7aQby&si=QWeEIi3hV0ob-9H_
We’re thrilled to announce that the full lecture recordings from our 2025 Summer School on Spectroscopy and Electronic Structure of Transition Metal Complexes are now available on YouTube!
Held from September 7–12 at the Mülheim Chemistry Campus, this intensive program brought together emerging chemists to explore cutting-edge spectroscopic methods and computational tools, with a special focus on leveraging ORCA software for first-principles quantum chemical calculations.
Why watch these lectures?
Whether you’re a student diving into transition metal chemistry or a researcher refining your computational skills, these sessions offer:
Foundational & advanced knowledge: From ligand field theory to Mössbauer spectroscopy.
Expert insights: Lectures by leading scientists, including Frank Neese, Serena DeBeer, Dimitrios Pantazis, Daniel J SantaLucia, and more.
Featured topics:
🔹 Introduction to Computational Chemistry
🔹 Group Theory and Ligand Field Theory
🔹 Optical, X-ray, and Vibrational Spectroscopy
🔹 Spin Hamiltonians, Magnetochemistry, and EPR
🔹 MCD and Mössbauer Spectroscopy
Perfect for: Chemistry students passionate about both spectroscopic techniques and computational modeling, and how they combine to decode complex molecular behavior!
Find access to all lectures via the link in the comments!
Spread the word, share with peers, and revisit these invaluable resources anytime. Let’s keep the learning going!
Find all lectures here: https://youtube.com/playlist?list=PL4lkhPVEEMVijdazRAOmhQ7gGKBJ7aQby&si=QWeEIi3hV0ob-9H_
❤9
8th International Mini-Symposium on Molecular Machine Learning
🔗Register here: https://lnkd.in/dPTaN7D2
🗓 Date: January 15th, 2026
🕒 Time: 3:00 PM (UTC+1)
💻 Location: Online | Free of charge | Limited number of Participants
👇 Meet this years amazing set of speakers:
🔹Marwin Segler (Microsoft Research AI for Science, UK)
"Deep Learning for Molecules: The First Decade & what’s next"
🔹Kjell Jorner (ETH Zürich, Switzerland)
"Towards generative models for catalyst discovery"
🔹Esther Heid (TU Vienna, Austria)
"Machine-Learned Chemical Reactivity"
🔹Robert Paton (Colorado State University, USA)
"Data-Driven Approaches for Computational Organic Chemistry"
Please share with your network! We look forward to seeing you there.
🔗Register here: https://lnkd.in/dPTaN7D2
🗓 Date: January 15th, 2026
🕒 Time: 3:00 PM (UTC+1)
💻 Location: Online | Free of charge | Limited number of Participants
👇 Meet this years amazing set of speakers:
🔹Marwin Segler (Microsoft Research AI for Science, UK)
"Deep Learning for Molecules: The First Decade & what’s next"
🔹Kjell Jorner (ETH Zürich, Switzerland)
"Towards generative models for catalyst discovery"
🔹Esther Heid (TU Vienna, Austria)
"Machine-Learned Chemical Reactivity"
🔹Robert Paton (Colorado State University, USA)
"Data-Driven Approaches for Computational Organic Chemistry"
Please share with your network! We look forward to seeing you there.
❤6🔥2
Dear Computational and Quantum Chemistry followers ⚛️ :
During the last weeks, we have been receiving in the Comp Chem group a huge quantity of spam messages from fake accounts and bots with cryptocurrency scams and Not Safe for Work (NSFW) content. Clicking into those malicious links can lead to hacking, data stealing, identity theft and so on. In order to reduce the risk of dangerous content reaching this group, we've added two group manager bots which are constantly looking for any intrusion.
If you see a doubtful message, you can report it via Rose bot with the command
Let's keep ourselves safe on the internet and make the stay in the group the most pleasant to all members. Cheers and have a nice weekend.
During the last weeks, we have been receiving in the Comp Chem group a huge quantity of spam messages from fake accounts and bots with cryptocurrency scams and Not Safe for Work (NSFW) content. Clicking into those malicious links can lead to hacking, data stealing, identity theft and so on. In order to reduce the risk of dangerous content reaching this group, we've added two group manager bots which are constantly looking for any intrusion.
If you see a doubtful message, you can report it via Rose bot with the command
\report, or via tagging the admins @<admin>. Let's keep ourselves safe on the internet and make the stay in the group the most pleasant to all members. Cheers and have a nice weekend.
❤11
The other bot which is managing this group is Protectron. Protectron will ban automatically any user who sends cryptoscams or NSFW content.
UCSF ChimeraX version 1.11 has been released!
This will be the last release to support Red Hat Enterprise Linux 8 and
its derivatives.
ChimeraX includes user documentation and is free for noncommercial use.
Download for Windows, Linux, and MacOS from:
https://www.rbvi.ucsf.edu/chimerax/download.html
Updates since version 1.10.1 (July 2024) include:
- 2D Labels and Arrows GUI
- Boltz 2 structure prediction of proteins, nucleic acids, and
small-molecule ligands, with affinity prediction
- Boltz 2 batch ligand-binding predictions (many ligands to same receptor)
- General minimization (including ligands) with Minimize Structure tool
and "minimize" command
- new ViewDock interface (replacing previous ViewDockX)
- new types of trajectory plots: H-bonds, RMSD
- save trajectory plots to image file
- Thermal Ellipsoids GUI
- "chirality" command to report stereocenter chirality
- save/restore scenes including atomic, ribbon, and volume styles and
coloring; more to be added later
For details, please see the ChimeraX change log:
https://www.rbvi.ucsf.edu/trac/ChimeraX/wiki/ChangeLog
This will be the last release to support Red Hat Enterprise Linux 8 and
its derivatives.
ChimeraX includes user documentation and is free for noncommercial use.
Download for Windows, Linux, and MacOS from:
https://www.rbvi.ucsf.edu/chimerax/download.html
Updates since version 1.10.1 (July 2024) include:
- 2D Labels and Arrows GUI
- Boltz 2 structure prediction of proteins, nucleic acids, and
small-molecule ligands, with affinity prediction
- Boltz 2 batch ligand-binding predictions (many ligands to same receptor)
- General minimization (including ligands) with Minimize Structure tool
and "minimize" command
- new ViewDock interface (replacing previous ViewDockX)
- new types of trajectory plots: H-bonds, RMSD
- save trajectory plots to image file
- Thermal Ellipsoids GUI
- "chirality" command to report stereocenter chirality
- save/restore scenes including atomic, ribbon, and volume styles and
coloring; more to be added later
For details, please see the ChimeraX change log:
https://www.rbvi.ucsf.edu/trac/ChimeraX/wiki/ChangeLog
🔥3❤1
Mol* (/'molstar/) is a modern web-based open-source toolkit for visualisation and analysis of large-scale molecular data
High-performance graphics and data handling of the Mol* Viewer allow users to simultaneously visualise up to hundreds of (superimposed) protein structures, play molecular dynamics trajectories, render cell-level models at atomic detail with tens of millions of atoms, or display huge models obtained by I/HM such as the Nuclear Pore Complex.
https://molstar.org
High-performance graphics and data handling of the Mol* Viewer allow users to simultaneously visualise up to hundreds of (superimposed) protein structures, play molecular dynamics trajectories, render cell-level models at atomic detail with tens of millions of atoms, or display huge models obtained by I/HM such as the Nuclear Pore Complex.
https://molstar.org
❤8👌1