Stable and Accurate Orbital-Free Density Functional Theory Powered by Machine Learning

First neural-network kinetic energy functional that makes orbital-free DFT both chemically accurate and variationally stable for diverse organic molecules, including drug-like systems. Trained on perturbed densities to guarantee convergence. Widely discussed in Q1 2026; model and code publicly available.

🔬 Join the TURBOMOLE development team!

A PhD position in computational chemistry is open in the Quantum Chemistry Group of Prof. Dr. Christof Hättig at Ruhr-Universität Bochum.

🎯 The project: Develop advanced coupled-cluster methods for electronically excited states in solution — with direct applications to emission and phosphorescence spectra of OLED and TADF materials.

💻 As part of this project, you will contribute directly to TURBOMOLE, implementing cutting-edge coupled-cluster response theory and modern solvation models into one of the world's leading quantum chemistry packages.

🌐 The position is embedded in the RESOLV Cluster of Excellence, offering an outstanding interdisciplinary research environment with strong international collaborations.

📋 Key details:
▸ Start: October 2026
▸ Duration: 3 years (E13 TV-L, 67%)
▸ Application deadline: July 31, 2026

👉 Apply here: https://lnkd.in/d5dPPyJt

🔎 Source: https://shorturl.at/IHCwY

📢 Switzerland – PhD Position in Computational Chemistry at University of Zurich

🏛 University: University of Zurich
🌍 Country: Switzerland

🎓 Fields:
Chemistry, Computational Chemistry, Physics, Materials Science, Machine Learning


Are you driven to advance the frontiers of computational chemistry and eager to work at the intersection of chemistry, physics, biology, and materials science? If you are seeking a highly interdisciplinary PhD opportunity that leverages cutting-edge computational methods and machine learning, this position at the University of Zurich may be the ideal next step in your academic journey.


This PhD position is centered on advanced computational chemistry, with a particular emphasis on the development and application of innovative methodologies. The research spans several cutting-edge areas: machine learning (ML)-enhanced catalysis and spectroscopy, ML-assisted sampling techniques, embedding approaches for complex systems, and the study of complex liquids and interfaces. These topics are of significant importance in contemporary science, as they enable a deeper understanding of chemical processes at the molecular level and drive innovation in materials science, energy, and biotechnology. By integrating machine learning with traditional computational methods, the project aims to solve complex scientific challenges and pave the way for new discoveries in catalysis and spectroscopy.

🔗 Find out more and Apply Now:

⚠️ Android users: If it doesn't open, use "Open in browser"

https://phdfinder.com/2026/04/15/switzerland-phd-position-in-computational-chemistry-at-university-of-zurich/



━━━━━━━━━━

POSTDOCTORAL POSITION IN THEORETICAL/COMPUTATIONAL CHEMISTRY – BOSTON UNIVERSITY

We are seeking a highly motivated postdoctoral researcher to join our group in the Department of Chemistry at Boston University.

This project focuses on developing computational approaches to simulate redox reactions and electron transfer in biological systems, integrating multiscale modeling with advanced quantum chemistry methods, including polarizable QM/MM approaches enhanced by machine learning.

Qualifications:
•Ph.D. in Theoretical Chemistry or a related field
•Strong programming skills (Python and C++)
•Experience in biomolecular simulations and electronic structure methods
•Familiarity with machine learning in computational chemistry (preferred)

The initial appointment is for one year, with the possibility of extension upon mutual agreement.

To apply or request further information, please contact Ksenia Bravaya (bravaya@bu.edu). Applications should include a brief description of current research, a CV, and a list of three references.

Learn more about our research: https://lnkd.in/ee6CvV5E

Source: https://shorturl.at/DZZ9p

QMCPACK v4.2.0 — high-performance quantum Monte Carlo

Repository: https://github.com/QMCPACK/qmcpack

Key features
* Advanced Variational, Diffusion, and Auxiliary-Field QMC implementations
* Hybrid parallelism (MPI/OpenMP) with GPU acceleration (CUDA/HIP/SYCL)
* Optimized for large-scale HPC environments

Practical implications
* Enables benchmark-quality correlated electronic structure beyond CCSD(T) regimes
* Increasingly relevant for solid-state and strongly correlated systems
* Serves as a reference method for validating ML and DFT approaches

xyzrender: Publication-quality molecular graphics.

Render molecular structures as publication-quality SVG, PNG, PDF, and animated GIF from XYZ, mol/SDF, MOL2, PDB, SMILES, CIF, cube files, quantum chemistry input or output — from the command line or from Python/Jupyter.

https://github.com/aligfellow/xyzrender

LightCone is a web-based molecular visualization and structure editing tool developed by QuantaBricks. It is designed to be user-friendly, allowing high-school students to engage with molecular structures without prior tutorials. The tool supports various features such as editing bond lengths, angles, and dihedrals, as well as fragment selection with rotation and translation. LightCone is permanently free and runs entirely in the browser, ensuring that no data ever leaves the user's device. The upcoming desktop version will add additional features like large-trajectory rendering and molecular orbital visualization.

https://lightcone.quanta-bricks.com

(here is the paper about it) 😉

https://www.nature.com/articles/s41557-026-02123-8

Want to try the new g-xTB on your ORCA (or stand-alone)?

Grab the development version binary here:
https://github.com/grimme-lab/g-xtb

Unpack it.

Inside your ORCA directory, back up your current otool_xtb:

cp otool_xtb otool_xtb.bkp

Replace it with the new xTB binary:

cp ../xtb otool_xtb

You are now ready to use it.

In your ORCA input, just add:

! XTB

%xtb
  XTBINPUTSTRING "--gxtb"
end

Million-atom all-electron quantum chemistry reaches biomolecular scale

A new *Communications Chemistry* paper reports an all-electron Hartree–Fock/divide-and-conquer framework applied to biomolecular systems with tens of millions of atoms and more than 150 million electrons. The result is not high-accuracy quantum chemistry: it uses a minimal basis and aggressive approximations. Its importance is different: it shows that whole-system approximate quantum-mechanical descriptors for huge biological assemblies are becoming technically feasible.

[https://www.nature.com/articles/s42004-026-02038-y](https://www.nature.com/articles/s42004-026-02038-y)

Registration is open for GPMLFF 2026 — a 3-day workshop on General-Purpose Machine-Learned Force Fields: in theory & practice. Researchers entering the field are especially welcome!

Lectures and hands-on tutorials with a great speaker lineup: Klaus-Robert Müller, Gabor Csányi, Stefan Chmiela, Leonardo Medrano Sandonas, Yury Lysogorskiy, Ilyes Batatia, Arslan Mazitov, Adil Kabylda

📅 13–15 July 2026
📍 Hybrid — Luxembourg (50 on-site) + online
✅ Free registration — sign up by 1 June: https://gpmlffworkshop.github.io/
🏆 Poster prizes from JACS, JCTC, and Chemical Science

Fun Fact of the Day

The Gaussian-and-plane-wave (GPW) strategy used in CP2K avoids explicit four-center integrals by mapping densities onto plane-wave grids while retaining localized Gaussian orbitals for the Kohn–Sham states. That hybridization is one major reason CP2K scales unusually well for condensed-phase DFT. 😉

ASH: a multi-scale, multi-theory modelling program ⚛️

Tired of being locked into one code's ecosystem? Meet ASH — a Python-based computational chemistry framework built around one elegant idea:

Separate the Hamiltonian from the job type.

Your QM/MM theory is one object. Your job (opt, freq, MD, scan, NEB...) is another. Mix and match freely.

What you can do:
- Single-point calculations
- Geometry optimizations
- Surface scans & relaxed PES
- Nudged Elastic Band (NEB) for reaction paths & TS
- Molecular dynamics
- Numerical frequencies
- QM/MM and ONIOM hybrid setups, built in minutes

Interfaces galore — drive your favorite code from a single Python script:
ORCA · xTB · CP2K · Psi4 · PySCF · ccpy · Dalton · MRCC · CFour · MNDO · TeraChem · QUICK · Gaussian · NWChem

Perfect for automating workflows, multi-scale modelling, and benchmarking across methods without rewriting half your pipeline.

🆓 Free & open-source on GitHub:
https://github.com/RagnarB83/ash

🚀 Try it live in Colab:
[ASH in Google Colab](https://colab.research.google.com/drive/11-FG7eTElCvcMNZiTIEXcdWjcR4YWRS-#scrollTo=ViPg1cGuck_a)

Ketcher 🧪

Ketcher is a free, open-source, web-based chemical structure editor for drawing molecules and reactions.

It includes:

• Fast 2D chemical structure drawing
• 3D structure visualization
• Template libraries
• Stereochemistry support
• Reaction drawing and atom-to-atom mapping
• R-group and S-group tools
• Structure cleanup
• CIP descriptor calculation
• Molecular formula and molecular weight calculation
• OCR for recognizing chemical structures from images
• Import/export support for Molfile, RXN, SMILES, CML, SDF, CDX/CDXML, InChI, InChIKey, and more
• SVG rendering for clean browser display
• Easy integration into custom web applications, including React-based projects

A serious tool for chemists, teachers, students, and developers. ⚗️

Stop paying for chemistry editors before checking what open source can already do. 🔓

Repository: https://github.com/epam/ketcher

Use it online: https://lifescience.opensource.epam.com/KetcherDemoSA/index.html