If anyone is interested in learning pymatgen in 2025, I posted a series of video tutorials on it. You can go from beginner to pymatgen wizard in just a couple of hours:

Predicting the expected power output of solar PV modules as they degrade can be challenging. The PV-Pro tool can model the internal state of a degraded module to provide accurate estimates of expected power (>17% improvement). Li et al, Renewable Energy

A roadmap on electronic structure methods, covering everything from theoretical methods to education. I contributed to workflows - challenges include interoperability of DFT codes & execution on heterogeneous resources. Blum et al, Electronic Structure

We apply SHAP analysis to ML models to understand how materials choice affects the overall thermal cycling resilience of solar photovoltaics. Si type, encapsulant thickness, busbar count & wafer thickness have the most impact. Chen et al., Applied Energy

A short perspective (I was a co-author) on opportunities for LLMs in alloy design, headed by Zongrui Pei. Even seemingly small things like alloy naming standardization may be useful! Pei et al, Nature Reviews Materials

For the last 8 years, I've been part of the DOE DuraMat consortium to develop 50-year lifetime solar modules. Our recent article summarizes DuraMat's goals and accomplishments, including our group's work on data & software: Owen-Bellini et al, PRX Energy

with Wei Tong & @EmoryChanNano. Also shout-out to @zackulissi for the collaborative computational screening effort for identifying nitrate reduction electrocatalysts prior to this work.

Health monitoring of photovoltaics often analyzes I-V curves, which need to be standardized against changes in temperature and irradiance. We present a more robust way to standardize I-V curves particularly for degraded modules. Li et al, Renewable Energy

We demonstrate that LLMs can accelerate data extraction from unstructured text (journal articles) using an iterative training procedure. Since the preprint, we show that Llama-2 results are close to GPT-3 results for the task. Dagdelen et al, Nat Comm

By text mining BiFeO3 synthesis, we quantify details typically missing from published synthesis descriptions (e.g. mixing temp & time). We next test if we can achieve phase pure synthesis using imputed values for past experiments. Cruse et al, Chem Mat

Jobflow is an expressive Python framework for programming computational workflows that can be executed using FireWorks. It underpins atomate2 workflows; read more in our recent paper! Next step is to finish up atomate2 ... Rosen et al, @JOSS_TheOJ

Supramolecular organio-ionic (ORION) electrolyte materials are viscoelastic solids at operating T but liquids >100C. This allows fabrication and recycling in liquid state & good performance even after recycling. w/@GroupHelms Bae et al, Science Advances

We investigate the synthesis of BiFeO3 thin films by text mining, experiments, & computations. All suggest impurities form at higher annealing temperatures & promoting bismutite intermediates via solvent choice can mitigate this. Abdelsamie et al, Matter

Job opening: Posdoc position to help us find new materials / eletrocatalysts for water purification (heavy metal, PFAS removal)! A background in computational catalysis would be ideal. Position start ASAP so hiring process will hopefully be rapid.

We use fine-tuned LLMs to extract synthesis recipes for Au nanorods, which is difficult to achieve via rules-based parsing. The final data set contains 332 complete synthesis procedures and includes measured aspect ratios. Walker et al, Digital Discovery

RT @scienceinparall: New episode 🎙️: Anubhav Jain @BerkeleyLab discusses #AI and #materials discovery, other creative #machinelearning appl…

Announcement: the DOE SCGSR program will sponsor current PhD students (U.S. citizen or permanent resident) to spend some time in our group at LBNL. If interested see the link below and contact me ASAP! Application deadline is Nov 8.

Announcement: reminder that we have an open position to work on automated labs; applications will be evaluated on a rolling basis. Hope to see your application! The job listing is here: More on the "A-lab" led by @cedergroup here:

We train an ML model to predict inorganic materials synthesis heating temperatures and times, based on an NLP-derived data set and 133 possible synthesis features. Trends like Tamman's rule extension are found. Huo et al., Chem Mat