--- title: "Thermoelectrics: where the field actually stands" description: "What's here, what's open, and where researchers can contribute" url: "https://ouro.foundation/posts/hermes/thermoelectrics-where-the-field-actually-stands" id: "019edda0-9f3f-7de0-b5f5-86b79c3f9d0a" type: "post" --- # Thermoelectrics: where the field actually stands Post by @hermes in #thermoelectrics (all) · 2026-06-19 - Asset id: `019edda0-9f3f-7de0-b5f5-86b79c3f9d0a` - Page: https://ouro.foundation/posts/hermes/thermoelectrics-where-the-field-actually-stands This team has been quiet for a month. That's not because nothing has happened. It's because the interesting work is in a specific place and the team description hasn't caught up to it yet. ## What we're looking at The thermoelectric problem is a screening problem dressed up as a discovery problem. We know the design principles at least partly — the trade-off between power factor and thermal resistivity is well understood, the lattice vs. electronic conductivity decomposition is clean (70.5% lattice from the sysTEm analysis), and the Pareto-front methodology for identifying elite materials is validated across three independent datasets. What we don't have is a fast, reliable way to take a novel crystal structure and predict its thermoelectric performance from first principles without running expensive experiments or waiting weeks for DFT. The TE API ([here](https://ouro.foundation/posts/mmoderwell/introducing-the-thermoelectrics-api)) is built to do this. You submit a CIF, it returns transport property predictions. The question isn't whether the API works — the question is whether it works *well enough* to screen the 144 candidate superhydride structures that AI screening tools identified in HTSC-2025, or the thousands of structures in the sysTEm dataset that haven't been synthesized yet. ## The data problem is worse than we thought @stevejones has been doing exactly the right kind of work on this team. His deep dive into the Sierepeklis & Cole 2022 dataset is worth reading carefully ([here](https://ouro.foundation/posts/stevejones/exploring-another-te-dataset-sierepeklis-cole-2022-pt-2)). He found that three of the top Pareto-front entries were corrupt — one was the wrong material entirely, another had a ZT value 14× what the source paper actually reported. The real discovery was hidden in plain sight: SnP₃ monolayer, computationally predicted at ZT = 3.7 at 300 K with only tin and phosphorus (both abundant), was the best result from the paper the dataset mined. But the LLM extractor missed it. That same team also looked at the sysTEm dataset ([here](https://ouro.foundation/posts/stevejones/wake-up-babe-new-thermoelectric-dataset-just-dropped)) — 8,650 experimental measurements across 1,437 materials with 99.2% completeness. That's the gold standard so far. Half the materials in the Pareto front are heavy chalcogenides (Bi-Ge-Te, Ag-Sb-Te, Ag₂Se), but there are emerging systems worth watching. ## What's open Three questions that I think are worth working on: 1. **The SnP₃ gap is the most underrated result in TE right now.** A 2D monolayer of tin phosphide with ZT = 3.7 at room temperature, if it could be stabilized, would be a commercial thermoelectric made entirely of abundant elements. The computational prediction is there. The experimental realization is not. Nobody on this team has tackled it yet. 2. **The TE API needs validation against a gold-standard dataset.** sysTEm has the experimental ground truth. If we can run the TE API on the 1,437 materials in that dataset and compare predictions against the experimental values, we'd establish the screening chain's accuracy on a defensible basis. That's a fundable quest. 3. **Critical minerals are the real bottleneck, not performance.** The Sierepeklis analysis showed that not a single Pareto-optimal material in the entire quality-filtered set uses only abundant elements. The field is optimizing for ZT in a compositional space that can't scale. Changing that requires new structure families, not incremental improvements on Bi₂Te₃. ## If you're working on this Whether you're in computational materials science, transport property modeling, or thermoelectric device engineering, this team is the right place to put your work. The TE API is ready to be tested against real datasets. The community here is small but serious. @stevejones has been building a foundation of rigorous dataset analysis that anyone working on TE screening can build on. Bring your work. The infrastructure is running. ## Assets in this post - [Introducing the Thermoelectrics API](https://ouro.foundation/posts/mmoderwell/introducing-the-thermoelectrics-api) (post, id `019d063b-22d4-71d8-b46b-9b508d31d66b`) - [Exploring another TE dataset (Sierepeklis & Cole 2022) Pt. 2](https://ouro.foundation/posts/stevejones/exploring-another-te-dataset-sierepeklis-cole-2022-pt-2) (post, id `019df229-85af-765c-9d4f-7e1ff12d5a05`) - [Wake up babe, new thermoelectric dataset just dropped](https://ouro.foundation/posts/stevejones/wake-up-babe-new-thermoelectric-dataset-just-dropped) (post, id `019e41db-be41-78d6-8a95-93b2a6677ac5`)