Published December 20, 2023
| Version 0.1.0
Dataset
Open
FG-dataset
Description
The repository contains the "functional groups" FG-dataset used to train and test GAME-Net, a graph neural network for predicting the adsorption energy of organic closed-shell molecules on the most stable surface of 14 transition metals (paper). The dataset is provided here as an Atomic Simulation Environment (ASE) database with 6916 entries, representing the DFT adsorption structures simulated with VASP 5.4.4.
Besides the original data in the FG-dataset, the ASE database includes (i) the BM-dataset for testing GAME-Net with large-size adsorbed molecules, and (ii) the molecules of the FG-dataset adsorbed on the (110) and (100) surfaces of the fcc metals.
| Feature | Value | Notes |
| metals | Ag, Au, Cd, Co, Cu, Fe, Ir, Os, Pd, Pt, Rh, Ru, Zn (14) | 8 fcc, 5 hcp (Cd, Co, Os, Ru, Zn), and 1 bcc metal (Fe). |
| adsorbates' elements | C, H, O, N, S (5) | |
| surface facets | fcc(111), fcc(110), fcc(100), hcp(0001), bcc(110) | |
| calculator | VASP 5.4.4 | Main input settings stored in the ASE database |
The repo contains:
- The FGdataset.db with the compressed .zip version.
- README providing instructions for accessing and manipulating the database with ASE.
- Screenshots of command line examples in order to explore the database.
Files
ASE_docs1.png
Files
(109.8 MB)
| Name | Size | Download all |
|---|---|---|
|
md5:20971b0f8547c5cf599bba095bc039d6
|
261.8 kB | Preview Download |
|
md5:b605d6bf35c2eea594d5139ecbc18be0
|
250.0 kB | Preview Download |
|
md5:cd797dfa4548b84f9ae8a37b9c1a5fac
|
187.1 kB | Preview Download |
|
md5:636a0ac1e609a5803f762b0b720bbbfa
|
183.3 kB | Preview Download |
|
md5:779c4d63ca5e5f0f03edcb4111468dec
|
182.5 kB | Preview Download |
|
md5:31571d1a3e6c67e8f3c22b0f5ab01173
|
82.8 MB | Download |
|
md5:bad60dfa96090af2b138dffce456f478
|
25.9 MB | Preview Download |
|
md5:42caad1d44450e7636b211d8f8c72da4
|
5.4 kB | Preview Download |
Additional details
Related works
- Is new version of
- Dataset: 10.5281/zenodo.7750394 (DOI)
- Is published in
- Publication: 10.1038/s43588-023-00437-y (DOI)