PhononBench – Phonon Spectrum Calculation Service

Fast phonon spectrum calculations and dynamical-stability evaluation for crystalline materials

Phonon Spectrum Calculation Service

PhononBench is currently offered as a free service for academic research. If this tool contributes to your work, we would greatly appreciate a citation of our related publications (see citation details below). The information collected below is used only for aggregated usage statistics and will not be made public. Your support through citations and acknowledgements directly encourages our continued development of fast and reliable computational tools.

User Information

eg:Renmin University of China - Xiao-Qi Han - hanxiaoqi@ruc.edu.cn

Step 1: Upload CIF File

Step 2: Download Results

Download the ZIP archive after the calculation is finished.

Typical turnaround time: a few minutes per structure (depending on system size).

Tips:

The labels of high-symmetry points in the phonon dispersion are automatically generated by Phonopy. For highly symmetric structures, the naming may differ from conventional DFT literature conventions (e.g., the Γ point may be labeled with a different symbol). This does not affect the correctness of the calculation, and users may relabel the high-symmetry points according to standard DFT conventions if needed.
PhononBench automatically performs structure relaxation and full phonon spectrum calculations following a standard DFT-based workflow, integrated with established tools such as ASE and Phonopy. To enable efficient large-scale evaluation, force calculations are accelerated using a validated universal machine-learning interatomic potential, while preserving consistency with conventional DFT phonon results.Once completed, PhononBench returns a comprehensive results package, including phonon dispersion plots, phonon density of states (DOS), phonon data files, and the relaxed CIF structure. The phonon band data are provided in plain-text `.dat` format for easy re-plotting using common tools such as matplotlib or Origin. If you encounter any issues during submission or calculation, please contact us at hanxiaoqi@ruc.edu.cn.
For systems containing fewer than 20 atoms, the calculation typically completes within approximately two minutes when the server is idle. During periods of high server load, tasks will be queued and the completion time may be longer; once finished, the results will be available for download. For systems with more than 40 atoms, please contact us by email to arrange a separate calculation due to computational resource limitations.
PhononBench uses a universal machine-learning interatomic potential to accelerate phonon calculations. The uMLIP performs well for most insulators and semiconductors, and its predictions are reliable in ~95% of cases as a rough stability indicator. However, it may still produce inaccuracies for certain materials or structures. Therefore, a “dynamically unstable” result here does not necessarily mean the material is truly unstable—DFT validation is recommended for critical cases.
Elements beyond the first 89 in the periodic table are not supported and will result in an error.
Only standard CIF files are supported. The first two lines of the CIF file should be similar to # generated using pymatgen and data_*.
If you encounter the message "Upload failed: no task ID returned", please verify that your CIF file follows the standard format.
Since its release, PhononBench has attracted widespread attention and positive feedback from both the academic and industrial communities worldwide. The platform has been continuously used by faculty members, students, and algorithm engineers from numerous internationally renowned research institutes, universities, and enterprises, who have highly evaluated its computational efficiency, stability, and reliability in practical applications.

We will review reported issues and handle them promptly.

How to Cite

PhononBench is a key enabling tool within our AI-driven inverse materials design paradigm, InvDesFlow. PhononBench was developed as a standalone platform and provides an essential capability for phonon-based dynamical stability assessment, which is a critical step in inverse materials discovery workflows. The overall InvDesFlow methodology is introduced in our review AI-Driven Inverse Design of Materials , and further developed in InvDesFlow and InvDesFlow-AL . If you use PhononBench or related components in your research, please consider citing these works.

@misc{han2025phononbench,
    title={PhononBench: A Large-Scale Phonon-Based Benchmark for Dynamical Stability in Crystal Generation},
    author={Xiao-Qi Han and Peng-Jie Guo and Ze-Feng Gao and Zhong-Yi Lu},
    year={2025},
    eprint={2512.21227},
    archivePrefix={arXiv},
    primaryClass={cond-mat.mtrl-sci},
    url={https://arxiv.org/abs/2512.21227}}
@article{InvDesFlow-AL,
    author = {Xiao-Qi Han and Peng-Jie Guo and Ze-Feng Gao and Hao Sun and Zhong-Yi Lu},
    title = {InvDesFlow-AL: active learning-based workflow for inverse design of functional materials},
    journal = {npj Computational Materials},
    year = {2025},
    volume = {11},
    number = {1},
    pages = {364},
    doi = {10.1038/s41524-025-01830-z},
    url = {https://doi.org/10.1038/s41524-025-01830-z},
    issn = {2057-3960},
    date = {2025/11/24}}
@article{InvDesFlow,
    title = {InvDesFlow: An AI-Driven Materials Inverse Design Workflow to Explore Possible High-Temperature Superconductors},
    journal = {Chin. Phys. Lett.},
    volume = {42},
    number = {4},
    pages = {047301},
    year = {2025},
    doi = {10.1088/0256-307X/42/4/047301},	
    url = {http://cpl.iphy.ac.cn/en/article/doi/10.1088/0256-307X/42/4/047301},
    author = {Xiao-Qi Han and Zhenfeng Ouyang and Peng-Jie Guo and Hao Sun and Ze-Feng Gao and Zhong-Yi Lu}}
@article{AI4Mreview,
    title = {AI-Driven Inverse Design of Materials: Past, Present, and Future},
    journal = {Chin. Phys. Lett.},
    volume = {42},
    number = {2},
    pages = {027403},
    year = {2025},
    doi = {10.1088/0256-307X/42/2/027403},	
    url = {http://cpl.iphy.ac.cn/en/article/doi/10.1088/0256-307X/42/2/027403},
    author = {Xiao-Qi Han and Xin-De Wang and Meng-Yuan Xu and Zhen Feng and Bo-Wen Yao and Peng-Jie Guo and Ze-Feng Gao and Zhong-Yi Lu}}