How the Hydrogen Diffusion Database works

The Hydrogen Diffusion Database is an interactive tool for exploring hydrogen diffusivity data from the literature. You can filter datasets by material, method, or temperature range, compare series directly in the browser, and export plots or data for your own work.

Open Explorer Contribute Data
On this page: Overview Using the explorer Data quality and scope Contributing data Citation and contact

Overview

This database collects hydrogen diffusivity data from the literature and makes it easier to compare, visualize, and reuse. It is designed for quick exploration in the browser, but also for more serious use in modeling, validation, and research workflows.

Instead of digging through many separate papers just to compare a few values, you can filter the database, inspect the plotted series, and export the result in a form that is easy to cite and reuse.

Using the explorer

The main workflow is intentionally simple: choose what you want to see, plot it, adjust the display if needed, and export the result.

1. Filter or select series

Use the filter panel to narrow the dataset by source, material, method, or temperature window. If you already know what you want, you can also select series directly.

2. Plot the result

Press Plot Filtered to generate the plot. This gives you a clean view of the currently selected datasets and makes comparison much easier.

3. Adjust the display

Use the plot options to change units, scale, grid lines, labeling, monochrome mode, or axis limits, depending on whether you want a quick inspection or a publication-ready figure.

4. Export image or data

Export figures for reports and presentations, or download the filtered data for your own analysis. PNG is practical for images, while CSV or JSON is better for downstream work.

For first-time use, the quickest route is simply: filter -> plot -> export.

Filters, in plain language. Start broad and narrow down. Typical flow: pick a material class, then a temperature window, then the exact method or model type you want. The filters are there so you can recreate what a paper claims without digging through every PDF again.
  • Literature compilations: include or exclude curated collections so you can compare them against individual papers.
  • Material class / grade: group by alloy family and then drill into specific grades.
  • Temperature window + year: trim down time period and operating range before plotting.
  • Model type / measurement method: separate permeation experiments from diffusion fits or carrier-gas extraction data.
  • Chemical composition, reported as, studied effect, source: check how the data is framed and where it came from.
Plot options you will actually use. Toggle between Kelvin and °C, decide whether the plot shows fitted envelopes, and choose a numbered legend if you want to align the figure with the original sources. Axis limits and monochrome mode are there for quick exports into reports or slides.

Data quality and scope

The database is built to be useful, but also traceable. Each dataset is tied back to its literature source so that the provenance stays visible and easy to verify. We attach DOIs or source links wherever possible because transparency matters.

  • Only peer-reviewed, open-access sources are included, with citations attached directly to the data.
  • Flagged outliers or unconfirmed values can be shown separately if you want to inspect them.
  • The database does not extend Arrhenius fits beyond the valid range stated by the original source.
No extrapolation by the database. Curves are plotted only within the temperature range that is explicitly supported by the original source. If a publication does not state a valid range clearly enough, that dataset may be excluded or handled conservatively. When a paper does not state a range but the context is obvious, we apply the most conservative assumption, Examples: If a paper mentions a 150 °C experimental method but never states the validity range for the Arrhenius fit, it was excluded in the initial dataset. Permeation experiments were assumed to be at room temperature when otherwise unspecified because from the paper this was obvious and a editorial oversight. Author-direct submissions are most welcome as this provides the most reliable validity range explicitly; if values look like outliers or units seem off, we flag them and get back to you for clarification.

Contributing data

Contributions are welcome. The preferred route is the contribution form, because it keeps submissions structured and formats them in a way that plugs straight into the site.

To keep the database trustworthy and reusable, submissions should come from peer-reviewed open-access sources and include clear publication metadata, model parameters, and the valid temperature range whenever possible.

If you want to send data by email, that is fine too — just know that the form is much prefered because it is much faster and easier for us to process because it auto-prepares the fields we need. For questions or corrections, email is preferred.

Citation and contact

If you use the database in your own work, please cite the database or website and also cite the original publications behind the datasets you use. When companion papers are available, add here.

The public database archive is available on Zenodo under CC BY 4.0: https://doi.org/10.5281/zenodo.18980187

If you spot an error, missing context, or a citation issue, please contact Denis@Czeskleba.com. I am also activly looking for collaborators to review and audit already included papers.