When chemical potential continuity fails: kinetic interface models for hydrogen isotope transport

Manuscript source and all simulation code for the paper by Remi Delaporte-Mathurin and James Dark (MIT).

The paper

Macroscopic hydrogen transport codes model material interfaces with local thermodynamic equilibrium (LTE), imposing continuity of chemical potential as a per-species algebraic constraint. That condition assumes fast interfacial equilibration, a single exchange pathway between the two sides, and a carrier species known in advance on each side. The paper argues that the last two are the more consequential failures, and that neither survives at a metal/molten salt interface.

We replace the constraint with reversible reaction channels on the interface obeying mass action, detailed balance fixing each ratio of rate constants from the same thermodynamic data that parameterises LTE. LTE is recovered as the fast-kinetics limit of a single channel, so the framework generalises LTE and does not compete with it. A Damköhler number delimits validity within a channel, a branching ratio between channels. The framework is implemented in FESTIM and applied to a representative nickel/FLiBe system, where hydrogen partitions kinetically between molecular and fluoride carriers and the apparent interfacial law drifts between Sieverts and Henry with loading and salt redox state.

Repository layout

Path	What it is
main.tex	The whole manuscript (elsarticle, single file).
references.bib	Bibliography. Read the verification notice near the top before editing.
scripts/	Every simulation, verification and figure script in the paper.

Every figure and every number quoted in the paper is produced by a script in scripts/, and each script writes its PDF next to itself; main.tex sets \graphicspath to those directories, so figures are included by bare filename. scripts/README.md maps each directory and each figure to the section it backs, and states the Damköhler convention shared by the scripts and the paper.

Broadly:

• 0-schematics/: illustrative figures, no solve.
• 1-first-order-interface/: Model 1, first-order exchange on a 1D two-slab problem, with its analytical steady state and the LTE-limit verification.
• 2-higher-order-reactions/: Model 2, recombination into a molecular carrier, the isotopologue channel set, and the analytical (Da, B) regime map.
• 3-competing-channels/: Model 3, recombination and fluorination competing on one interface, and the apparent-exponent verification.
• 4-hyperion/: the dimensional Ni/FLiBe application, all parameters pinned in parameters.py.
• reference/: read-only snapshot of the FESTIM interface module the paper describes.

Running the scripts

cd scripts
conda env create -f environment.yml
conda activate interface-models-env
pip install --no-deps git+https://github.com/festim-dev/FESTIM@interface-flux

The interface classes used here (InterfaceFlux, InterfaceReaction) live on the interface-flux branch of FESTIM and are not in a release yet, hence the extra install on top of the conda environment. --no-deps will not pull in io4dolfinx, which that branch also needs, so install it separately if the import fails.

Each script is standalone and self-describing: run it directly and it produces its figure and, where relevant, the CSV of the sweep it performs.

python 1-first-order-interface/verification_lte_limit.py

The committed CSVs are the outputs of those runs, so a rerun can be diffed against them.

Building the manuscript

latexmk -pdf main.tex

The class options are [preprint,11pt] for drafting and [5p] for submission.

Building requires a local LaTeX installation (elsarticle, siunitx v3); the manuscript is otherwise built on Overleaf, which this repository is synced to.

Licence

Released under the MIT licence, see LICENSE.