Breakthrough

Multi-component PFAS ion exchange breakthrough simulator, built on the EPA's own Homogeneous Surface Diffusion Model (HSDM).

Most breakthrough tools model one PFAS compound in isolation. Real source water contains many, each binding the resin with a different strength. Short-chain compounds break through first and can be physically displaced off the resin by longer-chain compounds arriving later, pushing effluent temporarily above influent concentration. This tool models that competition directly, using EPA's transport equations, not a single-species curve fit.

What it does

Given the PFAS species you actually measured, their concentrations, and your column geometry, it predicts:

• Breakthrough time — the day effluent first crosses a regulatory limit or chosen threshold, per species
• Limiting species — which compound actually governs changeout, which is often not the one with the strictest regulatory limit
• Bed volumes — the standard unit for comparing IX systems independent of scale
• Cost — resin charge, changeouts per year, cost per 1,000 gallons treated
• Required bed depth — solved backwards from a target service life

Pages

Page	Purpose
Dashboard	Compliance status and bed-life summary at a glance
Simulator	Run a multi-species competitive breakthrough simulation
Design solver	Solve for the bed depth needed to hit a target service life
Bed & flow	Column geometry, resin specs, and commercial resin presets
Selectivity data	Full editable parameter table with provenance for every value
Selectivity calculator	Estimate selectivity for a species with no published value
Alkalinity converter	Convert pH + alkalinity to the bicarbonate input the model needs
Film transfer estimator	Gnielinski correlation for the film transfer coefficient, kL

Parameter provenance

Every selectivity coefficient (KxA) in the app is labeled by where it came from:

• EPA measured — five PFCAs (PFBA, PFPeA, PFHxA, PFHpA, PFOA), taken directly from EPA's published example workbook, Shiny-IEX/Examples/example_input_medium.xlsx, in the USEPA/Water_Treatment_Models repository. Molecular weights from PSDM/PFAS_properties.xlsx in the same repo.
• Derived — three PFSAs (PFBS, PFHxS, PFOS), set at 100x the same-chain-length PFCA value, following Liu et al., Strong Base Anion Exchange Selectivity of Nine Perfluoroalkyl Chemicals Relevant to Drinking Water, ACS ES&T Water (2023), which reports sulfonate selectivity roughly two orders of magnitude above carboxylates at equal chain length.
• Extrapolated — PFNA, from a log-linear regression fitted to EPA's own PFCA series (R² = 0.963, 0.403 log₁₀ units per CF₂).

This parameter set reproduces the published PFAS breakthrough order from NEWMOA's ion exchange case history review exactly: PFBA < PFPeA < PFHxA < PFHpA < PFOA < PFNA < PFBS < PFHxS < PFOS.

All values are editable in Selectivity data, and the Selectivity calculator page provides three independent ways to estimate a value that isn't published: from chain length, from a measured column result, or scaled from a reference compound.

Resin presets

Specifications for three commercial PFAS-selective resins, taken from manufacturer product data sheets:

Resin	Capacity	Matrix	Functional group
CalRes 2301 (Calgon Carbon)	0.51 eq/L min	Macroporous	Tributylamine
AmberLite PSR2 Plus (DuPont)	≥ 0.7 eq/L	Gel	Quaternary amine
Purofine PFA694E (Purolite)	not published¹	Gel	Complex amino

¹ Purolite's public data sheet does not list total exchange capacity. A placeholder value is used and flagged in the app; confirm with the vendor before relying on it.

Engine

The physics is unmodified EPA code: ixpy.hsdmix, solving the multi-component HSDM transport equations via orthogonal collocation. This app builds the input workbook in memory and calls that engine directly — no reimplementation of the underlying chemistry.

Running locally

git clone <this-repo>
cd <this-repo>
pip install -r requirements.txt
git clone https://github.com/USEPA/Water_Treatment_Models.git
streamlit run app.py

The EPA repository must be cloned alongside app.py since the model imports directly from it.

Files

app.py      Page routing and UI
core.py     Simulation logic, PFAS/resin parameter database
style.py    Theme
requirements.txt
.streamlit/config.toml   Streamlit theme configuration

Limitations

This is a screening tool, not a substitute for pilot testing. Derived and extrapolated selectivity values carry real uncertainty — the ordering they produce is validated against published data, but absolute bed-volume predictions for the derived sulfonates have not been checked against column data. Cost estimates cover resin media only and exclude vessels, installation, labor, and spent-resin disposal.