Know what the cellwill do beforeyou machine it.
PROTONYX solves the electrochemistry of your design — Tafel kinetics, Nernst thermodynamics, membrane and mass transport, Faraday yield — and returns a polarisation curve, a loss budget in millivolts, and a written scientific assessment. Before fabrication, not after.
| 01 | Capital in a prototype that misses spec | £50–200k | Machining, membrane, catalyst, assembly, test rig time |
| 02 | Calendar lost per rebuild cycle | 6 weeks | And the diagnosis is usually inconclusive |
| 03 | Loss terms quantified in millivolts | 5 | Anode and cathode activation, membrane ohmic, interfacial contact, mass transport |
| 04 | Subsystem mechanisms modelled on top | 4 | Acoustic, electrostatic, magnetic and mist — each acting through the five |
| 05 | Time to a complete assessment | < 30 s |
Figures are indicative of single-cell R&D builds at 200–800 cm² active area. Your own history will be more accurate — the console keeps it.
Solved, not estimated
Tafel kinetics at both electrodes. Nernst corrected for temperature and differential pressure. Membrane conductivity is yours to supply, and we tell you when the figure you entered is the 30 °C reference value rather than one for your operating temperature. Faraday yield on your active area.
A loss budget you can act on
Five loss terms — anode activation, cathode activation, membrane ohmic, interfacial contact, mass transport — each in millivolts at your target current density, ranked, with the component responsible named. The four subsystem mechanisms are modelled as their effect on those terms, not as separate ones.
Interpretation, in writing
The computed numbers are read back as a senior electrochemist would read them: what dominates, why your material choices caused it, what to change first, and what will degrade.
Ranked by leverage
Every input re-solved against realistic alternatives and ranked by millivolts of swing — so you know which decisions matter and which you can stop arguing about. Paste the rig sweep afterwards and the model grades itself.
Specify
Cell, membrane, anode, cathode, current collectors, operating point, scale-up and targets — plus the acoustic, electrostatic and magnetic subsystems where your plan includes them. Every field is optional; what you leave blank is filled from literature and declared on the result.
Solve
Sixty-point polarisation sweep scaled to your operating point, with every overpotential term resolved separately.
Interpret
The numbers are explained against your components, with ranked changes and expected deltas in mV or kWh/kg.
Iterate, then validate
Overlay any number of variants, fabricate the one you can already defend, then paste the measured sweep back in and see the residual in millivolts.
Standard PEM, and the ones nobody models
Mist and aerosol feed, acoustic streaming, electret dipole alignment, magnetic OER facilitation. Treated as physics with literature-referenced corrections and an explicit confidence grade — not silently ignored, and not oversold.
What is most likely to happen if you build this?
Answered with computed physics, a ranked loss budget and a written verdict — before the first component is machined.