← All decksPrepared for Captura · DOC pilot simulator · July 2026

A working model of your BPMED plant, driven by our engine.

This is MESSAI's ion-transport, membrane, and techno-economic engine pointed at Captura's bipolar-membrane electrodialysis process. Move the levers your engineers actually tune. The model recomputes CO₂ capture, energy intensity, and cost per ton live, and finds the cost-minimizing operating point. The pilot deliverable is this, calibrated on your data.

Illustrative engine shipped · coefficients not yet calibrated on Captura data

01 · Operating pointEight levers, four outputs

—$/t CO₂

Levelized cost

band —

—kWh/t

Energy intensity

— plant draw

—t/yr

CO₂ captured

— membrane

—V/cell pair

Stack voltage

thermo + ohmic

Process

Current density 1000 A/m²

Current efficiency 85 %

Feedstock · seawater

Dissolved inorganic C 2.10 mM

Membrane stack

Area-specific resistance 5.0 Ω·cm²

Stack cost $1200/m²

Membrane lifetime 6.0 yr

Economics

Renewable electricity $40/MWh

Plant scale 1,000 t/yr

02 · Cost curveLevelized cost vs. current density

Low current density is membrane-capex limited: the stack is spread over too little output. High current density is energy limited: ohmic voltage dominates. The optimum sits where the two trade off.

cost curve (±25 % band) operating point cost-minimizing optimum

Levelized cost, $/t CO₂, against stack current density in A/m². Shaded band is an illustrative ±25 % placeholder for the conformal interval.

—

Cost component	$/t CO₂	Share
Energy · stack electrical	—	—
Membrane stack · capex over lifetime	—	—
Balance of plant · fixed opex, scale economies	—	—

03 · Under the hoodShipped MESSAI modules

The pilot recalibrates each of these against Captura's pilot data.

Module	What it does here
`transport · nernst-planck`	1-D ion-transport + carbonate speciation solver. Maps current density and seawater DIC to acid delivery and CO₂ release, the acidification/degas step.
`component-catalog · membrane`	Typed BPM / CEM / AEM library. ASR, permselectivity and lifetime drive stack voltage and energy intensity.
`ml · priors + conformal`	Calibrated uncertainty. The ± band is a placeholder here; the pilot replaces it with conformal intervals fit on your measurements.
`tea · process`	Techno-economic layer. Resolves the energy-vs-membrane trade-off into $/ton and locates the cost-minimizing current density.

What this is not

Illustrative. The engine is shipped; the coefficients are not yet calibrated on Captura data.
Fixed assumptions: 0.80 V per cell pair BPM water-dissociation potential, 1.10 mol H⁺ per mol CO₂ before utilisation, balance-of-plant $130/t at 1,000 t/yr with a 0.22 scale exponent.
The ±25 % band is a placeholder, not a fitted conformal interval.
Seawater DIC range 1.8–2.4 mM maps to 75–85 % acid utilisation by a linear rule, not a speciation fit.