Design a bioelectrochemical resource-recovery system

Start with a wastewater stream. Get a ranked design from the lab catalog via the 5-physics-family router, open its dossier and P&ID sheet, and see a real ML prediction with conformal bounds, an OOD flag, and a prior-trust badge. Trace every number to its source artifact.

reads from
  • /api/ml/predict — 15-block response
  • holdout-coverage-2026-05-21.json — 940 OOS obs
  • calibration.json — 517 obs, random split (not paper-disjoint)
  • wastewater/*-baseline.json — 6 process-TEAs
  • @messai/component-catalog — 6 lab-catalog twins
  • @messai/pid-schematic — ISA-5.1 sheets per twin

Section 1 · Influent characterisation

Define the wastewater stream

Six industrial archetypes or set parameters within their literature ranges. Each slider shows hard bounds, typical band, sub-ranges by facility type, and the active preset's target value.

Routed family · anodic oxidation
5,000 m³/day
10 m³/day100,000 m³/day
Literature
100–50,000 m³/day
Metcalf & Eddy 5e Tbl 3-15
Median
5,000 m³/day
Preset
5,000 m³/day
Brewery effluent
2,500 mg/L
100 mg/L25,000 mg/L
Literature
200–15,000 mg/L
Metcalf & Eddy / Renou 2008
Median
2,500 mg/L
Preset
2,500 mg/L
Brewery effluent
40 mg/L
0 mg/L2,500 mg/L
Literature
10–1,500 mg/L
Metcalf & Eddy / Tervahauta 2014
Median
40 mg/L
Preset
40 mg/L
Brewery effluent
8 mg/L
0 mg/L200 mg/L
Literature
2–80 mg/L
Metcalf & Eddy 5e
Median
8 mg/L
Preset
8 mg/L
Brewery effluent
25.0 °C
5.0 °C45.0 °C
Literature
15.0–35.0 °C
Logan 2008 §4.2
Median
25.0 °C
Preset
25.0 °C
Brewery effluent
6.8
4.0 9.5
Literature
6.0–8.5
Patil 2011 / Logan 2008
Median
7.0
Preset
6.8
Brewery effluent

Section 2 · System suggestion · lab catalog designs

Pick a coupled architecture

Ranked by influent fit using the 5-physics-family router. Every card is a digital twin from the lab catalog; click one to drive every downstream panel.

Section 2b · System dossier · from the lab catalog

MFC + Anaerobic Digester (electricity from wastewater)

The Bio-Battery Train

PilotscaleTRL 4–50.5 – 50 m³ MFC stack downstream of a 10 000 m³ ADAnodic oxidationProvisionalWaterElectricityBiogas

A two-stage flowsheet that pairs an anaerobic digester (for bulk COD reduction + biogas recovery) with a downstream microbial fuel cell that polishes residual COD into direct electrical current. The AD does the thermodynamic heavy lifting; the MFC monetises what’s left as electrons rather than gas, with no biogas-handling burden on the back end.

Key differentiator

AD alone leaves 25–50 % of the influent COD behind in the digestate. An MFC polish step recovers a portion of that residual COD as electricity at electrode-surface efficiencies that exceed dark-fermentation H2 routes. Bridges the parameter-sweep page’s Butler-Volmer + Monod outputs (sweep export → MESSanUnit) directly into a full-flowsheet TEA.

How it works

Influent enters the AD; biogas exits the dome to CHP; digestate flows by gravity into the MFC anode chamber where exoelectrogens (Geobacter / mixed consortia) attach to a carbon-cloth or graphite-felt anode. Substrate oxidation releases electrons that travel through an external circuit to an air cathode (typical) and protons migrate through a separator. Power is dumped to a buck-boost charger that conditions the MFC output to grid-tie or battery voltage.

Performance envelope

MFC power density0.1–2 W/m² anode area
MFC coulombic efficiency15–40 % on real wastewater
Combined COD removal75–95 % (AD + MFC)
Electrical output1–10 GWh/yr at 500 m² anode
LCoE (current scale)$100–500/kWh (not yet competitive)

Research status

MFCs remain at TRL 4–5 outside niche sensor applications. The MFC+AD combination is published (Logan group, Penn State 2010s) but no pilot plants are continuously operated. Active research: scaling anode area while preserving CE, cheaper cathode catalysts, hybrid AD-MFC operating strategies.

Key parameters

MFC anode area50–10 000 m²
MFC power density0.1–2 W/m²
MFC coulombic efficiency15–40 %
AD HRT5–15 d (shorter — AD is upstream)

Reactor geometry

Left: cylindrical AD tank (see "wastewater-ad" for sizing). Right: rectangular MFC chamber with planar anode (left wall) and air cathode (right wall, typically Pt/C on carbon cloth or activated-carbon-on-stainless-mesh). Top: external circuit with adjustable load resistor; LED indicator on the wire visualises current.

Applications

  • Wastewater treatment with energy recovery (food / dairy / brewery effluents)
  • Decentralised treatment at sites without grid access
  • BOD sensors at WWTP outfalls (MFC voltage tracks BOD)
  • Educational / public-engagement reactors
  • Combined heat (AD biogas) + power (MFC) micro-grids

Scale

Pilotscale

TRL

4–5

Active volume

0.5 – 50 m³ MFC stack downstream of a 10 000 m³ AD

Loading 3D reactor scene…

Section 4 · Live prediction · /api/ml/predict

Real ML stack, honest bounds

idlein-corpuscal · n/a
Power density
mW/m²
COD removal
%
Predictor doesn't emit a COD block on this path. See the I/O analysis section below for a population-prior estimate.
Voltage
V
Current density
A/m²

Section 5 · Recovery balance

What this system actually recovers

Daily mass + energy flux from first-principles balance on the influent and the live prediction. Hover any band for source.

Net electricity · 2.544 kWh/m³
COD removed
10000 kg/d
80% removal
MES electricity
2.4 kWh/d
from MES cells
Biogas → CHP
12720 kWh/d
35% CHP electrical eff.
Nutrients
0.0 kg/d
struvite + NH₃
  • · Using literature-midpoint fallback for power density.
  • · COD removal from archetype default (80%).

Section 7 · Design recommendations

Apply a recommended change

Each lift is a fitted corpus effect with its n and 95 % CI. Stage one to see it as a ghost overlay in 3D; apply to commit it to the scenario.

live prediction unavailable — lifts shown as relative % only

Section 7b · Raise confidence

33 evidence gaps for MFC + Anaerobic Digestion

Each row is a design lever the corpus cannot yet quantify with high confidence, the study that would close it, and the literature search that looks for it.

  • external resistancepower density128 papers · cross-paper · effect not distinguishable from zero · low
    Suggested study: vary external resistance; hold temperature, pH, HRT, COD_in, anode material fixed; measure power density + coulombic efficiency; ≥ 3 conditions per study.
    openalex: external resistance "microbial fuel cell" power density
  • pHpower density110 papers · cross-paper · effect not distinguishable from zero · low
    Suggested study: vary anolyte pH; hold temperature, HRT, COD_in, R_ext, anode material fixed; measure power density + coulombic efficiency; ≥ 3 conditions per study.
    openalex: anolyte pH "microbial fuel cell" power density
  • temperatureCOD removal108 papers · cross-paper · effect not distinguishable from zero · medium
    Suggested study: vary reactor temperature; hold pH, HRT, COD_in, R_ext, anode material fixed; measure COD removal + coulombic efficiency; ≥ 3 conditions per study.
    openalex: reactor temperature "microbial fuel cell" COD removal
  • pHCOD removal105 papers · cross-paper · effect not distinguishable from zero · low
    Suggested study: vary anolyte pH; hold temperature, HRT, COD_in, R_ext, anode material fixed; measure COD removal + coulombic efficiency; ≥ 3 conditions per study.
    openalex: anolyte pH "microbial fuel cell" COD removal
  • influent CODCOD removal94 papers · cross-paper · effect not distinguishable from zero · low
    Suggested study: vary influent COD concentration; hold temperature, pH, HRT, R_ext, anode material fixed; measure COD removal + coulombic efficiency; ≥ 3 conditions per study.
    openalex: influent COD concentration "microbial fuel cell" COD removal
  • influent CODcoulombic efficiency62 papers · cross-paper · effect not distinguishable from zero · low
    Suggested study: vary influent COD concentration; hold temperature, pH, HRT, R_ext, anode material fixed; measure coulombic efficiency + power density; ≥ 3 conditions per study.
    openalex: influent COD concentration "microbial fuel cell" coulombic efficiency

Section 7c · What to measure

Measurement plan for MFC + Anaerobic Digestion

Ranked from 255 MFC papers that report a recovery target: levers a study should vary, the context variables the corpus expects alongside them, and the outcomes to report. Coverage and correlation are descriptive; only the effect column is a fitted causal-grade estimate.

parameterrolereported infitted effectgaps it closesρ vs target (pooled)
anode materiallever0% · 0+0.017 → power density · low · n=564 · hold in 64
hydraulic retention timehlever12% · 30−0.029 → COD removal · low · n=4315 · hold in 113-0.25 (11)
pHlever33% · 85−0.086 → ammonium removal · medium · n=1014 · hold in 114+0.55 (15)
temperature°Clever22% · 56+0.0087 → COD removal · medium · n=10814 · hold in 114-0.25 (103)
COD Concentrationmg/Llever34% · 86+4.6e-7 → COD removal · low · n=9413 · hold in 115+0.43 (69)
External LoadΩlever28% · 71−0.0000037 → coulombic efficiency · medium · n=608 · hold in 120+0.78 (8)
Power DensitymW/cm²covariate54% · 137+0.74 (30)
voltageVcovariate44% · 111+0.48 (13)
Open Circuit VoltageVcovariate35% · 90-0.45 (10)
Current DensitymA/cm²covariate34% · 87+0.56 (26)
Reactor VolumemLcovariate31% · 79-0.20 (73)
Internal ResistanceΩcovariate30% · 77-0.71 (12)
Cod Removal%outcome83% · 211
coulombic efficiency%outcome43% · 110
Ammonium-N Removed%outcome9% · 22
Phosphorus Removal%outcome4% · 10

source: apps/web/public/data/computed/research/measure-priorities.json · 2026-09-12

Section 8 · Sensitivity ladder

Which knob moves the needle

Within-paper Bayesian effects from within-paper-effects.json. Each β is the population-level effect on the target after partialling out paper-level confounds. Sign matters — green increases, red decreases.

Targets · cod_removal · power_density
Anode Material:Graphite-BrushPower Density
β = -3.55
n = 2, [-21.89, 14.79]
Anode Material:Cnt-MatPower Density
β = -0.39
n = 3, [-2.75, 1.97]
Anode Material:Graphene-FoamPower Density
β = +0.33
n = 4, [-0.15, 0.82]
HrtCod Removal
β = +0.26
n = 8, [-0.51, 1.02]
Temperature CPower Density
β = -0.25
n = 6, [-0.65, 0.11]
PhPower Density
β = -0.11
n = 110, [-0.27, 0.07]
Hrt HCod Removal
β = -0.03
n = 43, [-0.08, 0.02]
PHCod Removal
β = +0.02
n = 33, [-0.00, 0.05]

Mechanistic closure · direction-validated levers

Does the physics closure predict the direction (sign) of a within-design change better than a coin flip (50%)? These are the only axes that validated. Direction agreement only — not magnitude.

Substrate → power density
63%
dir. acc · closure forward sweep
pH → power density
62%
dir. acc · closure forward sweep
External resistance → current density
71–78%
dir. acc · load-line operating point

Not shown: current↔substrate and current↔pH are anti-skillful (the closure predicts the wrong direction), so we exclude them rather than surface a misleading bar. The richer interactive version lives at /lab/design.

Section 9 · Calibration · honest caveats

What this model is — and isn't — calibrated for

⚠ Conformal coverage guarantee is provisional

The split-conformal q̂ were calibrated against holdout-priors-2026-05-21, but the served predictor is v2-sota-2026-05-15. The interval math (log-space + physical clamp) is correct, but the marginal-coverage guarantee holds only once the constants are regenerated against the served fit. Every /api/ml/predict response carries this same version_check.ok = false.

OOS 95% coverage
97.98% target ≥ 93%
n = 940 held-out
MFC stratum coverage
no data yet · target ≥ 90%
No MFC stratum in the current calibration artifact yet; using the hierarchical prior with wider bounds until paired holdout obs land.
Expected calibration error
no data yet · target ≤ 0.05
ECE not yet computed for this corpus — awaiting calibration artifact refresh.
Parameter
System
Coverage
n
power_density_areal
MFC
97.9%
234
current_density_areal
MFC
97.5%
163
current_density_areal
MEC
100.0%
69
coulombic_efficiency
MFC
100.0%
135
coulombic_efficiency
MEC
100.0%
64
cod_removal
MFC
97.3%
220
cod_removal
MEC
92.7%
55

What we don't claim

  • MDC, MES, MNRC, MMRC, MBES are not in the 97.98% OOS holdout — analytical predictors only.
  • TEA numbers are literature midpoints; not a contracted quote. Site engineering adds 20–40%.
  • COD removal predictions are coarse for high-strength industrial streams with toxic shocks.
  • No predictions of micropollutant removal, antibiotic resistance, or pathogen fate.
  • Long-term biofouling, electrode degradation > 12 months are absent from the corpus.

Section 10 · Mini-TEA · MFC + Anaerobic Digestion

What it costs

From the archetype's Process-TEA baseline. Discount rate 0.1, lifetime 15 yr. Honest disclaimer: literature midpoints, not a contracted quote.

Scaled · 1.00× ref flow (5,000 m³/d)
CapEx
$60.6M
Annual OpEx
$4.2M
LCOE
$5.15/kWh
LCO-water
$5.34/m³
NPV @ life
−$92.0M
Payback
No payback within 15-yr life
GWP
-0.28 kgCO₂e/m³
Fossil energy
-3.34 MJ/m³
  • · Electrode capex $1,200/m² (carbon-cloth, Logan 2008 inflated).
  • · Performance inputs are stub literature midpoints — wire to a sweep export for real numbers.
  • · Higher 10 % discount rate reflects MFC scale-up risk (Santoro 2017).

Section 11 · For industrial partners · your stream in, value out

What your stream could become

Pick the row that matches what you discharge or emit. Each wastewater row is a modeled system with a published baseline and selects that design above; each prediction comes with an interval and an honest flag when your stream sits outside the data we have.

Your inputSystemWhat comes outNotes
MFC + anaerobic digestion
The Bio-Battery Train · TRL 4–5
waterelectricitybiogas
Brewery, dairy, food processing. Most data-rich archetype.
MEC + anaerobic digestion
The Hydrogen Train · TRL 3–4
waterH₂biogas
~0.5 V applied; hydrogen instead of power.
MFC + struvite side-stream
The Phosphorus Mine · TRL 8–9
waterelectricitystruvite (P)
MgCl₂ dosing; 75 % P recovery in the baseline model.
MFC + NH₃ stripping
The Ammonia Stripper · TRL 8–9
waterelectricityNH₃ (N)
N-fertilizer feedstock; 65 % N recovery in the baseline model.
MFC + electrochemical polishing
The Color Eater · TRL 7–8
waterelectricitymetals
Mining, plating, electronics effluent.
Conventional activated sludge
The Industry Workhorse · TRL 9
water only
The incumbent baseline every case is priced against.
Ambient air
Bipolar-membrane electrodialysis DAC
CO₂ streamcompressed CO₂
Presets at 0.15 t/yr bench, 1 kt/yr pilot, 100 kt/yr commercial. Same core, different inlet.
Mine ventilation methane
Methanotrophic biofilter
CH₄ oxidized
Early model, not yet calibrated. Ask if relevant.

What a feasibility study looks like

Six to eight weeks from stream data to a decision-grade answer. The output is not a brochure; it is the number, its interval, the plant economics, and the one or two bench measurements that would move the answer most.

Weeks 1–2

Characterize

You share flow, COD, N, P, metals, temperature, and what you pay today. We map it to the corpus and flag where your stream is outside the data.

Weeks 2–4

Route and predict

The stream is routed to the right physics family and archetype. Every output carries a calibrated interval and a prior-trust badge.

Weeks 4–6

Price it

CAPEX, OPEX, levelized cost or net present value over 25 years, against your incumbent. Cost inputs are labeled literature or assumed, line by line.

Weeks 6–8

Name the test

A ranked measurement plan: the one or two bench measurements that would move the answer most, with the variance each one explains.

What we will tell you before you ask

  • Waste-to-value predictions are strongest for MFC systems, where held-out coverage was measured. Sparse archetypes carry a visible out-of-distribution flag in the prediction strip above.
  • Recovery fractions in the balance are baseline-model constants, not measurements of your stream: 75 % P as struvite and 65 % N as NH₃ until a partner sample replaces them.
  • Cost inputs are labeled literature or assumed, line by line in the mini-TEA. A partner quote replaces an assumption with a fact.
  • The DAC and mine-methane models have not been calibrated against a running plant. They are listed so you know they exist, and they say so on their own pages.

Partners are the calibration. The first studies are priced accordingly.

Start with one stream

Send a flow rate and a lab sheet. Within a week you get a routed archetype, a first interval, and a scope for the study.

6 modeled wastewater designs25-year TEAconformal intervals
founders@messai.io

every number on this page traces to a source file. no black boxes.

Spec: docs/superpowers/specs/2026-05-26-resource-recovery-demo-design.md. Live audit dashboard at /proof. Design catalog at /lab/catalog. Lab CAD + parametric editor at /lab.