MESSAI.iO

How MESSAI reads your work

Let’s look at your papers together.

Type your name. We find your recent work, score each paper against the reproducibility minimum, and show you the small reporting fixes that make it comparable.

Public OpenAlex profile only · nothing is stored · a snapshot to build on, not a verdict.

Or pick your focus directly:

5.1 orders
Spread of reported areal power density (5th–95th percentile, 2,083 values from 394 papers). About 7 % of values exceed 20 W m⁻² and are likely unit or extraction errors.
11%
of power-density papers also carry the electrode area and the external load in our extractions — the minimum context to compare them.
0.2/3
Mean auto-checked score of 753 recent MFC papers (2023+); 88 % score 0 — partly because our extraction misses these fields, not only the papers.

Measured on the MESSAI corpus (staging), 2026-10-03. Counts are what our extraction captured — lower bounds on what papers report.

Worked example · one paper, end to end

What this page does with a real paper — including what it gets wrong.

Every number below comes from the corpus or is computed from it; nothing is illustrative. Corpus facts measured 2026-10-03; the score in step 2 is live.

  1. 01

    The paper

    Combined anaerobic–aerobic bioelectrochemical system for ammonium, sulfide and carbon removal from ethanol stillage J. Chem. Technol. Metall. 58(2), 2023 · corpus page · PDF ↗ · doi:10.59957/jctm.v58i2.53

    An anaerobic digester feeding a two-chamber bioelectrochemical cell, run in five modes: no circuit, MFC, and MEC at 0.6, 0.8 and 1.0 V. The extractor pulled 162 distinct facts from it.

  2. 02

    What the reproducibility scorer reads

    It sees 8 canonical parameters (substrate, current_density, power_density, currentDensity, powerDensity, cod_removal, ammonium_removal, h2_production_rate) — the scorer reads only facts linked to the parameter ontology, so the other facts are invisible to it. “Missing” on a report card therefore means not detected, not not reported.

    Study type: Wastewater treatment / resource recovery — the device alone (an MEC) would have filed it as “MEC — hydrogen / gases”, but what it measured is ammonium and COD removal, so the measured outcomes decide. Readiness against that checklist: 25%.

    Critical items not detected: hydraulic retention time, electrode surface area, external resistance / load.

  3. 03

    Checks the numbers have to survive

    • passStated removal vs Table 3. Every removal the text states matches (anode − cathode) / anode from Table 3 within 1 percentage point.
    • flagEvery value appears in its quote. The extractor recorded 75 % for 0.8 V (MEC), but its quote only says “followed by MEC0.8V” — no number. Table 3 gives 75.6 %.
    • failPower ÷ current gives a possible voltage. 50.1 W m⁻² at 133.6 mA m⁻² implies 375 V across one cell — impossible (a single cell stays near or below ~1 V). Most likely the power is 50.1 mW m⁻², which gives 0.38 V.
    • can’t tellRemoval attributable to a mechanism. Removal is measured across an aerobic cathode, where nitrification, NH₄⁺ migration and NH₃ stripping all remove ammonium. Without nitrite/nitrate in the cathode effluent, the electrode area and the flow, the share due to migration cannot be separated.
    • boundFaraday ceiling on NH₄⁺ migration. At 1.5–2.9 A m⁻² (MEC 1.0 V), migration can move at most 18.8–36.4 g N m⁻² d⁻¹, even if NH₄⁺ carried every charge.
    • flagEvery linked parameter means what it says. The one “H₂ production rate” the scorer sees is “7.9 L of biogas in 11 days” — a cumulative volume of mixed biogas linked to a hydrogen rate. It no longer decides the study type, but the link itself is wrong.
  4. 04

    The gap

    Removal climbs from 54 % with no circuit to 91 % at 1.0 V, so the circuit clearly matters. But removal is measured across an aerobic cathode, where nitrification, NH₄⁺ migration and NH₃ stripping all remove ammonium — and the facts extracted from the paper hold no nitrite or nitrate concentrations, no electrode area and no flow. From the corpus alone, no one can say how much of the removal the current did.

  5. 05

    The next experiment

    Isolate migration: a 2² factorial + 3 centre points over current density (2.5–10 A m⁻²) and the NH₄⁺ share of the feed’s cation charge (0.3–0.7), with nitrite, nitrate and cations measured in both chambers, the area and flow logged, and the nitrogen balance closed. Full run sheet on the nitrogen page →

  6. 06

    What we predict before it runs

    Migration alone can move at most 10.6 g N m⁻² d⁻¹ (90 %: 8.7–12.6) at 2.5 A m⁻² with 30 % NH₄⁺, and 91.9 (76.8–107.8) at 10 A m⁻² with 70 % — ceilings from Faraday’s law and ionic transport numbers, fitted to nothing. Energy per kg N falls from 5.7 to 2.6 kWh per volt of cell voltage as the NH₄⁺ share rises. Observed ÷ predicted is the migration efficiency this paper could not report.

    Pre-registered · run sheet SHA-256 1ac828f157891efe… · seed 20261003