Resonant Ionic Momentum Transfer — a proposed solid-state marine propulsion concept, openly red-teamed and analytically falsified by its own project
This project proposed RIMT: converting the wetted hull of a marine vessel into a distributed solid-state engine by applying a MHz-frequency travelling-wave potential through a sub-surface micro-electrode lattice, "ratcheting" the ions of seawater's Electrical Double Layer (EDL) to generate thrust. The disclosure (v1.0–v1.2) was published with an unusual posture: every load-bearing assumption was flagged, a companion adversarial critique was commissioned against the project's own interest, and the author committed in advance to publishing a negative result.
The negative result arrived — from the project itself, analytically, at zero hardware cost. A follow-up first-principles audit (2026-06-12) resolved the two questions the critique had left open as Fatal risks, and both resolve against ship-scale viability. The whitepaper is now v2.0 — Proposal and Analytical Falsification: one continuous record of the concept, the red-team, the physics that killed it, and what survives.
-
Seawater is a conductor at 2 MHz — the driving field is screened away. Seawater's charge-relaxation frequency is ≈ 1.08 GHz; at the 2 MHz carrier, conduction dominates displacement by ≈ 540×, so the liquid only sustains the tangential field needed to carry the capacitively-available current: E∥ ≈ 2 kV/m, not the MV/m-scale electrostatic estimate. The physically consistent EDL coupling is α ≈ 1.8×10⁻⁴ — numerically the two-capacitor divider the paper had disclosed as its pessimistic "floor", because the two are the same physics. The consistent electro-osmotic slip is ≈ 40 µm/s (Péclet number ≈ 0.12: thermal diffusion beats the directed transport).
-
Electro-osmosis is a velocity boundary condition, not a deliverable stress — and slip-surface propulsion has a hard speed ceiling. A control-volume momentum balance shows deliverable thrust equals the momentum flux exported into the wake: ≈ 0.45 N/m² even granting the model's own optimistic slip velocity — far below the design target. The corollary is a "conveyor-belt" theorem: self-propulsion equilibrium gives U_max ≈ 2·v_slip, independent of hull area — centimetres per second here. The only electro-osmotically propelled vehicle in the literature (Hansen et al., a 5 kg UUV) reached 2.2 cm/s, exactly the scale this analysis predicts.
The arguments are independent (one electrostatics, one fluid mechanics) and each alone closes the propulsion case at first order. A lever-by-lever sweep (frequency, waveform, pitch, dielectric, voltage) finds no escape within the disclosed technology envelope. The proposed bench experiment remains in the paper as the empirical confirmation of the predicted null (≪ 1 mN on the test plate — a µN-class balance is required).
- Anti-biofouling active hull skin — the most viable use of the engineered hardware (electrode lattice + dielectric coating + GaN drive): micro-scale EDL agitation as a fouling-prevention mechanism, decoupled from any thrust claim. Concept note:
biofouling/RIMT-antifouling-concept.md. - A cm/s micro-propulsion niche — consistent with the Hansen datum; positioning of small neutrally-buoyant platforms, not ships.
- The conveyor-belt ceiling theorem (U_max ≈ 2·v_slip) — a portable result for any proposed slip-surface propulsor.
- The closed-form loss algebra of the tile model, and the red-team methodology itself, which worked exactly as intended.
Before the falsification, the project commissioned and published a steelmanned attempt to break its own concept: critique/RIMT-adversarial-critique.md. Its two Fatal-risk findings — the thrust-density momentum cross-check (§2.1) and seawater concentration-suppression of the coupling (§2.3) — are precisely the two arguments the audit later confirmed analytically. Its permanent Verdicts stand unchanged; only its Status tracking moved (see its §7.1 resolution log). If you read one companion document, read this one.
Project-Leviathan-RIMT/
├── README.md This file
├── whitepaper/
│ ├── RIMT-whitepaper.md v2.0 — Proposal and Analytical Falsification
│ ├── RIMT-whitepaper.html / .pdf Print exports
│ ├── export_html.py Markdown → print HTML (live docs only)
│ └── figures/ SVG figures + generator (build_figures.py)
├── critique/
│ ├── RIMT-adversarial-critique.md The red-team analysis (Verdicts permanent;
│ │ its two Fatal-risk findings were confirmed)
│ ├── RIMT-adversarial-critique.html / .pdf
│ └── critique_checks.py Independent re-derivations (stdlib only)
├── simulations/
│ ├── rimt_simulation.py First-order analytical models (the v1.2
│ │ model record + falsification helpers)
│ ├── test_rimt_simulation.py Unit test suite
│ └── requirements.txt
├── infographic/
│ ├── RIMT-infographic-data.md Canonical JSON data source
│ ├── RIMT-infographic.html Interactive "Illustrated Companion" — a
│ │ guided tour of the technology, the physics,
│ │ the engineering, and what the analysis
│ │ concluded (the falsification, honestly told)
│ ├── template.html Layout / interaction shell (data injected in)
│ └── build_infographic.py JSON + template → HTML renderer
├── biofouling/
│ ├── RIMT-antifouling-concept.md Active anti-biofouling hull skin (B-01) —
│ │ the surviving application; concept note,
│ │ no propulsion claim
│ └── RIMT-antifouling-concept.html / .pdf Print exports
└── archive/ Superseded / shelved — NOT live claims
├── README.md What is in here and why
├── comparison/ v1.2 RIMT-vs-conventional comparison,
│ frozen with a superseded banner
└── briefs/ Unpublished v1.2 audience briefs, shelved
Everything at the top level is live and reflects the falsification. Everything under archive/ is the frozen pre-falsification record — kept because the project's history is the point, but not to be cited as current.
Requires Python ≥ 3.9 and NumPy.
pip install -r simulations/requirements.txt
python simulations/rimt_simulation.py
cd simulations && pytest test_rimt_simulation.py -vThe simulation suite is kept deliberately: it faithfully encodes the v1.2 first-order model whose internal values (the efficiency headline, the solved drive voltage, the thrust density) the whitepaper now discusses as model-internal quantities falsified at the physics layer — the arithmetic was always correct; the coupling and momentum assumptions were not. The deterministic outputs reproduce bit-for-bit across the supported toolchain (Python 3.12 reference, NumPy ≥ 1.21, pytest ≥ 7; full run and test pass each take under a second).
| Quantity | Value | Significance |
|---|---|---|
| Seawater charge-relaxation frequency | ≈ 1.08 GHz (σ = 4.8 S/m) | At f_c = 2 MHz the liquid is a conductor (conduction/displacement ≈ 540) |
| Conduction-limited tangential field | E∥ ≈ 2 kV/m | The only field the electrolyte sustains at the available current |
| First-principles EDL coupling | α ≈ 1.8×10⁻⁴ | = the two-capacitor divider; no physical headroom above it |
| Consistent electro-osmotic slip | v_eo ≈ 40 µm/s | Smoluchowski at the screened field |
| Consistent Péclet number | Pe ≈ 0.12 | Diffusion-dominated — no deterministic ionic control |
| Deliverable thrust | ≈ 0.45 N/m² (at the model's own optimistic slip) | Wake momentum export, c_f·ρ·U·v_slip |
| Speed ceiling | U_max ≈ 2·v_slip (≈ cm/s) | Conveyor-belt theorem; hull-area independent |
| External corroboration | 2.2 cm/s (Hansen et al., 5 kg EO-propelled UUV) | The only EO vehicle in the literature sits exactly at this scale |
| Debye length in seawater | κ⁻¹ ≈ 0.39 nm | 0.6 M NaCl, 20 °C — the EDL is sub-nanometre |
| Carrier frequency | 2 MHz (2–5 MHz band) | Faradaic-suppression design window of the proposal |
| Electrode geometry | w = 10 µm, g = 5 µm, λ = 30 µm | Interdigitated lattice of the proposal |
| Dielectric stack | Ta₂O₅, 500 nm ALD | The Faradaic blocker; survives into the anti-biofouling thread |
The v1.2 whitepaper's own call for validation said it plainly: "a measured falsification, published openly, would be a more useful contribution than this disclosure itself." The falsification arrived analytically instead of experimentally, which makes it cheaper, reproducible by anyone with textbook physics, and immediately useful to anyone evaluating electro-osmotic propulsion at scale. Negative results that are specific, quantitative, and openly derived keep other teams from spending real money on the same dead end — and the conveyor-belt ceiling generalises beyond this project. All earlier versions remain permanently citable on Zenodo; nothing was retracted silently.
Szabó, G. (2026). Solid-State Marine Propulsion via Resonant Ionic Momentum Transfer (RIMT): Proposal and Analytical Falsification [Technical note]. Zenodo. https://doi.org/10.5281/zenodo.20361267
@misc{szabo2026rimt,
author = {Szabó, Gábor},
title = {Solid-State Marine Propulsion via Resonant Ionic Momentum Transfer ({RIMT}):
Proposal and Analytical Falsification},
year = {2026},
howpublished = {Zenodo},
doi = {10.5281/zenodo.20361267},
url = {https://doi.org/10.5281/zenodo.20361267},
note = {Concept DOI; resolves to the latest version. CC BY-SA 4.0.}
}The DOI above is the concept DOI — it always resolves to the latest version on Zenodo (the falsification, v2.0). Earlier versions, including the pre-falsification v1.2 disclosure, remain archived and individually citable there.
Gábor Szabó (sole author, concept originator)
AI research assistance acknowledged in the whitepaper Acknowledgments: Google Gemini, Claude (Anthropic). Per standard academic authorship norms, AI tools are not listed as authors.
This work is released under the Creative Commons Attribution-ShareAlike 4.0 International License (CC BY-SA 4.0). You are free to share and adapt this material for any purpose, provided you give appropriate credit and distribute any derivatives under the same license.
This repository constitutes a defensive publication. All material herein is placed in the public domain as prior art under CC BY-SA 4.0, effective from the date of the Zenodo DOI assignment — including the falsification analysis itself and the surviving application threads. The defensive-publication mechanism depends on a patent examiner's diligence in locating and citing the prior art; final patentability determinations rest with individual patent offices and courts in each relevant jurisdiction. This notice records the author's intent that the design space be kept open; it is not a unilateral declaration of unpatentability.