The mechanism, the permit, and who else reads the instruments.

What the physics gives you, and what it costs to size

Below roughly 1,000 nanometers a bubble stops behaving like a bubble. Buoyancy no longer dominates, the charge on the surface keeps neighbors apart, and the combined interface available for gas to cross into water becomes enormous for the volume delivered. That is the whole reason the treatment stage fits on a hull rather than in a works.

The elevenfold mass-transfer result in the table is one rig, one water, one measurement. Across the wider aeration literature the reported enhancement factors run over a broad range, and eleven sits high in it. For a technical evaluator the consequence is a sizing question rather than a mechanism question. Size a fleet against the top of that range and you have a cheaper program than one sized against the middle — and a thinner margin when a bad August arrives. Alarivean sizes against the middle.

The two NOAA releases, and the four letters

Neither NCCOS release concerns a SeaBreather. The 2018 work names a system called NABAS. The 2020 system abbreviates to NBOT, standing for Nanobubble Ozone Technology, which is the same four letters this network uses for Nano Bubble Oxidation Technology — a different owner's hardware on a different duty. What the pair establish is that this approach clears what it is aimed at, at pond scale and in a ballast tank, without poisoning the receiving water.

Work done under a completed NOAA agreement

Alarivean completed the term of a Cooperative Research and Development Agreement with NOAA, entered through AMURI. That term has ended. No CRADA is running today, and the past tense is the only correct one for it. CRADAs are private instruments and are never posted, so no public copy exists to retrieve — true while it ran, and true now. Work under it with the agency scientist Dr Peter Moeller established something specific: the treatment sterilizes a toxin by disabling the functional groups responsible for the toxicity, rather than dropping the material out of the water as a precipitate. Anyone who has had to answer the question of where a contaminant ended up will recognize why that distinction is the interesting one.

The working relationship outlasted the paperwork. Dr Moeller continues as an advisory and verification voice on the research, which is a looser arrangement than an agreement and an honest description of what it is.

The original CRADA was built around one particular nanobubble generator, and that piece of equipment has since been left behind. Research after it has run with Arizona State University — Dr Sergi Garcia-Segura, Dr Jesús Morón-López and colleagues — with the Middle East Desalination Research Center, and with more. Those facilities suited the scale the company wanted to test next. That was the reason for the move, and there is no more interesting story behind it.

Dissolved metals follow the same rule as everything else this platform treats: neutralized, not removed, and no PFAS destruction is claimed anywhere in the program. A hydroxyl radical does not destroy a metal, and the organic chemistry bound to one is where the work lands — a 2023 review of advanced oxidation and heavy metal complexes finds the ligand comes apart and the metal is released rather than removed. Read it with the same caution as everything else in this register: it is wastewater work, the bulk of it on one model compound, and nobody has taken it to open water.

A research agreement is not an endorsement, and a finished one is not either. NOAA does not endorse commercial products, has not been asked to, and nothing in this network should be read as a federal agency putting its name behind a hull.

The Florida permit

In September 2024 Alarivean, Inc. of Scottsdale, Arizona received Florida Department of Environmental Protection permit FLOA00062, issued by FDEP's South District Office under Chapter 403 of the Florida Statutes with a five-year window, to field-test an ozonated-seawater craft producing 1,000–2,000 gallons per minute for red tide mitigation. The trial is conditioned on active Karenia brevis above 100,000 cells per liter.

That craft is a modified personal watercraft, a fraction of the size of a SeaBreather hull. Its permitted 1,000–2,000 gallons per minute is a fact about that trial vessel and describes nothing about the class.

Two outside bodies documented it: the Sarasota Bay Estuary Program, which gave the application technical support and deliberately no money, and the Florida nonprofit START. The estuary program's public position is that efficacy in open water is unproven, and it is — nothing has been published, by anyone in this category.

The technical support without the funding is the working part of that arrangement. A program that has taken nothing from Alarivean can publish whatever the water says. What it has asked for is the measuring role once open-water work begins, so that anything claimed past the mouth of that bay rests on readings its own scientists took. Alarivean has asked for the same thing.

Bromate, and whose limit applies

Ozone finds the bromide in seawater far more readily than the chloride, and the hypobromite that follows can go on to bromate. That reaction rate is why an oxidation duty is run enclosed and instrumented rather than out in the open. Bromate is a stable end-product and cannot be quenched back out once it exists, which makes governing its formation the whole of the job.

You will not find a fixed number for it in this register, and that is a decision rather than an oversight. The value that binds a discharge is the value that applies to the receiving water, agreed with the consenting authority and written into the service level agreement before a system deploys. A drinking-water standard is a poor test for open sea, and the desalination intake the question usually turns on carries bromate handling of its own already.

Formation is assessed per water body, because it is not a given. Quantity needs a bromide threshold that salt and brackish water do not always meet, and ozone is spent indiscriminately across the dissolved carbon in the column, so bromide competes for it. Real-time instruments cannot see a stable end-product, so independent laboratory work confirms it and the reporting line runs to the regulator that issued the permit. The client reads the same feeds and can put an active deployment on hold on any suspected infraction until it is cleared to their satisfaction.

The two open questions

Radical chemistry. Moleaer and Arizona State University reported reactive oxygen species from injected nanobubbles in 2020; a controlled 2023 study by Chae and colleagues in ACS ES&T Engineering found generation minimal, if it occurred at all, at ambient conditions. No part of a SeaBreather program is costed on radical chemistry.

Open-water efficacy. No open-water result has been published — not this platform's, and not a competitor's. What exists is pond scale and ballast-tank scale. The calibration phase is where a client gets a number on their own water: a bounded area, viability parameters agreed in writing beforehand, and an independent institution taking the readings. A figure for a named bay follows the site read, with its assumptions printed alongside it.

Which is also why the commercial terms are shaped the way they are. A subscription is written against a performance specification, and performance outside it ends the subscription. After the minimum subscription period a client who remains unconvinced can end it too, with no reason required and nothing to prove. Anyone taking a position this early on evidence should not also be taking a position on being stuck.