An R&D Proposal · From the Floor · June 2026

Salt, Water, Power

There is a battery that runs 120,000 cycles on salt and water, doesn’t burn, and costs 30–60× less than lithium by mass. Emerson already makes every tool to build it. I work the floor where the lithium boxes go out. Here is the case.

120,000 cycles · Chen et al., Nat. Commun. 2026 (CityU HK · SUSTech) · Mg2+/Ca2+ chloride brine, pH 7.0
Connor Scanlan · Material Attendant · Emerson Shakopee
The ask · $1.8M · 12 months · Shakopee NE corner, L1

How the cell works

hover or tap any part
≈ 2.2 V 120,000 cycles · pH 7 COP anode (−) MgCl₂ (aq) · pH 7 CuFe-PBA cathode (+) quinone ↔Mg²⁺ alkoxide Mg²⁺ → · ← Cl⁻ Fe³⁺ → Fe²⁺Mg²⁺ insertion Cu Cu e⁻ → electrons →
 

 

 
Salt percolates. Ions migrate. The cell breathes — and the lamp comes on. · 2.2 V · 120,000 cycles

How to Read This

The Proposal is the load-bearing document. The Evidence pages prove it. The Reference pages answer objections. The Signal pages are optional.

15 min · Skeptical
Proposal › three force-boxes
Then the FAQ’s Aquion-trap card. Risks register is Section F if you want it after.
45 min · Curious
Proposal end-to-end
The full case. Executive summary through the ask.
Already inside
Where We Are Now
Living state — spine, tributaries, what shipped last.
Long story
A History of Salt
Save for after. Five frames, five minutes. Why this craft belongs here.
The Pitch
The argument, in one document.
Start here
The Argument
The Proposal
The full case for Phase 1.
Executive summary, the science, the product stack (nine revenue surfaces including Plantweb Insight Aqueous and Battery-as-a-Service), the self-hosted pilot, the market, the supply chain, the financial frame, the IP playbook, the champions, and the ask in detail. If you read only one section: the Executive Summary force-boxes — box 5 (Plantweb-anchored revenue), box 7 (the diesel-free hyperscale data-center claim), and box 8 (financial outlook: $485M–$4.2B annual at year 10) carry the argument in four minutes.
v17 · ~45 min full read · 4 min for the three boxes Read
The Evidence
Proof the pitch is real.
Featured · WebXR
The Volume, in Motion
Tour the Lab
A Three.js scene of the ~10,000 SF lab as it will be built. Walk through it in your browser.
Orbit, zoom, click any of the 19 instruments for an in-line spec card with Emerson product line + purpose + status. Press T for a cinematic tour. Press F for first-person walk mode (WASD + mouse). Tap VR to step inside on Meta Quest or any WebXR-capable device. Equipment glows when running, pulses when in fault. Same code path that wires to Plantweb Insight telemetry in v3.
WebXR · click · T · F · VR Tour
The Place
The Floor Plan
The lab from above, with 45 field photos.
~10,000 SF Phase 1 footprint on L1 of the Shakopee Shell Space (field-measured 17 Jun 2026: 135'-4" × 112'-4" per floor, ~30,000 SF two-floor envelope). Four zones: glovebox NW / cycler NE / wet SE / CHO+break SW. Field documentation: 45 photos including the actual posted Storage Request Form, the reserved Irvine IT space, the perimeter windows.
SVG plan + 45 field photos · ~8 min See
The Method
The Lab Manual
The bench-grade procedure.
18 sections, 92 cells, four C-rates, twelve-month timeline. Hex-TADD-COP synthesis, electrolyte preparation (4.0M & saturated 5.8M tested), Byproduct & Recovery Strategy, Energy Budget. The canonical Phase 1 success-criteria table — tiered kill floor / target / stretch — lives here and is quoted by every other document.
v3 · ~30 min Read
The Volume
Peek Inside
A 30° isometric.
Nineteen pieces of equipment as extruded volumes: HPS-17000 cycler, glovebox, fume hood, electrowinning rig, solvent still. Cu recovery loop annotated.
SVG · ~2 min See
The Reference
For the skeptic and the safety officer.
Asked & Answered
FAQ
~80 questions across 7 sections.
Every question Connor asked himself, and every question a reviewer might. Each gets three layers: the one-sentence headline, the plain-English version, and the full engineering case. “I asked” badges mark questions that became proposal improvements. If a reviewer raises any objection, this page probably already answers it — in three depths.
~80 questions · 7 sections · ~25 min Read
Learn the science
The Reading Textbook
Twelve chapters. The chemistry, the cell, the cycle, the kill modes — university-style, tight.
A short textbook that teaches the aqueous magnesium-ion battery from first principles. What a battery is, why lithium burns, what changes when water is the electrolyte, why magnesium, what a Prussian-blue analogue does, how a single cycle works, why 120,000 cycles is possible, and what could kill it. Built to read in 25 minutes and to be pitchable from memory after.
12 chapters · ~25 min · left-rail TOC Read
Ops & Maintenance
The Operations Manual
How the lab runs. Built for the floor.
Six role cards with daily/weekly/monthly responsibilities. Three zone walkthroughs. The Ventilation Envelope (ACH, pressure cascade, recombiner, daylight + glazing, occupant amenity). 30-year service schedule. Twelve emergency procedure cards. 13 cited regulatory standards. Reference, not pitch. Read after the proposal lands — the manual proves the lab is real before it exists.
10 sections · 6 roles · ~35 min Read
Companion wiki
The Glossary
Every acronym, instrument, and concept — in plain English.
The proposal is dense with science-coded language: kVA, PXI, ICP-OES, NaOCl detox, NMP solvent, DeltaV, CuFe-PBA, Hex-COP, MgCl2, LCOS, BaaS, VCM, IBC, NFPA 2/55/70, OSHA PSM, ACH, Δp, K-13, AHJ. Every term gets a plain-English definition you can read in 20 seconds. Across the rest of the hub, look for dotted underlines — click any term to get the popover. Wave 30 Phase 2 ships 100 entries. Phase 3 expands toward ~150-200.
100 entries · equipment + concepts + regulatory · searchable Browse
The Signal · optional
If you only have a meeting room, the world is moving, or you want the long story.
If you only have a meeting room
The 12-Minute Pitch
The slides version.
23 slides. Same argument as the proposal, rendered for a room. Read the proposal instead if you have 45 minutes.
23 slides · ~12 min Open
The long game
The Carlsbad Play
Phase 4: vertical integration on the brownfield next to the largest desal plant in North America.
~492 metric tons of magnesium discharged unrecovered to the Pacific every day. 95 acres of brownfield directly adjacent. The honest carbon math, the five conditions for net-negative per kWh delivered, the ~$100–200M federal + CA incentive stack, and the named gaps. Gated on Phase 1–3 success. No Phase 1 capital committed. The long game named out loud.
9 sections · ~12 min · 2028 horizon Read
External signal
The Pulse
The thesis is moving in the world.
Three-column intelligence dashboard: community resistance to data-center build-outs, brine-recovery developments, aqueous-battery research. Auto-refreshed nightly. Every item links back to its publication.
3 columns · auto-refresh · RSS See
End here
A 5-minute essay
A History of Salt
Read after the proposal, if you want to know why salt, why now, why here.
A quieter argument. Why this craft belongs in a company that has spent its history building things that last. Five places where humans coaxed power out of salt and water; one place where the chemistry is just beginning.
5 frames · ~5 min Read
For the team
Where We Are Now
Living state. Bookmark this if you’re inside the program.
The spine, the tributaries, and the manifest of what shipped last. A coast-to-coast timeline from Research through Commercial Demo. Click any dot to expand.
7 phases · 8 tributaries · living Open
›  The working repository · changelog + design decisions Open if you want to see how this was built
Changelog · the working log
2026-04-14 06:12
Chen et al. read three times. Saturated MgCl₂, 120,000 cycles, pH 7.0. Why is nobody from Emerson at this? Proposal v1 drafted that night.
2026-04-26 22:30
Logan Woolery (cousin, Global PM · Industrial Wireless Instrumentation, the upstream sensor portfolio that feeds Plantweb) briefed by phone. Plantweb route to Stokes locked. Train (CSO) noted as nuclear option.
2026-05-07 05:42
The 12-Minute Pitch v1 (23 slides) + protocol v2 audit B+→A+ after 21 fixes. Floor plan SVG locked the NE corner. Rejected adding a Voice-of-God section to the protocol; kept the flat technical register.
2026-05-09 05:57
Walkie-talkie wired: ntfy.sh/connor-proposal → cron-fired Claude session. Topic name is case-sensitive (learned the hard way).
2026-05-11 02:42
Proposal v13: 5 grafts. Considered cutting the "Why I'm Writing This" first-person box. Kept it. Audit confirmed: only first-person voice in 2,150 lines.
2026-05-11 23:14
Brand declared: Salt, Water, Power. Rejected: Aqueous (too startup), The Permanent Electrolyte (overclaims), Saltwater (under-claims; chemistry is MgCl₂/CaCl₂). Crescendo rebuilt; 11 photos recovered from v10 before v11 pruning. Page XI light-themed: the exception.
2026-05-12 14:00
v17 reframed: Plantweb Insight Aqueous as the SaaS spine, $30M–$100M ARR by year 10. Walkthrough v2: walls + ceiling + first-person walk mode + per-equipment click panels + digital-twin emissive seed. Hub privileges the walkthrough.
Decisions · what we picked / what we rejected
QuestionPickedRejected (and why)
Brand nameSalt, Water, PowerAqueous (startup-y), The Permanent Electrolyte (overclaims hedged language), Saltwater (chemistry is MgCl₂, not NaCl)
Phase 1 envelope$1.0M–$1.8M$1.1M–$2.0M (introduced by zero-discharge additions; rolled back to match slides; honor the original ask)
Lead documentProposal first · the three force-boxes carry the argument in four minutesSlides-first (too rehearsed for a reading room), Crescendo first (no context). The deck remains, demoted to The Signal · “if you only have a meeting room.”
Outro of CrescendoCream colophon, signature only"Phase 1 is where we add the next page" outro (audit: "salesman tap-tapping the mic after the artist has left the stage")
Hub framingPrivate reading room for invited readers"Central destination for Emerson employees" (overreach; undermines Material Attendant credibility)
Visual + economic framingDark Salt Kitchen palette · Cu as Stewardship ServiceLight whitepaper register (would have looked like every other internal pitch); 70/30-split CRM module (per-installation cash 2–3 orders too small, so Cu becomes a sleeve of the SaaS, not its own product)
Connor Scanlan
Material Attendant · Shakopee · 2026

Built on nights and weekends. If the math is wrong, please show me. If the math is right, please move.

Not reviewed or sanctioned by Emerson Electric. A private working draft. Please do not circulate without permission.